- Email: [email protected]
Hospital merger control in Germany, the Netherlands and England: Experiences and challenges
Health Policy 120 (2016) 16–25
Contents lists available at ScienceDirect
Health Policy journal homepage: www.elsevier.com/locate/healthpol
Review
Hospital merger control in Germany, the Netherlands and England: Experiences and challenges Andreas Schmid a , Marco Varkevisser b,∗ a b
Health Management (MiG), University of Bayreuth, Germany Institute of Health Policy & Management (iBMG), Erasmus University Rotterdam, The Netherlands
a r t i c l e
i n f o
Article history: Received 28 July 2015 Received in revised form 30 October 2015 Accepted 2 November 2015 Keywords: Hospital mergers Competition policy International comparison
a b s t r a c t Aiming at the efficiency enhancing and quality improving effects of competition, various steps have been undertaken to foster competition in hospital markets. For these mechanisms to work, robust competition policy needs to be enacted and enforced. We compare the hospital markets in Germany, the Netherlands and England regarding their experience with competition and put a special focus on merger control and the stringency of its implementation. Elaborating on the differences in merger control practice we find that despite very similar goals the respective agencies apply very different approaches and take fundamentally different routes when balancing proclaimed benefits of mergers with potential risks of consolidated markets. While the German competition authority has a strong focus on maintaining the preconditions for competition, in the Netherlands we find over the past decade a much stronger focus on hypothesized countervailing buyer power, accepting in turn highly concentrated markets. In England we find the currently most comprehensive analysis of proposed mergers in combination with a clearly positive assessment of the effects of patient choice and competition on prices and quality. All agencies are still reluctant to implement merger simulation models or similarly advanced econometric methods in their appraisal. One very likely reason is a lack of country specific empirical evidence on these matters. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction A growing number of countries have introduced or strengthened competition among health care providers, including hospitals, to improve the functioning of their health care systems. For hospital competition to be successful, the existence of a sufficient number of alternative hospitals for patients and payers to choose from is required. The empirical literature clearly shows that mergers between rivals in concentrated markets are likely to
∗ Corresponding author. E-mail address: [email protected] (M. Varkevisser). http://dx.doi.org/10.1016/j.healthpol.2015.11.002 0168-8510/© 2015 Elsevier Ireland Ltd. All rights reserved.
increase prices. The results for the effect on quality seem to depend on the price setting mechanism. In markets where prices are set by hospitals the empirical evidence is mixed, but competition between hospitals operating in markets with regulated prices has generally been found to have a positive effect on quality [1]. It is therefore argued that when the market mechanism is used for maximizing the public healthcare interests competition enforcement should be strict [2]. In the final opinion of the Expert Panel on effective ways of investing in Health (EXPH), it is recently concluded that the introduction of competition between providers of health care requires, among other conditions, the enforcement of competition rules to prevent the creation, strengthening and abuse of dominant
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25
positions [3]. However, hospital mergers may also benefit patients. For example, there is some evidence that reaching certain volume thresholds improves quality for specific complex hospital services [4]. Overall, the relevant question from the patient’s perspective is whether the benefits of hospital consolidation outweigh the negative effects caused by reduced market competition. Using a semi-structured narrative approach, this paper compares and discusses the experiences with and challenges for hospital merger control in three countries with competition among hospitals: Germany, the Netherlands, and England. In our country comparison, the focus is first on institutional differences related to hospital competition. This is the basis for the subsequent description of the respective approaches used for ex ante hospital merger control, as well as the comparative discussion of results achieved and challenges faced by the competition authorities. Table 1 provides some relevant key indicators for the three countries (i.e. including the United Kingdom rather than England because of the OECD statistics used here), highlighting already some of the structural differences and similarities. Each of the countries included in the paper has experienced consolidation of the hospital industry. In Germany, in addition to a more “pro-market” attitude in health politics, hospital merger activity is predominantly fuelled by payment reforms and shrinking financial resources of municipalities owning public hospitals. The impact of those mergers on the hospital market structure is twofold [5]. First, the formation of multi-site hospital systems on a local level. Second, on a supra-regional level hospital chains arise that are active in various local hospital markets across the country. Consolidation in England has also been driven by financial pressure. However, whether a hospital is merged or not depends not just on (financial) performance but also on national politics related to the National Health Service (NHS). That is, to avoid unpopular closures merger activity between geographically co-located hospitals may also be initiated by the government [6]. In the Netherlands, the most important merger motives for hospitals – as well as other providers of health care – are quality improvements associated with the use of minimum volume thresholds for complex procedures and increased bargaining clout vis-à-vis payers followed by financial and efficiency considerations [7]. 2. Hospital competition 2.1. Germany Over the past 15 years, encouraged by official advisory bodies such as the German Advisory Council on the Assessment of Developments in the Health Care System [8–10] and the Monopoly Commission [11], policy makers have fostered the role of competition in the German hospital sector in multiple ways. For example by increasing transparency through mandatory quality reporting and allowing for some degree of selective contracting between sickness funds and providers. In this context, competition is always intended to achieve higher levels of efficiency and to increase quality of care [12]. The greatest impact on the
17
cost side came from the introduction of the reimbursement system based on Diagnosis Related Groups (DRG) in 2003/4, which forced providers into cost oriented yardstick competition as the case-based lump sum payment is related to the average costs across hospitals. Acute care hospitals are now reimbursed on a case basis and the regulated price is calculated as an average of costs across a sample of hospitals. Cost structures have become significantly more transparent and large numbers of hospitals faced severe financial challenges. A relatively small number of hospitals went out of business and there were a considerable number of mergers, many of which were aimed at keeping all existing hospital sites open. This is partly due to political pressure especially at the local and state level, where (public) hospitals are considered to be important factors for the local economy (e.g. because they are important employers) [13,14]. There is considerable heterogeneity in the degree of hospital market concentration across the 16 German states [15]. Reasons for this heterogeneity include differences in geographic settlement structures, the states’ different influence in hospital planning and the structural legacy differences between former East and West Germany. The average catchment area ranges from 108 km2 in North Rhine-Westphalia over 224 km2 in Bavaria to 703 km2 in Mecklenburg–Vorpommern – the latter two states both having large rural areas. While an average hospital in Bavaria serves just about 40,000 people, other states have larger institutions that serve – as in Saxony – up to 55,000 people on average [16]. Generally speaking, there is excess capacity in metropolitan areas while it is challenging to ensure adequate access to care for patients in rural areas [16]. By 2007 about 40% of German hospitals operated in concentrated or highly concentrated regional hospital markets, with a Herfindahl–Hirschman-Index (HHI) of 0.18 or higher. Concentration is generally higher in rural states, but some more urban areas also have highly concentrated markets [14]. There is even more variation when concentration is analyzed for different procedures; i.e. in disaggregated product markets. While the average HHI for selected elective surgical procedures (e.g. hip or knee replacement) is as low as 0.19, markets for pneumonia or maternity units have an average HHI of more than 0.3 [5]. While the cost oriented incentives of the yardstick based DRG system are extremely strong, the quality component is still trailing behind. Considerable efforts have been undertaken to foster quality based competition, e.g. by the aforementioned mandatory quality reporting of hospitals and an increasing number of websites that provide quality related information. However, according to a recent study by Emmert et al., despite an overall positive trend, only about a third of their respondents were aware that such information existed in the first place [17]. Despite the low awareness of quality indicators, the role of patients is crucial as at the individual provider level increased patient volume leads to higher revenues in the long run. Physicians, including both general practitioners and independent medical specialists, are advised to include the two nearest suitable hospitals in their referral (§§ 39(2), 73(4) SGB V). If patients do not choose a hospital recommended in the referral they could be charged any
18
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25
Table 1 Relevant key indicators for Germany, the Netherlands and the United Kingdom. Germany
Netherlands
United Kingdom
Total population Land area (km2 ) Population density (population per km2 )
81,900,000 357,000 229
16,800,000 33,700 497
63,700,000 242,500 263
Type of health system Heath expenditure (% GDP) Health expenditure per capita ($) Acute care hospitals beds (per 1000 population)
Bismarck 11.0 4,819 5.3
Bismarck 11.1 5,131 3.3
Beveridge 8.5 3,235 2.3
Source: OECD.Stat (accessed on 15 October 2015).
additional costs – an option that is not enforced in practice. The competition authority and the respective courts agree that despite these formal restrictions patients can de facto exercise free hospital choice. The Bundeskartellamt (BKartA – Federal Cartel Office) therefore focuses on competition for patients. The fact that patients and not sickness funds are the relevant customers of hospitals, however, negates a potential conflict of § 69 SGB V (Book V of the Social Code) granting some exceptions from competition law for certain relations between sickness funds and providers [18,19]. In summary, German hospitals compete on the basis of costs with fixed prices and – still on a low level but increasingly – on quality. Merger control focuses on the patients as the customers that need to be protected while third party payers are not relevant in these considerations. 2.2. The Netherlands After decades of top-down health care rationing policies, the Netherlands has opted for a system of regulated competition and private insurance with wide-ranging reforms implemented since the mid-2000s to strengthen the role of market mechanisms [26]. In 2006, competition among health insurers was reinforced with the introduction of the Health Insurance Act making private health insurance mandatory for every Dutch citizen. The basic idea behind the introduction of regulated competition was to give risk-bearing private health insurers appropriate incentives to act as prudent buyers of health services on behalf of their customers. To that end, the insurers are expected to negotiate with health care providers about the price, quality, and/or volume of care allowed. To strengthen insurers’ bargaining strength vis-à-vis providers, they are allowed to contract selectively and use financial incentives for channelling patients to preferred providers. In addition to the fundamental reform of the health insurance system, a gradual deregulation of provider markets was also carried out. This deregulation included the market for hospital services. In anticipation of the major 2006 reform of the Dutch health care system, in February 2005 insurers were allowed to negotiate contracts (including prices) with individual hospitals for a number of routine hospital services, such as cataract surgery and hip replacement. These hospital-insurer negotiations were facilitated by the introduction of a newly developed hospital product classification system based on the diagnoses and subsequent treatment of patients. Since then each patient admitted to a Dutch hospital or visiting a hospital’s
outpatient clinic is categorized into a Diagnosis Treatment Combination (DTC). Each DTC includes all hospital activities and services (both inpatient and outpatient) associated with the patient’s care, from his initial consultation or examination to the final check-up. The proportion of freely negotiable hospital services (labelled the ‘B-segment’) was stepwise expanded from about 10% in 2005 to about 34% of total hospital revenue in 2009. To increase the incentives for efficiency, the room for price negotiations between hospitals and health insurers was further expanded in 2012 to about 70% of hospital turnover. For the remainder of hospital production (labelled the ‘A-segment’), containing the more complex treatments, prices per DTC are still determined by the Dutch Healthcare Authority (NZa) which is the independent administrative body supervising all healthcare markets in the Netherlands. In addition to the insurer-hospital negotiations, hospitals also compete directly for patients. Since the introduction of the new Dutch health care system patients are encouraged to make an active choice between alternative providers, for example by the provision of consumer information about hospital quality. Though the publicly available hospital quality ratings are still far from perfect, empirical research indicates that at least to some extent patients are sensitive to differences in observed hospital quality [27,28]. However, this research also clearly reveals that travel time is still the most important choice determinant. Hence, hospital competition in the Netherlands seems to be restricted to a few neighbouring hospitals. This is also reflected in the fact that the average Dutch hospital has a market share of 50% in its catchment area [29]. The NZa placed any zip code in a hospital’s catchment area if earnings from that zip code were at least D 1. 2.3. England Government initiatives over the last couple of decades have introduced competition and choice in the provision of NHS services. There are two models of competition in the NHS in England: quality competition in the market, which occurs in routine elective (planned) services where patients have a choice between hospitals; and competition for the market, where healthcare providers compete on quality, and sometimes price, to be selected by commissioners as the sole provider of services to patients. In the NHS in England, healthcare services are agreed, planned and monitored by a number of different commissioning bodies. Most of the NHS commissioning budget is now
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25
managed by 211 clinical commissioning groups (CCGs). These are groups of general practices that are responsible for planning healthcare services for their patients and population. Commissioners often have a wider set of choices than patients and so recent merger control has tended to focus on the impact on competition in the market. Patients have a choice of provider in respect of their first consultant-led outpatient appointment for elective care and for maternity services. The costs of all NHS patient care is fully funded by the government, with no out-ofpocket payments by patients, and patients can only access acute healthcare by referral from their general practitioner (GP). Consequently the GP plays an important role advising the patient on which hospital to choose. The NHS provides information to help patients choose, with the advice of their GP. These include the ‘Choose and Book’ website and the NHS Choices website, which shows the Care Quality Commission’s report on quality of care as well as “Trip Advisor” style feedback from patients. Empirical evidence has shown that in addition to travel time, quality of care and waiting times influences hospital choice [20,21]. Therefore, from the demand side perspective the conditions are in place for quality competition to provide incentives for hospitals to compete for patient/GP referrals. These demand side incentives are complemented by incentives on the supply side. Under the Payment by Results (PbR) regime, which covers the majority of acute healthcare, hospitals are paid a fixed national tariff per treatment so that their income depends on the number of patients that they treat. Hospitals in England are managed by acute trusts, some of which have achieved Foundation Trust status. Foundation Trusts are required to provide certain services, but are afforded a degree of operational autonomy and can keep their surpluses and borrow to invest in new and improved service. This gives NHS Foundation Trusts incentives to maximize their income by taking steps to attract patients for profitable specialties, for example by maintaining and improving service quality. NHS Foundation Trusts are regulated by Monitor, which is required to protect and promote the interests of people who use the NHS by promoting the provision of healthcare services. As providers of publicly funded NHS services for patients, Foundation Trusts have many different objectives and healthcare professionals and managers want to deliver high-quality care for their patients. However, these organizations also have the objective of ensuring they receive sufficient income to cover costs. The quality of healthcare provided by a hospital is the outcome of many different decisions that are made at different levels across an organization trading off different factors. The effect of competition on quality where prices are fixed centrally is to focus these spending decisions such that account is taken of the factors that matter to patients and GPs. Empirical studies of the effect of quality competition in the NHS in England provide some evidence that competition is associated with improved quality and efficiency for patients diagnosed with acute myocardial infarction (AMI) [22–24]. One paper looked at the impact of competition on management quality in NHS hospitals and found that management quality is strongly correlated with
19
financial and clinical outcomes (such as survival rates from emergency heart attack admissions) [25]. It is found that higher competition is positively correlated with management quality and that adding a rival hospital increases management quality and increases heart attack survival rates. 3. Hospital merger control 3.1. Germany The regulatory framework for merger control in Germany, including hospital mergers, is given by the Gesetz gegen Wettbewerbsbeschränkungen (GWB – Act against Restraints of Competition) which is enforced by the Bundeskartellamt (BKartA). This law was amended by the 8th Amendment of the GWB in 2013, aligning the substantive test with EU standards, i.e. the SIEC-Test (Significant Impediment to Effective Competition). For most cases, the change is not too fundamental, as the prior core criteria, i.e. the creation or strengthening of a dominant market position, remains the standard example within the SIEC-Test for horizontal mergers [30]. The SIEC-Test allows for the consideration of relevant customer benefits, but it remains to be seen if the BKartA will give more weight to this aspect in its future practice. Currently it seems unlikely that the authority would clear an otherwise harmful merger on the basis of merger-specific benefits to patients. While the evaluation of a potential dominant position is a comprehensive analysis, taking account of numerous factors that determine the actual competitive situation in the market such as barriers for market entry or exit, financial situation of competitors and potential future competition, the merging hospitals’ combined market share is an important first indicator (§ 18(3) GWB). For single companies the threshold for this market power presumption was raised from 33.3% to 40% by the 8th amendment of the GWB (§ 18(4) GWB). For a group up to three firms the threshold is 50% and for up to five firms 66.6% (collective or oligopolistic market dominance). It is important to note that not only full mergers but also acquisitions are subject to the full assessment even if full control is not acquired. For example, also joint ventures and minority shareholding can fulfil the criteria outlined in detail in § 37 GWB. The turnover thresholds for merger control, deciding whether the merger falls under the jurisdiction of the BKartA, are of particular relevance for the hospital sector. The threshold is supposed to ensure that only mergers of macro-economic relevance are monitored by the BKartA. Among other thresholds, § 35 GWB states that the aggregated turnover of the involved companies must exceed D 500 million. In the hospital sector that is characterized by regional markets and comprises a large number of local and regional hospital systems this has been a concern, as many mergers that are likely to create market power in a regional market remain under the radar. However, up to now suggestions for improvement, such as using a turnover multiplier to assess hospital mergers [11], have not been implemented. This may result in a rather uncomfortable situation. That is, when two independent local hospitals merge without exceeding the turnover threshold
20
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25
Box 1 The Elzinga/Hogarty approach is easy to apply and only requires commonly available patient flow data. It begins with a narrowly defined market and then expands the boundary until threshold conditions are met for both imports and exports of hospital services. In various US court cases, the analysis focused on identifying geographic areas such that both statistics are either below 25% (“weak market”) or 10% (“strong market”). The central methodological problem underlying the approach is that the presence of a minority of travelling patients does not necessarily discipline hospitals from exercising market power over the silent majority of non-travelling patients.
their consolidation will not be subject to merger control scrutiny, even though it may lead to a dominant position in the relevant market. On the other hand, when one of the merging hospitals belongs to a hospital chain that also owns hospitals in other parts of the country the threshold is certainly exceeded for this economic entity and therefore the merger will always be assessed by the competition agency while it does not have an impact on local/regional hospital competition. Furthermore, there is some controversy around the fact that for municipalities that own a hospital the turnover of affiliated municipal companies such as Sparkasse (mutual savings bank) or waste management are also taken into account [11]. To calculate the market shares – which are calculated on the number of patients treated – the relevant market needs to be defined. For the geographic delineation of hospital markets the BKartA looks at patient flows on zip code level and tries to identify largely self-contained areas. The starting points are the hospitals of the merging entities. In a first step, the catchment area is determined from a supply side perspective and areas with homogeneous market conditions that differ from neighbouring areas are identified. In a second step, the share of patients from these areas that are served by the corresponding hospitals are calculated from a demand side perspective. The focus is on the inflow of patients into the aforementioned areas. Other than the widely criticized Elzinga/Hogarty test (Box 1), this approach usually results in rather narrowly defined geographic markets [31]. There are no official thresholds for the definition of self-contained areas and for reasons of confidentiality only aggregated data are published. The law also gives the BKartA the flexibility to consider further aspects in the evaluation of the competitive situation on an ad hoc basis [32]. The BKartA uses a cluster market approach to define relevant product markets, defining all services under the label “acute inpatient hospital care”. A slightly more detailed analysis will be conducted on an ad hoc basis, for example if one of the competitors is a dedicated orthopaedic or cardiac hospital. Rehabilitation and ambulatory care are considered to be separate markets. This cluster market approach results in rather broad product markets, despite the availability of very detailed transaction data. As a result, the BKartA opts for a higher risk of under-enforcement rather than over-enforcement. This has been discussed critically
by courts [19], but to date the BKartA has been very successful in defending its decisions. Hence, in current practice market definition has been determinative. The competition authority has the flexibility to also consider other factors affecting the post-merger level of competition, but so far these factors are of secondary importance at most. 3.2. The Netherlands In the Netherlands, the general Competition Act (Mw) provides the regulatory framework for hospital merger control. The Mw was introduced in 1998 and is modelled after the competition rules in the EC Treaty and subsequent legislation. It includes a prohibition on cartels; a prohibition on the abuse of a dominant position; and a preventive merger control regime. It is enforced by the Netherlands Authority for Consumers and Markets (ACM), an independent administrative body; i.e. ACM is part of the Dutch central government but does not belong to any ministry. Under the Mw, mergers between any firms (including hospitals) whose combined and individual turnovers exceed the thresholds that are in force are subject to notification and prior approval by ACM. This requirement applies to all mergers between companies whose combined turnover has exceeded D 113 million in the preceding calendar year, with at least D 30 million realized in the Netherlands by at least two of the companies involved. Thresholds for the health care industry were temporarily lowered in January 2008 to D 55 million and D 10 million respectively. This was because geographic markets for health care tend to be small and the emerging competition in the Dutch health care system is fragile since competition among providers has only gradually been introduced since 2005. A third threshold was added to prevent these thresholds from applying to mergers involving companies whose health care services are only a small part of their business. This requires that for at least two of the companies involved turnover from health care services alone must exceed D 5.5 million. The lowered health care specific thresholds are currently still effective. The ACM starts a general review when it is notified about a proposed hospital merger. If the merger is not likely to be anticompetitive, the ACM clears the merger. Otherwise it decides that a license is required. A more substantial assessment of the proposed merger follows when the merging hospitals do need a license and also apply for it. According to Section 41.2 of the Mw, ACM will eventually prohibit a merger “if, as a result of the proposed concentration, effective competition on the Dutch market or a part thereof would be appreciably impeded, specifically as a result of the creation or strengthening of a dominant economic position.” On the basis of supranational European guidelines, a dominant position is defined as “a position of one or more undertakings that enables them to prevent effective competition being maintained on the Dutch market or a part thereof, by giving them the power to behave to an appreciable extent independently of their competitors, their suppliers, their customers or end-users” (Mw, Section 1.i). When assessing the likely competitive effects of a proposed hospital merger, ACM takes into account the non-binding opinion of the Dutch Healthcare Authority
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25
(NZa). In its opinion, the NZa uses merger simulation to predict the price effect of the consolidation. Note that Dutch hospitals also require formal permission from the NZa for their merger plans. This requirement, however, focuses on the merger process and the accessibility of some essential hospital services (24/7 emergency care) and does not involve an assessment of the likely competitive effects. Before 2004 proposed hospital mergers were not assessed in the Netherlands since it was argued that hospitals were not able to compete due to strict supply and price regulation. This changed when in January 2004 the competition authority published a position document stating that as a result of institutional and regulatory changes in the Dutch health care system the economic and legal context by then offered scope for competition among hospitals. Until July 2015, 27 hospital mergers have been subject to an ex ante assessment of which the first 26 cases were all cleared after an initial or more substantial investigation. Only very recently ACM for the first time prohibited a proposed hospital merger involving two hospital groups in the south-western part of the Netherlands, near the cities of Dordrecht and Gorinchem, because they are strong competitors of each other [35]. Obviously, it is too soon to say whether this decision marks the beginning of a more stringent approach to hospital mergers in the Netherlands. As a result of the permissive policy that can be observed over the past decade, the number of general hospitals has steadily declined over the years from 90 in 2004 to little more than 80 in 2014 and, based on the merger plans already made public, local and regional hospital markets are likely to become even more concentrated. It may therefore not come as a surprise that hospital merger control in the Netherlands has been criticized for being too permissive [31,36]. In some cases, mergers are even permitted on very questionable grounds and without a proper definition of the relevant product and geographic market. Without further investigation in hospital merger decisions typically two separate product markets are identified: one for inpatient hospital services and one for outpatient hospital services. For defining the relevant geographic market ACM starts by looking at patient flows. After this the size of the geographical market is most often left open-ended because it is argued that current patient flows are not indicative for patients’ willingness to travel in the future when quality differences across hospitals are expected to become more transparent. The approval of the merger between two hospitals in the city of Tilburg clearly illustrates the fundamental flaws of hospital merger control in the Netherlands [36]. The Tilburg merger was cleared by ACM in November 2012. While based on patient flow data combined market shares ranged between 60–70% and 70–80%, and substantial price increases were predicted by the NZa, the merger was cleared using speculative claims about patients’ willingness to travel and health insurers’ future countervailing buyer power, and with the acceptance of a non-binding temporary price ceiling. In this case, ACM based its decision on the opinions put forward by the health insurers without carefully examining these. The fact that, as can be concluded from current patient flows, other hospitals did not exert competitive pressure on the merging hospitals was
21
ignored. In its new guidelines for assessing mergers and collaborations in health care, issued in 2013, ACM officially confirms this questionable approach: “When assessing a concentration’s implications, the arguments put forward by insurers and patient organizations will be central. The quantitative analyses based on patient flows with regard to the extent to which the merger hospitals exert competitive pressure on each other primarily act as a first scan.” [37]. 3.3. England After taking over many of the functions of the Competition Commission and certain consumer functions of the Office of Fair Trading, as amended by the Enterprise and Regulatory Reform Act 2013, from 1 April 2014 the Competition and Markets Authority (CMA) is responsible for merger control in England, including mergers of NHS Foundation Trusts. Monitor has an important role advising the CMA on possible relevant patient benefits which could outweigh any substantial reduction in hospital competition and choice. The CMA must decide whether a relevant merger situation gives rise to a substantial lessening of competition (SLC) [33]. Patient interests are at the heart of any merger review. The CMA must decide whether a merger is overall in the interests of patients and takes into account potential improvements to services. Choice for patients and commissioners, and competition between hospitals to attract them, helps to ensure that providers have incentives to improve the quality and efficiency of their services. The introduction of greater choice for patients and commissioners and the greater independence of providers means that merger control has become more relevant and the CMA must make sure that changes improve outcomes for people who use the services. Sometimes a merger between competing hospitals may decrease the hospital’s incentives to improve their services, leading to worse outcomes for patients. On the other hand, sometimes a merger may be the best way of delivering certain benefits to patients in a timely manner. The merger review process is designed to examine both of these aspects and determine on balance what is in the overall interests of patients. Until July 2015, the CMA has reviewed six NHS hospital mergers. Four of these were cleared but one, the proposed merger of Royal Bournemouth and Christchurch NHS FT and Poole NHS FT, was prohibited following an indepth investigation. Currently, one merger is the subject of an in-depth investigation. The CMA and Monitor work closely together to ensure the process is useable and efficient and focuses on getting the right outcome. Monitor provides early upfront advice to FTs considering mergers. It is hoped that this early engagement will result in fewer “problematic” mergers which are not in the overall interests of patients. The CMA and Monitor are also working together to ensure that the process is well understood and useful; the CMA has recently published guidance on the review of healthcare mergers, and the CMA and Monitor have jointly published a short guide for hospitals [34]. The CMA has developed a methodology for the review of hospital mergers. In all five of the decisions since 2012, for the purposes of its competitive assessment it has treated
22
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25
each medical specialty as a separate market, with separate inpatient and outpatient services within each specialty and separate elective and non-elective services. Demand side factors give the starting point of narrowly delineated product markets, covering each different service. Supply side factors, particularly the existence of resources such as clinicians, equipment and facilities that are common across services, allow for the aggregation of individual services into broader specialty level markets. The CMA has not widened the relevant product market beyond specialties because of limits to supply side substitution from one specialty into another, including the need to invest in new staff, facilities and equipment as well as the difficulties of small scale expansion that arise because a non-negligible proportion of costs are indivisible and hospitals need to achieve economies of scale. The CMA has adopted a pragmatic approach to geographic market definition, with a focus on the competitive assessment. It has used isochrones based on catchment areas as the starting point for the competitive assessment. Catchment areas show the area from which a large proportion of patients originate. For example, in the RBCH/PH merger, the CMA found 80% catchment areas of about 20 min and used isochrones (lines on a map showing the points at a given travel time) around each hospital to identify the geographic areas from which the hospitals attracted the majority of their patients and therefore the area over which they competed. There can be significant variation across specialties and the CMA has typically adopted sensitivity analysis to capture these differences. As a general principle, a catchment area is typically narrower than the area over which a hypothetical monopolist could profitably sustain a small but significant price increase (or quality deterioration). As a result, the starting point for the competitive assessment is the areas defined by catchment areas analysis, with account taken of constraints from hospitals located further away. The assessment of the impact of hospital mergers presents particular challenges, particularly to using standard merger control techniques, because (i) regulation and institutional design both play a significant role in shaping providers’ incentives, (ii) prices are centrally determined for many services, and (iii) competition in the market focuses on non-price variables. In this context the CMA has used analysis of referral patterns, patient and GP surveys and internal documents to inform its assessment of the competitive effects of recent NHS hospital mergers. Analysis of GP referral patterns exploits the rich patientlevel data available to provide a relative measure of closeness of competition, conceptually similar to a diversion ratio which is the proportion of sales lost by firm A captured by firm B when firm A increases its price. A high diversion ratio between two firms indicates that they are close alternatives for consumers and therefore the closer competitors. The methodology relies on aggregation of patient-GP decisions to the GP practice level to get a measure of referrals to each local hospital. These GP referrals give a measure of the number or proportion of referrals that may shift to each alternative hospital and provides a ranking of hospitals in the patient/GP’s choice set. Aggregating these rankings by hospital gives an indication
Table 2 Hospital competition in Germany, the Netherlands and England.
Hospital price competition Hospital quality competition Patient choice of provider Room for selective contracting by payers
Germany
Netherlands
England
No
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
Very limited
Allowed
Allowed
of which hospitals are good alternatives for the patients and GPs currently choosing that hospital and therefore an indication of the closeness of competition. In recent cases the CMA has used a cut-off of 30% to identify specialties for further investigation. In RBCH/Poole, the CMA extended this analysis in a number of ways. These included looking at the evolution of hospital referral shares over time and with a particular focus on geographically marginal GP practices. The CMA used natural experiments, looking at the impact on patient/GP choices of quality shocks. The CMA also looked at the proportion of the merger hospitals’ referrals that were from GP practices with choices other than the merger hospitals and also the impact of the merger on the number of marginal GP practices. Taken together with qualitative evidence from internal documents and the views of commissioners, this evidence allowed the CMA to assess how closely the merging hospitals competed as well as the strength of competition from other local hospitals resulting in the decision to prohibit the merger. Some observers were concerned that the prohibition would undermine the financial stability of the hospitals involved. In reaction, the Department of Health commented that the review of mergers is a matter for the independent competition authorities and that they must make their decisions in the best interests of patients. 4. Discussion In Germany, the Netherlands, and England the hospitals are expected to compete. However, from the first part of our comparative analysis it follows that both similarities and differences exist between the three countries. These are summarized in Table 2. Since it can be concluded that in all three countries hospital competition is still emerging and therefore fragile, preventing the creation or strengthening of dominant positions through hospital merger control is of crucial importance. However, the characteristics of the sector seem to present competition authorities with some special challenges and it is sometimes argued that legal frameworks might need to be more flexible in order to be effective. At the same time decision makers need to ensure that the application of competition law is effective and remains consistent and effective across different sectors. This results in a number of important challenges, especially because competition policy in the hospital sector is in a relatively early stage of evolution and thus far from fully developed yet.
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25
In all three countries, the main drivers of hospital mergers include financial pressures and ambitions to concentrate complex surgeries at fewer hospitals to improve quality. As a result, hospital merger activity is part of a debate that involves an assessment of broader public policy questions that go beyond competition: the need to restructure the provision of hospital services to ensure greater efficiency and to meet wider public policy objectives such as improved health care quality. At the same time hospital mergers also reduce market competition, which in the health economics literature generally has been found to lower prices and improve health care quality. This complex context raises the question of the role of competition law and the best way to achieve a trade-off between these different aims. One important response is to ensure the independence of decision makers. In Germany, the BKartA has a very high degree of autonomy and its aims are solely targeted on the protection of competition. It has a very clear and unambiguous stance and defends competition and the protection of consumer interests against interests of other stakeholders. In the Netherlands, the ACM also is an independent administrative body. However, rather than restricting itself to protecting competition the competition authority is very permissive. In its assessment of proposed hospital mergers over the past decade, the impact on market outcomes is mainly based on the shortsighted and subjective views of stakeholders, which may not be well placed to assess the long term effects that might come about from a loss of competition. In England the CMA closely looks at the expected outcomes for patients taking account of the potential for relevant patient benefits. The CMA’s assessment, highlighting the role of patient choice and its effect particularly on non-price outcomes. Summarizing, while all three competition authorities have a shared framework and approach, they interpret it very differently. There are, however, also some important institutional differences among the three countries, notably in respect of the role of competition for the market as compared to competition in the market. In England both models of competition exist, though the CMA tends to focus on the effect of hospital mergers on competition for patients because the commissioners who act as buyers usually have a wider set of options than individual patients. In Germany only competition in the market matters, as selective contracting does not yet play a notable role. This may or may not change in the future. The most recent proposal for hospital sector reform (Krankenhausstrukturgesetz, KHSG, 2015) – currently being discussed in the legislative process – includes a paragraph with a vague and tentative framework for more options to concluded selective contracts focusing on quality criteria. The scope is limited to testing this approach for future decisions on a broader implementation [38,39]. In the Netherlands, the ACM has taken the speculative view that insurers possess considerable countervailing power and that they could deal with highly concentrated hospital markets. Even accounting for these differences there remain very heterogeneous expectations about potential merger outcomes. The BKartA is able to maintain a clear position by – while being in line with SIEC – maintaining a strong focus on the
23
Box 2 In the United States, the Federal Trade Commission (FTC) faced a string of defeats in hospital merger cases in the 1990s while applying a “traditional” approach. The turning point was the Evanston/Highland Park case in 2007 in which the FTC brought two major breakthroughs [41,42]. First, it established the relevance of competition for the market and second, based on comprehensive retrospective merger analysis it cleared the way for the acceptance of merger simulation models. While in the US merger simulation has become widely accepted economic evidence in hospital merger control (e.g. ProMedica/St. Luke’s [43]), this is not yet the case in Europe. In Germany, the robust country specific evidence on modelling hospital choice and analyses of effects of competition on quality [15,44] is lacking. But also in the Netherlands – where scientific studies about modelling patient choice for assessing hospital competition are available [27,28,45] – and England – which currently conducts the most detailed analysis of hospital markets – choice modelling is currently not included in current merger control practice. However, at least in England, the comparatively high number of studies with UK data [6,20,23,24] seems to have strengthened the approach that patient choice and the resulting competition does have positive effects on prices and quality, fostering a consequent merger control regime.
pre-merger competitive conditions. However, this requires considerable confidence and trust in the effectiveness of the pre-existing conditions and is increasingly difficult to uphold in times in which (i) big data is widely used for sophisticated analyses of consumer behaviour [40] and (ii) consumers and politicians want to see evidence on outcomes rather than to rely on the predictions of economic theory. While the latter has become widely accepted in the United States, Germany, the Netherlands and England do not (yet) systematically apply economic modelling in hospital merger control (Box 2). There is clearly an urgent need for European research in this field, particularly in the form of retrospective hospital merger studies, similar to those undertaken by the US Federal Trade Commission [46–48], and structural models of hospital competition, such as the recent model of Gowrisankaran et al., that can be used for merger simulation [49]. The results of this research will result in a better understanding of the functioning of hospital markets and are likely to have a direct impact on hospital merger control. 5. Concluding remarks Observing a clearly pro-competition approach to health policy in Germany, the Netherlands, and England we wanted to evaluate the extent to which this rhetoric is reflected in merger control – a fundamental corner stone of competition policy – in the respective hospital markets. In all three countries, competition has been introduced and/or strengthened in hospital markets over the past decade and policy makers as well as competition authorities have prioritized different modes of competition. The major
24
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25
difference between the approaches adopted by the authorities is evident in the very heterogeneous assessment of potential merger outcomes. While England and Germany – based on recent decisions – seem to handle these cases rather restrictive, the Netherlands accept very high levels of market concentration. This stance on expected outcomes seems to be even more important than the analytical method applied. This is particularly true since once markets have consolidated it is extremely difficult, if not impossible, to restructure a sector to address issues arising from market power. As a result, the decisions taken today by competition authorities shape the health care landscape of the future. Acknowledgements We would like to thank Kate Collyer and Rupprecht Podszun for their valuable contributions and comments as well as all participants of the invitational conference on competition and antitrust in hospital markets (22–23 September 2014, University of Bayreuth) for the interesting discussions. References [1] Gaynor M, Town RJ. Competition in health care markets. In: Pauly MV, Barros PP, McGuire TG, editors. Handbook of health economics, vol. 2, 1st ed. Amsterdam/Boston: Elsevier/North Holland; 2012. p. 499–637. [2] Loozen E. Public healthcare interests require strict competition enforcement. Health Policy 2015;119:882–8. [3] Hentschker C, Mennicken R. The volume-outcome relationship and minimum volume standards – empirical evidence for Germany. Health Economics 2015;24:644–58. [4] EXPH. Competition among health care providers – investigating policy options in the European Union: European Union Expert Panel on effective ways of investing in Health (EXPH); 2015, May 7. Available from: http://ec.europa.eu/health/ expert panel/opinions/docs/008 competition healthcare providers en.pdf [accessed 25.06.15]. [5] Hentschker C, Mennicken R, Schmid A. Defining hospital markets – an application to the German hospital sector. Health Economics Review 2014;4:1–17. [6] Gaynor M, Laudicella M, Propper C. Can governments do it better? Merger mania and hospital outcomes in the English NHS. Journal of Health Economics 2012;31:528–43. [7] Postma J, Roos A. Why healthcare providers merge. Health Econ Policy Law 2015:1–20. [8] SVR. Gutachten 2000/2001 des Sachverständigenrates für die Konzertierte Aktion im Gesundheitswesen: Bedarfsgerechtigkeit und Wirtschaftlichkeit – Band II – Qualitätsentwicklung in Medizin und Pflege. Bundestags-Drucksache 2001;14/5661. [9] SVR. Gutachten 2005 des Sachverständigenrates zur Begutachtung der Entwicklung im Gesundheitswesen: Koordination und Qualität im Gesundheitswesen. Drucksache 2005;15/5670. [10] SVR. Gutachten 2012 des Sachverständigenrates zur Begutachtung der Entwicklung im Gesundheitswesen: Wettbewerb an der Schnittstelle zwischen ambulanter und stationärer Gesundheitsversorgung. Bundestags-Drucksache 2012;17/10323. [11] Monopolkommission. Weniger Staat. mehr Wettbewerb: Gesundheitsmärkte und staatliche Beihilfen in der Wettbewerbsordnung, Siebzehntes Hauptgutachten der Monopolkommission gemäß § 44 Abs. 1 Satz 1 GWB 2006/2007. Bonn; 2008. Available from: http://www.monopolkommission.de/haupt 17/gesamtfassung h17. pdf [accessed 01.12.08]. [12] Wasem J, Geraedts M. Qualität durch Wettbewerb. In: Klauber J, Geraedts M, Friedrich J, Wasem J, editors. Krankenhaus-Report 2011: Schwerpunkt: Qualität durch Wettbewerb. Stuttgart: Schattauer; 2011. p. 3–17. [13] Preusker UK, Müschenich M, Preusker S. Darstellung und Typologie der Marktaustritte von Krankenhäusern – Deutschland 2003 – 2013: Gutachten im Auftrag des GKV-Spitzenverbandes; 2014.
[14] Schmid A, Ulrich V. Consolidation and concentration in the German hospital market: the two sides of the coin. Health Policy 2013;109:301–10. [15] Schmid A. Konsolidierung und Konzentration im Krankenhaussektor: Eine empirische Analyse der Marktstruktur unter Berücksichtigung des Krankenhausträgers. Baden-Baden: Nomos; 2012. [16] SVR. Gutachten 2014 des Sachverständigenrates zur Begutachtung der Entwicklung im Gesundheitswesen: Bedarfsgerechte Versorgung – Perspektiven für ländliche Regionen und ausgewählte Leistungsbereiche. Drucksache 2014;18/1940. [17] Emmert M, Hessemer S, Meszmer N, Sander U. Do German hospital report cards have the potential to improve the quality of care? Health Policy 2014;118:386–95. [18] BKartA. Beschluss des BKartA vom 10.05.05 B 10 – 123/04 – RhönKlinikum AG/Kreiskrankenhäuser Bad Neustadt, Mellrichstadt. Bonn; 2005. Available from: http://www.bundeskartellamt.de/ wDeutsch/download/pdf/Fusion/Fusion05/B10-123-04.pdf [accessed 02.12.08]. [19] OLG Düsseldorf Beschluss des Oberlandesgerichts Düsseldorf vom 11.04.2007 VI-Kart 6/05 (V). Düsseldorf; 2007. Available from: http://www.justiz.nrw.de/nrwe/olgs/duesseldorf/j2007/VI Kart 6 05 V beschluss20070411.html [accessed 01.12.08]. [20] Beckert W, Christensen M, Collyer K. Choice of NHS-funded hospital services in England. Economic Journal 2012;122:400–17. [21] Sivey P. The effect of waiting time and distance on hospital choice for English cataract patients. Health Economics 2012;21: 444–56. [22] Cooper Z, Gibbons S, Jones S, McGuire A. Does competition improve public hospitals’ efficiency? Evidence from a quasi-experiment in the English National Health Service: CEP Discussion Paper number 1125; 2012. [23] Cooper Z, Gibbons S, Jones S, McGuire A. Does hospital competition save lives? Evidence from the English NHS patient choice reforms. Economic Journal 2011;121:F228–60. [24] Gaynor M, Moreno-Serra R, Propper C. Death by market power: reform, competition, and patient outcomes in the National Health Service. American Economic Journal: Economic Policy 2013;5:134–66. [25] Bloom N, Propper C, Seiler S, van Reenan J. The impact of competition on management quality: evidence from public hospitals. Centre for Economic Performance, LSE, Discussion Paper No 983, May 2010 (revised February 2013); 2013. [26] Schut FT, Van de Ven Wynand PMM. Rationing and competition in the Dutch health-care system. Health Economics 2005;14: S59–74. [27] Varkevisser M, van der Geest Stéphanie A, Schut FT. Do patients choose hospitals with high quality ratings? Empirical evidence from the market for angioplasty in the Netherlands. Journal of Health Economics 2012;31:371–8. [28] Beukers Puck DC, Kemp Ron GM, Varkevisser M. Patient hospital choice for hip replacement: empirical evidence from the Netherlands. European Journal of Health Economics 2014;15:927–36. [29] NZa. Medisch specialistische zorg 2014: weergave van de markt 2010–2014 (deel A). Utrecht: Nederlandse Zorgautoriteit; 2014. [30] BKartA. Leitfaden zur Marktbeherrschug in der Fusionskontrolle: Stand 29. März 2012. Bonn; 2012. [31] Varkevisser M, Schut FT. The impact of geographic market definition on the stringency of hospital merger control in Germany and the Netherlands. Health Economics, Policy and Law 2012;7: 363–81. [32] Kuchinke BA, Kallfass HH. Die Praxis der räumlichen Marktabgrenzung bei Krankenhauszusammenschlüssen in den USA und in Deutschland: Eine wettbewerbsökonomische Analyse. Zeitschrift für Wettbewerbsrecht 2007;07:319–37. [33] CC, OFT. Merger Assessment Guidelines: September 2010, CC2 (Revised), OFT1254, A joint publication of the Competition Commission and the Office of Fair Trading; 2010. [34] CMA. CMA guidance on the review of NHS mergers: 13 July 2014; 2014. [35] ACM. ACM prohibits proposed merger between two Dutch hospital groups: press release 15, July 2015; 2015. [36] Loozen E, Varkevisser M, Schut E. Dutch Authority for Consumers and Markets fails to meet the standard of proof in recent hospital merger decisions. European Competition Law Review 2014;35: 16–22. [37] ACM. Assessing mergers and collaborations in hospital care. Available from: https://www.acm.nl/en/download/publication/ ?id=12304 [accessed 23.02.15].
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25 [38] KHSG. Gesetzentwurf der Fraktionen der CDU/CSU und SPD: Entwurf eines Gesetzes zur Reform der Strukturen der Krankenhausversorgung (Krankenhausstrukturgesetz – KHSG). Deutscher Bundestag, Drucksache 2015;18/5372. [39] Bund-Länder-AG. Eckpunkte der Bund-Länder-AG zur Krankenhausreform 2015. Berlin; 2014. Available from: http://www.bmg.bund.de/fileadmin/dateien/Downloads/B/Bund Laender Krankenhaus/Eckpunkte Bund Laender Krankenhaus.pdf [accessed 16.01.15]. [40] Haucap J. Wo die Anwendung des Kartellrechts reformbedürftig ist: FAZ; 12.04.2015. Available from: http://www.faz.net/-gqi-81zoj [accessed 22.04.15]. [41] Ashenfelter O, Hosken D, Vita M, Weinberg M. Retrospective analysis of hospital mergers. International Journal of the Economics of Business 2011;18:5–16. [42] Rice EL. Evanston’s Legacy: a prescription for addressing two-stage competition in hospital merger antitrust analysis. Boston University Law Review 2010;90:431–63. [43] FTC. ProMedica/St. Luke’s: FTC Matter 101 0167, Docket Number 9346; 2012.
25
[44] Steinhauser C. Patientengewinnung im Krankenhaus: Determinanten der Krankenhauswahl und ihre Beeinflussung – Eine theoretische und empirische Analyse. Hamburg: Dr. Kovac; 2015. [45] Varkevisser M, van der Geest SA, Schut FT. Assessing hospital competition when prices don’t matter to patients: the use of time-elasticities. International Journal of Health Care Financing and Economcis 2010;2010:43–60. [46] Tenn S. The price effects of hospital mergers: a case study of the Sutter–Summit transaction. International Journal of the Economics of Business 2011;18:65–82. [47] Haas-Wilson D, Garmon C. Hospital mergers and competitive effects: two retrospective analyses. International Journal of the Economics of Business 2011;18:17–32. [48] Thompson A. The effect of hospital mergers on inpatient prices: a case study of the New Hanover–Cape Fear transaction. International Journal of the Economics of Business 2011;18:91–101. [49] Gowrisankaran G, Nevo A, Town R. Mergers when prices are negotiated: evidence from the hospital industry. American Economic Review 2015;105:172–203.
Contents lists available at ScienceDirect
Health Policy journal homepage: www.elsevier.com/locate/healthpol
Review
Hospital merger control in Germany, the Netherlands and England: Experiences and challenges Andreas Schmid a , Marco Varkevisser b,∗ a b
Health Management (MiG), University of Bayreuth, Germany Institute of Health Policy & Management (iBMG), Erasmus University Rotterdam, The Netherlands
a r t i c l e
i n f o
Article history: Received 28 July 2015 Received in revised form 30 October 2015 Accepted 2 November 2015 Keywords: Hospital mergers Competition policy International comparison
a b s t r a c t Aiming at the efficiency enhancing and quality improving effects of competition, various steps have been undertaken to foster competition in hospital markets. For these mechanisms to work, robust competition policy needs to be enacted and enforced. We compare the hospital markets in Germany, the Netherlands and England regarding their experience with competition and put a special focus on merger control and the stringency of its implementation. Elaborating on the differences in merger control practice we find that despite very similar goals the respective agencies apply very different approaches and take fundamentally different routes when balancing proclaimed benefits of mergers with potential risks of consolidated markets. While the German competition authority has a strong focus on maintaining the preconditions for competition, in the Netherlands we find over the past decade a much stronger focus on hypothesized countervailing buyer power, accepting in turn highly concentrated markets. In England we find the currently most comprehensive analysis of proposed mergers in combination with a clearly positive assessment of the effects of patient choice and competition on prices and quality. All agencies are still reluctant to implement merger simulation models or similarly advanced econometric methods in their appraisal. One very likely reason is a lack of country specific empirical evidence on these matters. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction A growing number of countries have introduced or strengthened competition among health care providers, including hospitals, to improve the functioning of their health care systems. For hospital competition to be successful, the existence of a sufficient number of alternative hospitals for patients and payers to choose from is required. The empirical literature clearly shows that mergers between rivals in concentrated markets are likely to
∗ Corresponding author. E-mail address: [email protected] (M. Varkevisser). http://dx.doi.org/10.1016/j.healthpol.2015.11.002 0168-8510/© 2015 Elsevier Ireland Ltd. All rights reserved.
increase prices. The results for the effect on quality seem to depend on the price setting mechanism. In markets where prices are set by hospitals the empirical evidence is mixed, but competition between hospitals operating in markets with regulated prices has generally been found to have a positive effect on quality [1]. It is therefore argued that when the market mechanism is used for maximizing the public healthcare interests competition enforcement should be strict [2]. In the final opinion of the Expert Panel on effective ways of investing in Health (EXPH), it is recently concluded that the introduction of competition between providers of health care requires, among other conditions, the enforcement of competition rules to prevent the creation, strengthening and abuse of dominant
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25
positions [3]. However, hospital mergers may also benefit patients. For example, there is some evidence that reaching certain volume thresholds improves quality for specific complex hospital services [4]. Overall, the relevant question from the patient’s perspective is whether the benefits of hospital consolidation outweigh the negative effects caused by reduced market competition. Using a semi-structured narrative approach, this paper compares and discusses the experiences with and challenges for hospital merger control in three countries with competition among hospitals: Germany, the Netherlands, and England. In our country comparison, the focus is first on institutional differences related to hospital competition. This is the basis for the subsequent description of the respective approaches used for ex ante hospital merger control, as well as the comparative discussion of results achieved and challenges faced by the competition authorities. Table 1 provides some relevant key indicators for the three countries (i.e. including the United Kingdom rather than England because of the OECD statistics used here), highlighting already some of the structural differences and similarities. Each of the countries included in the paper has experienced consolidation of the hospital industry. In Germany, in addition to a more “pro-market” attitude in health politics, hospital merger activity is predominantly fuelled by payment reforms and shrinking financial resources of municipalities owning public hospitals. The impact of those mergers on the hospital market structure is twofold [5]. First, the formation of multi-site hospital systems on a local level. Second, on a supra-regional level hospital chains arise that are active in various local hospital markets across the country. Consolidation in England has also been driven by financial pressure. However, whether a hospital is merged or not depends not just on (financial) performance but also on national politics related to the National Health Service (NHS). That is, to avoid unpopular closures merger activity between geographically co-located hospitals may also be initiated by the government [6]. In the Netherlands, the most important merger motives for hospitals – as well as other providers of health care – are quality improvements associated with the use of minimum volume thresholds for complex procedures and increased bargaining clout vis-à-vis payers followed by financial and efficiency considerations [7]. 2. Hospital competition 2.1. Germany Over the past 15 years, encouraged by official advisory bodies such as the German Advisory Council on the Assessment of Developments in the Health Care System [8–10] and the Monopoly Commission [11], policy makers have fostered the role of competition in the German hospital sector in multiple ways. For example by increasing transparency through mandatory quality reporting and allowing for some degree of selective contracting between sickness funds and providers. In this context, competition is always intended to achieve higher levels of efficiency and to increase quality of care [12]. The greatest impact on the
17
cost side came from the introduction of the reimbursement system based on Diagnosis Related Groups (DRG) in 2003/4, which forced providers into cost oriented yardstick competition as the case-based lump sum payment is related to the average costs across hospitals. Acute care hospitals are now reimbursed on a case basis and the regulated price is calculated as an average of costs across a sample of hospitals. Cost structures have become significantly more transparent and large numbers of hospitals faced severe financial challenges. A relatively small number of hospitals went out of business and there were a considerable number of mergers, many of which were aimed at keeping all existing hospital sites open. This is partly due to political pressure especially at the local and state level, where (public) hospitals are considered to be important factors for the local economy (e.g. because they are important employers) [13,14]. There is considerable heterogeneity in the degree of hospital market concentration across the 16 German states [15]. Reasons for this heterogeneity include differences in geographic settlement structures, the states’ different influence in hospital planning and the structural legacy differences between former East and West Germany. The average catchment area ranges from 108 km2 in North Rhine-Westphalia over 224 km2 in Bavaria to 703 km2 in Mecklenburg–Vorpommern – the latter two states both having large rural areas. While an average hospital in Bavaria serves just about 40,000 people, other states have larger institutions that serve – as in Saxony – up to 55,000 people on average [16]. Generally speaking, there is excess capacity in metropolitan areas while it is challenging to ensure adequate access to care for patients in rural areas [16]. By 2007 about 40% of German hospitals operated in concentrated or highly concentrated regional hospital markets, with a Herfindahl–Hirschman-Index (HHI) of 0.18 or higher. Concentration is generally higher in rural states, but some more urban areas also have highly concentrated markets [14]. There is even more variation when concentration is analyzed for different procedures; i.e. in disaggregated product markets. While the average HHI for selected elective surgical procedures (e.g. hip or knee replacement) is as low as 0.19, markets for pneumonia or maternity units have an average HHI of more than 0.3 [5]. While the cost oriented incentives of the yardstick based DRG system are extremely strong, the quality component is still trailing behind. Considerable efforts have been undertaken to foster quality based competition, e.g. by the aforementioned mandatory quality reporting of hospitals and an increasing number of websites that provide quality related information. However, according to a recent study by Emmert et al., despite an overall positive trend, only about a third of their respondents were aware that such information existed in the first place [17]. Despite the low awareness of quality indicators, the role of patients is crucial as at the individual provider level increased patient volume leads to higher revenues in the long run. Physicians, including both general practitioners and independent medical specialists, are advised to include the two nearest suitable hospitals in their referral (§§ 39(2), 73(4) SGB V). If patients do not choose a hospital recommended in the referral they could be charged any
18
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25
Table 1 Relevant key indicators for Germany, the Netherlands and the United Kingdom. Germany
Netherlands
United Kingdom
Total population Land area (km2 ) Population density (population per km2 )
81,900,000 357,000 229
16,800,000 33,700 497
63,700,000 242,500 263
Type of health system Heath expenditure (% GDP) Health expenditure per capita ($) Acute care hospitals beds (per 1000 population)
Bismarck 11.0 4,819 5.3
Bismarck 11.1 5,131 3.3
Beveridge 8.5 3,235 2.3
Source: OECD.Stat (accessed on 15 October 2015).
additional costs – an option that is not enforced in practice. The competition authority and the respective courts agree that despite these formal restrictions patients can de facto exercise free hospital choice. The Bundeskartellamt (BKartA – Federal Cartel Office) therefore focuses on competition for patients. The fact that patients and not sickness funds are the relevant customers of hospitals, however, negates a potential conflict of § 69 SGB V (Book V of the Social Code) granting some exceptions from competition law for certain relations between sickness funds and providers [18,19]. In summary, German hospitals compete on the basis of costs with fixed prices and – still on a low level but increasingly – on quality. Merger control focuses on the patients as the customers that need to be protected while third party payers are not relevant in these considerations. 2.2. The Netherlands After decades of top-down health care rationing policies, the Netherlands has opted for a system of regulated competition and private insurance with wide-ranging reforms implemented since the mid-2000s to strengthen the role of market mechanisms [26]. In 2006, competition among health insurers was reinforced with the introduction of the Health Insurance Act making private health insurance mandatory for every Dutch citizen. The basic idea behind the introduction of regulated competition was to give risk-bearing private health insurers appropriate incentives to act as prudent buyers of health services on behalf of their customers. To that end, the insurers are expected to negotiate with health care providers about the price, quality, and/or volume of care allowed. To strengthen insurers’ bargaining strength vis-à-vis providers, they are allowed to contract selectively and use financial incentives for channelling patients to preferred providers. In addition to the fundamental reform of the health insurance system, a gradual deregulation of provider markets was also carried out. This deregulation included the market for hospital services. In anticipation of the major 2006 reform of the Dutch health care system, in February 2005 insurers were allowed to negotiate contracts (including prices) with individual hospitals for a number of routine hospital services, such as cataract surgery and hip replacement. These hospital-insurer negotiations were facilitated by the introduction of a newly developed hospital product classification system based on the diagnoses and subsequent treatment of patients. Since then each patient admitted to a Dutch hospital or visiting a hospital’s
outpatient clinic is categorized into a Diagnosis Treatment Combination (DTC). Each DTC includes all hospital activities and services (both inpatient and outpatient) associated with the patient’s care, from his initial consultation or examination to the final check-up. The proportion of freely negotiable hospital services (labelled the ‘B-segment’) was stepwise expanded from about 10% in 2005 to about 34% of total hospital revenue in 2009. To increase the incentives for efficiency, the room for price negotiations between hospitals and health insurers was further expanded in 2012 to about 70% of hospital turnover. For the remainder of hospital production (labelled the ‘A-segment’), containing the more complex treatments, prices per DTC are still determined by the Dutch Healthcare Authority (NZa) which is the independent administrative body supervising all healthcare markets in the Netherlands. In addition to the insurer-hospital negotiations, hospitals also compete directly for patients. Since the introduction of the new Dutch health care system patients are encouraged to make an active choice between alternative providers, for example by the provision of consumer information about hospital quality. Though the publicly available hospital quality ratings are still far from perfect, empirical research indicates that at least to some extent patients are sensitive to differences in observed hospital quality [27,28]. However, this research also clearly reveals that travel time is still the most important choice determinant. Hence, hospital competition in the Netherlands seems to be restricted to a few neighbouring hospitals. This is also reflected in the fact that the average Dutch hospital has a market share of 50% in its catchment area [29]. The NZa placed any zip code in a hospital’s catchment area if earnings from that zip code were at least D 1. 2.3. England Government initiatives over the last couple of decades have introduced competition and choice in the provision of NHS services. There are two models of competition in the NHS in England: quality competition in the market, which occurs in routine elective (planned) services where patients have a choice between hospitals; and competition for the market, where healthcare providers compete on quality, and sometimes price, to be selected by commissioners as the sole provider of services to patients. In the NHS in England, healthcare services are agreed, planned and monitored by a number of different commissioning bodies. Most of the NHS commissioning budget is now
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25
managed by 211 clinical commissioning groups (CCGs). These are groups of general practices that are responsible for planning healthcare services for their patients and population. Commissioners often have a wider set of choices than patients and so recent merger control has tended to focus on the impact on competition in the market. Patients have a choice of provider in respect of their first consultant-led outpatient appointment for elective care and for maternity services. The costs of all NHS patient care is fully funded by the government, with no out-ofpocket payments by patients, and patients can only access acute healthcare by referral from their general practitioner (GP). Consequently the GP plays an important role advising the patient on which hospital to choose. The NHS provides information to help patients choose, with the advice of their GP. These include the ‘Choose and Book’ website and the NHS Choices website, which shows the Care Quality Commission’s report on quality of care as well as “Trip Advisor” style feedback from patients. Empirical evidence has shown that in addition to travel time, quality of care and waiting times influences hospital choice [20,21]. Therefore, from the demand side perspective the conditions are in place for quality competition to provide incentives for hospitals to compete for patient/GP referrals. These demand side incentives are complemented by incentives on the supply side. Under the Payment by Results (PbR) regime, which covers the majority of acute healthcare, hospitals are paid a fixed national tariff per treatment so that their income depends on the number of patients that they treat. Hospitals in England are managed by acute trusts, some of which have achieved Foundation Trust status. Foundation Trusts are required to provide certain services, but are afforded a degree of operational autonomy and can keep their surpluses and borrow to invest in new and improved service. This gives NHS Foundation Trusts incentives to maximize their income by taking steps to attract patients for profitable specialties, for example by maintaining and improving service quality. NHS Foundation Trusts are regulated by Monitor, which is required to protect and promote the interests of people who use the NHS by promoting the provision of healthcare services. As providers of publicly funded NHS services for patients, Foundation Trusts have many different objectives and healthcare professionals and managers want to deliver high-quality care for their patients. However, these organizations also have the objective of ensuring they receive sufficient income to cover costs. The quality of healthcare provided by a hospital is the outcome of many different decisions that are made at different levels across an organization trading off different factors. The effect of competition on quality where prices are fixed centrally is to focus these spending decisions such that account is taken of the factors that matter to patients and GPs. Empirical studies of the effect of quality competition in the NHS in England provide some evidence that competition is associated with improved quality and efficiency for patients diagnosed with acute myocardial infarction (AMI) [22–24]. One paper looked at the impact of competition on management quality in NHS hospitals and found that management quality is strongly correlated with
19
financial and clinical outcomes (such as survival rates from emergency heart attack admissions) [25]. It is found that higher competition is positively correlated with management quality and that adding a rival hospital increases management quality and increases heart attack survival rates. 3. Hospital merger control 3.1. Germany The regulatory framework for merger control in Germany, including hospital mergers, is given by the Gesetz gegen Wettbewerbsbeschränkungen (GWB – Act against Restraints of Competition) which is enforced by the Bundeskartellamt (BKartA). This law was amended by the 8th Amendment of the GWB in 2013, aligning the substantive test with EU standards, i.e. the SIEC-Test (Significant Impediment to Effective Competition). For most cases, the change is not too fundamental, as the prior core criteria, i.e. the creation or strengthening of a dominant market position, remains the standard example within the SIEC-Test for horizontal mergers [30]. The SIEC-Test allows for the consideration of relevant customer benefits, but it remains to be seen if the BKartA will give more weight to this aspect in its future practice. Currently it seems unlikely that the authority would clear an otherwise harmful merger on the basis of merger-specific benefits to patients. While the evaluation of a potential dominant position is a comprehensive analysis, taking account of numerous factors that determine the actual competitive situation in the market such as barriers for market entry or exit, financial situation of competitors and potential future competition, the merging hospitals’ combined market share is an important first indicator (§ 18(3) GWB). For single companies the threshold for this market power presumption was raised from 33.3% to 40% by the 8th amendment of the GWB (§ 18(4) GWB). For a group up to three firms the threshold is 50% and for up to five firms 66.6% (collective or oligopolistic market dominance). It is important to note that not only full mergers but also acquisitions are subject to the full assessment even if full control is not acquired. For example, also joint ventures and minority shareholding can fulfil the criteria outlined in detail in § 37 GWB. The turnover thresholds for merger control, deciding whether the merger falls under the jurisdiction of the BKartA, are of particular relevance for the hospital sector. The threshold is supposed to ensure that only mergers of macro-economic relevance are monitored by the BKartA. Among other thresholds, § 35 GWB states that the aggregated turnover of the involved companies must exceed D 500 million. In the hospital sector that is characterized by regional markets and comprises a large number of local and regional hospital systems this has been a concern, as many mergers that are likely to create market power in a regional market remain under the radar. However, up to now suggestions for improvement, such as using a turnover multiplier to assess hospital mergers [11], have not been implemented. This may result in a rather uncomfortable situation. That is, when two independent local hospitals merge without exceeding the turnover threshold
20
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25
Box 1 The Elzinga/Hogarty approach is easy to apply and only requires commonly available patient flow data. It begins with a narrowly defined market and then expands the boundary until threshold conditions are met for both imports and exports of hospital services. In various US court cases, the analysis focused on identifying geographic areas such that both statistics are either below 25% (“weak market”) or 10% (“strong market”). The central methodological problem underlying the approach is that the presence of a minority of travelling patients does not necessarily discipline hospitals from exercising market power over the silent majority of non-travelling patients.
their consolidation will not be subject to merger control scrutiny, even though it may lead to a dominant position in the relevant market. On the other hand, when one of the merging hospitals belongs to a hospital chain that also owns hospitals in other parts of the country the threshold is certainly exceeded for this economic entity and therefore the merger will always be assessed by the competition agency while it does not have an impact on local/regional hospital competition. Furthermore, there is some controversy around the fact that for municipalities that own a hospital the turnover of affiliated municipal companies such as Sparkasse (mutual savings bank) or waste management are also taken into account [11]. To calculate the market shares – which are calculated on the number of patients treated – the relevant market needs to be defined. For the geographic delineation of hospital markets the BKartA looks at patient flows on zip code level and tries to identify largely self-contained areas. The starting points are the hospitals of the merging entities. In a first step, the catchment area is determined from a supply side perspective and areas with homogeneous market conditions that differ from neighbouring areas are identified. In a second step, the share of patients from these areas that are served by the corresponding hospitals are calculated from a demand side perspective. The focus is on the inflow of patients into the aforementioned areas. Other than the widely criticized Elzinga/Hogarty test (Box 1), this approach usually results in rather narrowly defined geographic markets [31]. There are no official thresholds for the definition of self-contained areas and for reasons of confidentiality only aggregated data are published. The law also gives the BKartA the flexibility to consider further aspects in the evaluation of the competitive situation on an ad hoc basis [32]. The BKartA uses a cluster market approach to define relevant product markets, defining all services under the label “acute inpatient hospital care”. A slightly more detailed analysis will be conducted on an ad hoc basis, for example if one of the competitors is a dedicated orthopaedic or cardiac hospital. Rehabilitation and ambulatory care are considered to be separate markets. This cluster market approach results in rather broad product markets, despite the availability of very detailed transaction data. As a result, the BKartA opts for a higher risk of under-enforcement rather than over-enforcement. This has been discussed critically
by courts [19], but to date the BKartA has been very successful in defending its decisions. Hence, in current practice market definition has been determinative. The competition authority has the flexibility to also consider other factors affecting the post-merger level of competition, but so far these factors are of secondary importance at most. 3.2. The Netherlands In the Netherlands, the general Competition Act (Mw) provides the regulatory framework for hospital merger control. The Mw was introduced in 1998 and is modelled after the competition rules in the EC Treaty and subsequent legislation. It includes a prohibition on cartels; a prohibition on the abuse of a dominant position; and a preventive merger control regime. It is enforced by the Netherlands Authority for Consumers and Markets (ACM), an independent administrative body; i.e. ACM is part of the Dutch central government but does not belong to any ministry. Under the Mw, mergers between any firms (including hospitals) whose combined and individual turnovers exceed the thresholds that are in force are subject to notification and prior approval by ACM. This requirement applies to all mergers between companies whose combined turnover has exceeded D 113 million in the preceding calendar year, with at least D 30 million realized in the Netherlands by at least two of the companies involved. Thresholds for the health care industry were temporarily lowered in January 2008 to D 55 million and D 10 million respectively. This was because geographic markets for health care tend to be small and the emerging competition in the Dutch health care system is fragile since competition among providers has only gradually been introduced since 2005. A third threshold was added to prevent these thresholds from applying to mergers involving companies whose health care services are only a small part of their business. This requires that for at least two of the companies involved turnover from health care services alone must exceed D 5.5 million. The lowered health care specific thresholds are currently still effective. The ACM starts a general review when it is notified about a proposed hospital merger. If the merger is not likely to be anticompetitive, the ACM clears the merger. Otherwise it decides that a license is required. A more substantial assessment of the proposed merger follows when the merging hospitals do need a license and also apply for it. According to Section 41.2 of the Mw, ACM will eventually prohibit a merger “if, as a result of the proposed concentration, effective competition on the Dutch market or a part thereof would be appreciably impeded, specifically as a result of the creation or strengthening of a dominant economic position.” On the basis of supranational European guidelines, a dominant position is defined as “a position of one or more undertakings that enables them to prevent effective competition being maintained on the Dutch market or a part thereof, by giving them the power to behave to an appreciable extent independently of their competitors, their suppliers, their customers or end-users” (Mw, Section 1.i). When assessing the likely competitive effects of a proposed hospital merger, ACM takes into account the non-binding opinion of the Dutch Healthcare Authority
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25
(NZa). In its opinion, the NZa uses merger simulation to predict the price effect of the consolidation. Note that Dutch hospitals also require formal permission from the NZa for their merger plans. This requirement, however, focuses on the merger process and the accessibility of some essential hospital services (24/7 emergency care) and does not involve an assessment of the likely competitive effects. Before 2004 proposed hospital mergers were not assessed in the Netherlands since it was argued that hospitals were not able to compete due to strict supply and price regulation. This changed when in January 2004 the competition authority published a position document stating that as a result of institutional and regulatory changes in the Dutch health care system the economic and legal context by then offered scope for competition among hospitals. Until July 2015, 27 hospital mergers have been subject to an ex ante assessment of which the first 26 cases were all cleared after an initial or more substantial investigation. Only very recently ACM for the first time prohibited a proposed hospital merger involving two hospital groups in the south-western part of the Netherlands, near the cities of Dordrecht and Gorinchem, because they are strong competitors of each other [35]. Obviously, it is too soon to say whether this decision marks the beginning of a more stringent approach to hospital mergers in the Netherlands. As a result of the permissive policy that can be observed over the past decade, the number of general hospitals has steadily declined over the years from 90 in 2004 to little more than 80 in 2014 and, based on the merger plans already made public, local and regional hospital markets are likely to become even more concentrated. It may therefore not come as a surprise that hospital merger control in the Netherlands has been criticized for being too permissive [31,36]. In some cases, mergers are even permitted on very questionable grounds and without a proper definition of the relevant product and geographic market. Without further investigation in hospital merger decisions typically two separate product markets are identified: one for inpatient hospital services and one for outpatient hospital services. For defining the relevant geographic market ACM starts by looking at patient flows. After this the size of the geographical market is most often left open-ended because it is argued that current patient flows are not indicative for patients’ willingness to travel in the future when quality differences across hospitals are expected to become more transparent. The approval of the merger between two hospitals in the city of Tilburg clearly illustrates the fundamental flaws of hospital merger control in the Netherlands [36]. The Tilburg merger was cleared by ACM in November 2012. While based on patient flow data combined market shares ranged between 60–70% and 70–80%, and substantial price increases were predicted by the NZa, the merger was cleared using speculative claims about patients’ willingness to travel and health insurers’ future countervailing buyer power, and with the acceptance of a non-binding temporary price ceiling. In this case, ACM based its decision on the opinions put forward by the health insurers without carefully examining these. The fact that, as can be concluded from current patient flows, other hospitals did not exert competitive pressure on the merging hospitals was
21
ignored. In its new guidelines for assessing mergers and collaborations in health care, issued in 2013, ACM officially confirms this questionable approach: “When assessing a concentration’s implications, the arguments put forward by insurers and patient organizations will be central. The quantitative analyses based on patient flows with regard to the extent to which the merger hospitals exert competitive pressure on each other primarily act as a first scan.” [37]. 3.3. England After taking over many of the functions of the Competition Commission and certain consumer functions of the Office of Fair Trading, as amended by the Enterprise and Regulatory Reform Act 2013, from 1 April 2014 the Competition and Markets Authority (CMA) is responsible for merger control in England, including mergers of NHS Foundation Trusts. Monitor has an important role advising the CMA on possible relevant patient benefits which could outweigh any substantial reduction in hospital competition and choice. The CMA must decide whether a relevant merger situation gives rise to a substantial lessening of competition (SLC) [33]. Patient interests are at the heart of any merger review. The CMA must decide whether a merger is overall in the interests of patients and takes into account potential improvements to services. Choice for patients and commissioners, and competition between hospitals to attract them, helps to ensure that providers have incentives to improve the quality and efficiency of their services. The introduction of greater choice for patients and commissioners and the greater independence of providers means that merger control has become more relevant and the CMA must make sure that changes improve outcomes for people who use the services. Sometimes a merger between competing hospitals may decrease the hospital’s incentives to improve their services, leading to worse outcomes for patients. On the other hand, sometimes a merger may be the best way of delivering certain benefits to patients in a timely manner. The merger review process is designed to examine both of these aspects and determine on balance what is in the overall interests of patients. Until July 2015, the CMA has reviewed six NHS hospital mergers. Four of these were cleared but one, the proposed merger of Royal Bournemouth and Christchurch NHS FT and Poole NHS FT, was prohibited following an indepth investigation. Currently, one merger is the subject of an in-depth investigation. The CMA and Monitor work closely together to ensure the process is useable and efficient and focuses on getting the right outcome. Monitor provides early upfront advice to FTs considering mergers. It is hoped that this early engagement will result in fewer “problematic” mergers which are not in the overall interests of patients. The CMA and Monitor are also working together to ensure that the process is well understood and useful; the CMA has recently published guidance on the review of healthcare mergers, and the CMA and Monitor have jointly published a short guide for hospitals [34]. The CMA has developed a methodology for the review of hospital mergers. In all five of the decisions since 2012, for the purposes of its competitive assessment it has treated
22
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25
each medical specialty as a separate market, with separate inpatient and outpatient services within each specialty and separate elective and non-elective services. Demand side factors give the starting point of narrowly delineated product markets, covering each different service. Supply side factors, particularly the existence of resources such as clinicians, equipment and facilities that are common across services, allow for the aggregation of individual services into broader specialty level markets. The CMA has not widened the relevant product market beyond specialties because of limits to supply side substitution from one specialty into another, including the need to invest in new staff, facilities and equipment as well as the difficulties of small scale expansion that arise because a non-negligible proportion of costs are indivisible and hospitals need to achieve economies of scale. The CMA has adopted a pragmatic approach to geographic market definition, with a focus on the competitive assessment. It has used isochrones based on catchment areas as the starting point for the competitive assessment. Catchment areas show the area from which a large proportion of patients originate. For example, in the RBCH/PH merger, the CMA found 80% catchment areas of about 20 min and used isochrones (lines on a map showing the points at a given travel time) around each hospital to identify the geographic areas from which the hospitals attracted the majority of their patients and therefore the area over which they competed. There can be significant variation across specialties and the CMA has typically adopted sensitivity analysis to capture these differences. As a general principle, a catchment area is typically narrower than the area over which a hypothetical monopolist could profitably sustain a small but significant price increase (or quality deterioration). As a result, the starting point for the competitive assessment is the areas defined by catchment areas analysis, with account taken of constraints from hospitals located further away. The assessment of the impact of hospital mergers presents particular challenges, particularly to using standard merger control techniques, because (i) regulation and institutional design both play a significant role in shaping providers’ incentives, (ii) prices are centrally determined for many services, and (iii) competition in the market focuses on non-price variables. In this context the CMA has used analysis of referral patterns, patient and GP surveys and internal documents to inform its assessment of the competitive effects of recent NHS hospital mergers. Analysis of GP referral patterns exploits the rich patientlevel data available to provide a relative measure of closeness of competition, conceptually similar to a diversion ratio which is the proportion of sales lost by firm A captured by firm B when firm A increases its price. A high diversion ratio between two firms indicates that they are close alternatives for consumers and therefore the closer competitors. The methodology relies on aggregation of patient-GP decisions to the GP practice level to get a measure of referrals to each local hospital. These GP referrals give a measure of the number or proportion of referrals that may shift to each alternative hospital and provides a ranking of hospitals in the patient/GP’s choice set. Aggregating these rankings by hospital gives an indication
Table 2 Hospital competition in Germany, the Netherlands and England.
Hospital price competition Hospital quality competition Patient choice of provider Room for selective contracting by payers
Germany
Netherlands
England
No
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
Very limited
Allowed
Allowed
of which hospitals are good alternatives for the patients and GPs currently choosing that hospital and therefore an indication of the closeness of competition. In recent cases the CMA has used a cut-off of 30% to identify specialties for further investigation. In RBCH/Poole, the CMA extended this analysis in a number of ways. These included looking at the evolution of hospital referral shares over time and with a particular focus on geographically marginal GP practices. The CMA used natural experiments, looking at the impact on patient/GP choices of quality shocks. The CMA also looked at the proportion of the merger hospitals’ referrals that were from GP practices with choices other than the merger hospitals and also the impact of the merger on the number of marginal GP practices. Taken together with qualitative evidence from internal documents and the views of commissioners, this evidence allowed the CMA to assess how closely the merging hospitals competed as well as the strength of competition from other local hospitals resulting in the decision to prohibit the merger. Some observers were concerned that the prohibition would undermine the financial stability of the hospitals involved. In reaction, the Department of Health commented that the review of mergers is a matter for the independent competition authorities and that they must make their decisions in the best interests of patients. 4. Discussion In Germany, the Netherlands, and England the hospitals are expected to compete. However, from the first part of our comparative analysis it follows that both similarities and differences exist between the three countries. These are summarized in Table 2. Since it can be concluded that in all three countries hospital competition is still emerging and therefore fragile, preventing the creation or strengthening of dominant positions through hospital merger control is of crucial importance. However, the characteristics of the sector seem to present competition authorities with some special challenges and it is sometimes argued that legal frameworks might need to be more flexible in order to be effective. At the same time decision makers need to ensure that the application of competition law is effective and remains consistent and effective across different sectors. This results in a number of important challenges, especially because competition policy in the hospital sector is in a relatively early stage of evolution and thus far from fully developed yet.
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25
In all three countries, the main drivers of hospital mergers include financial pressures and ambitions to concentrate complex surgeries at fewer hospitals to improve quality. As a result, hospital merger activity is part of a debate that involves an assessment of broader public policy questions that go beyond competition: the need to restructure the provision of hospital services to ensure greater efficiency and to meet wider public policy objectives such as improved health care quality. At the same time hospital mergers also reduce market competition, which in the health economics literature generally has been found to lower prices and improve health care quality. This complex context raises the question of the role of competition law and the best way to achieve a trade-off between these different aims. One important response is to ensure the independence of decision makers. In Germany, the BKartA has a very high degree of autonomy and its aims are solely targeted on the protection of competition. It has a very clear and unambiguous stance and defends competition and the protection of consumer interests against interests of other stakeholders. In the Netherlands, the ACM also is an independent administrative body. However, rather than restricting itself to protecting competition the competition authority is very permissive. In its assessment of proposed hospital mergers over the past decade, the impact on market outcomes is mainly based on the shortsighted and subjective views of stakeholders, which may not be well placed to assess the long term effects that might come about from a loss of competition. In England the CMA closely looks at the expected outcomes for patients taking account of the potential for relevant patient benefits. The CMA’s assessment, highlighting the role of patient choice and its effect particularly on non-price outcomes. Summarizing, while all three competition authorities have a shared framework and approach, they interpret it very differently. There are, however, also some important institutional differences among the three countries, notably in respect of the role of competition for the market as compared to competition in the market. In England both models of competition exist, though the CMA tends to focus on the effect of hospital mergers on competition for patients because the commissioners who act as buyers usually have a wider set of options than individual patients. In Germany only competition in the market matters, as selective contracting does not yet play a notable role. This may or may not change in the future. The most recent proposal for hospital sector reform (Krankenhausstrukturgesetz, KHSG, 2015) – currently being discussed in the legislative process – includes a paragraph with a vague and tentative framework for more options to concluded selective contracts focusing on quality criteria. The scope is limited to testing this approach for future decisions on a broader implementation [38,39]. In the Netherlands, the ACM has taken the speculative view that insurers possess considerable countervailing power and that they could deal with highly concentrated hospital markets. Even accounting for these differences there remain very heterogeneous expectations about potential merger outcomes. The BKartA is able to maintain a clear position by – while being in line with SIEC – maintaining a strong focus on the
23
Box 2 In the United States, the Federal Trade Commission (FTC) faced a string of defeats in hospital merger cases in the 1990s while applying a “traditional” approach. The turning point was the Evanston/Highland Park case in 2007 in which the FTC brought two major breakthroughs [41,42]. First, it established the relevance of competition for the market and second, based on comprehensive retrospective merger analysis it cleared the way for the acceptance of merger simulation models. While in the US merger simulation has become widely accepted economic evidence in hospital merger control (e.g. ProMedica/St. Luke’s [43]), this is not yet the case in Europe. In Germany, the robust country specific evidence on modelling hospital choice and analyses of effects of competition on quality [15,44] is lacking. But also in the Netherlands – where scientific studies about modelling patient choice for assessing hospital competition are available [27,28,45] – and England – which currently conducts the most detailed analysis of hospital markets – choice modelling is currently not included in current merger control practice. However, at least in England, the comparatively high number of studies with UK data [6,20,23,24] seems to have strengthened the approach that patient choice and the resulting competition does have positive effects on prices and quality, fostering a consequent merger control regime.
pre-merger competitive conditions. However, this requires considerable confidence and trust in the effectiveness of the pre-existing conditions and is increasingly difficult to uphold in times in which (i) big data is widely used for sophisticated analyses of consumer behaviour [40] and (ii) consumers and politicians want to see evidence on outcomes rather than to rely on the predictions of economic theory. While the latter has become widely accepted in the United States, Germany, the Netherlands and England do not (yet) systematically apply economic modelling in hospital merger control (Box 2). There is clearly an urgent need for European research in this field, particularly in the form of retrospective hospital merger studies, similar to those undertaken by the US Federal Trade Commission [46–48], and structural models of hospital competition, such as the recent model of Gowrisankaran et al., that can be used for merger simulation [49]. The results of this research will result in a better understanding of the functioning of hospital markets and are likely to have a direct impact on hospital merger control. 5. Concluding remarks Observing a clearly pro-competition approach to health policy in Germany, the Netherlands, and England we wanted to evaluate the extent to which this rhetoric is reflected in merger control – a fundamental corner stone of competition policy – in the respective hospital markets. In all three countries, competition has been introduced and/or strengthened in hospital markets over the past decade and policy makers as well as competition authorities have prioritized different modes of competition. The major
24
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25
difference between the approaches adopted by the authorities is evident in the very heterogeneous assessment of potential merger outcomes. While England and Germany – based on recent decisions – seem to handle these cases rather restrictive, the Netherlands accept very high levels of market concentration. This stance on expected outcomes seems to be even more important than the analytical method applied. This is particularly true since once markets have consolidated it is extremely difficult, if not impossible, to restructure a sector to address issues arising from market power. As a result, the decisions taken today by competition authorities shape the health care landscape of the future. Acknowledgements We would like to thank Kate Collyer and Rupprecht Podszun for their valuable contributions and comments as well as all participants of the invitational conference on competition and antitrust in hospital markets (22–23 September 2014, University of Bayreuth) for the interesting discussions. References [1] Gaynor M, Town RJ. Competition in health care markets. In: Pauly MV, Barros PP, McGuire TG, editors. Handbook of health economics, vol. 2, 1st ed. Amsterdam/Boston: Elsevier/North Holland; 2012. p. 499–637. [2] Loozen E. Public healthcare interests require strict competition enforcement. Health Policy 2015;119:882–8. [3] Hentschker C, Mennicken R. The volume-outcome relationship and minimum volume standards – empirical evidence for Germany. Health Economics 2015;24:644–58. [4] EXPH. Competition among health care providers – investigating policy options in the European Union: European Union Expert Panel on effective ways of investing in Health (EXPH); 2015, May 7. Available from: http://ec.europa.eu/health/ expert panel/opinions/docs/008 competition healthcare providers en.pdf [accessed 25.06.15]. [5] Hentschker C, Mennicken R, Schmid A. Defining hospital markets – an application to the German hospital sector. Health Economics Review 2014;4:1–17. [6] Gaynor M, Laudicella M, Propper C. Can governments do it better? Merger mania and hospital outcomes in the English NHS. Journal of Health Economics 2012;31:528–43. [7] Postma J, Roos A. Why healthcare providers merge. Health Econ Policy Law 2015:1–20. [8] SVR. Gutachten 2000/2001 des Sachverständigenrates für die Konzertierte Aktion im Gesundheitswesen: Bedarfsgerechtigkeit und Wirtschaftlichkeit – Band II – Qualitätsentwicklung in Medizin und Pflege. Bundestags-Drucksache 2001;14/5661. [9] SVR. Gutachten 2005 des Sachverständigenrates zur Begutachtung der Entwicklung im Gesundheitswesen: Koordination und Qualität im Gesundheitswesen. Drucksache 2005;15/5670. [10] SVR. Gutachten 2012 des Sachverständigenrates zur Begutachtung der Entwicklung im Gesundheitswesen: Wettbewerb an der Schnittstelle zwischen ambulanter und stationärer Gesundheitsversorgung. Bundestags-Drucksache 2012;17/10323. [11] Monopolkommission. Weniger Staat. mehr Wettbewerb: Gesundheitsmärkte und staatliche Beihilfen in der Wettbewerbsordnung, Siebzehntes Hauptgutachten der Monopolkommission gemäß § 44 Abs. 1 Satz 1 GWB 2006/2007. Bonn; 2008. Available from: http://www.monopolkommission.de/haupt 17/gesamtfassung h17. pdf [accessed 01.12.08]. [12] Wasem J, Geraedts M. Qualität durch Wettbewerb. In: Klauber J, Geraedts M, Friedrich J, Wasem J, editors. Krankenhaus-Report 2011: Schwerpunkt: Qualität durch Wettbewerb. Stuttgart: Schattauer; 2011. p. 3–17. [13] Preusker UK, Müschenich M, Preusker S. Darstellung und Typologie der Marktaustritte von Krankenhäusern – Deutschland 2003 – 2013: Gutachten im Auftrag des GKV-Spitzenverbandes; 2014.
[14] Schmid A, Ulrich V. Consolidation and concentration in the German hospital market: the two sides of the coin. Health Policy 2013;109:301–10. [15] Schmid A. Konsolidierung und Konzentration im Krankenhaussektor: Eine empirische Analyse der Marktstruktur unter Berücksichtigung des Krankenhausträgers. Baden-Baden: Nomos; 2012. [16] SVR. Gutachten 2014 des Sachverständigenrates zur Begutachtung der Entwicklung im Gesundheitswesen: Bedarfsgerechte Versorgung – Perspektiven für ländliche Regionen und ausgewählte Leistungsbereiche. Drucksache 2014;18/1940. [17] Emmert M, Hessemer S, Meszmer N, Sander U. Do German hospital report cards have the potential to improve the quality of care? Health Policy 2014;118:386–95. [18] BKartA. Beschluss des BKartA vom 10.05.05 B 10 – 123/04 – RhönKlinikum AG/Kreiskrankenhäuser Bad Neustadt, Mellrichstadt. Bonn; 2005. Available from: http://www.bundeskartellamt.de/ wDeutsch/download/pdf/Fusion/Fusion05/B10-123-04.pdf [accessed 02.12.08]. [19] OLG Düsseldorf Beschluss des Oberlandesgerichts Düsseldorf vom 11.04.2007 VI-Kart 6/05 (V). Düsseldorf; 2007. Available from: http://www.justiz.nrw.de/nrwe/olgs/duesseldorf/j2007/VI Kart 6 05 V beschluss20070411.html [accessed 01.12.08]. [20] Beckert W, Christensen M, Collyer K. Choice of NHS-funded hospital services in England. Economic Journal 2012;122:400–17. [21] Sivey P. The effect of waiting time and distance on hospital choice for English cataract patients. Health Economics 2012;21: 444–56. [22] Cooper Z, Gibbons S, Jones S, McGuire A. Does competition improve public hospitals’ efficiency? Evidence from a quasi-experiment in the English National Health Service: CEP Discussion Paper number 1125; 2012. [23] Cooper Z, Gibbons S, Jones S, McGuire A. Does hospital competition save lives? Evidence from the English NHS patient choice reforms. Economic Journal 2011;121:F228–60. [24] Gaynor M, Moreno-Serra R, Propper C. Death by market power: reform, competition, and patient outcomes in the National Health Service. American Economic Journal: Economic Policy 2013;5:134–66. [25] Bloom N, Propper C, Seiler S, van Reenan J. The impact of competition on management quality: evidence from public hospitals. Centre for Economic Performance, LSE, Discussion Paper No 983, May 2010 (revised February 2013); 2013. [26] Schut FT, Van de Ven Wynand PMM. Rationing and competition in the Dutch health-care system. Health Economics 2005;14: S59–74. [27] Varkevisser M, van der Geest Stéphanie A, Schut FT. Do patients choose hospitals with high quality ratings? Empirical evidence from the market for angioplasty in the Netherlands. Journal of Health Economics 2012;31:371–8. [28] Beukers Puck DC, Kemp Ron GM, Varkevisser M. Patient hospital choice for hip replacement: empirical evidence from the Netherlands. European Journal of Health Economics 2014;15:927–36. [29] NZa. Medisch specialistische zorg 2014: weergave van de markt 2010–2014 (deel A). Utrecht: Nederlandse Zorgautoriteit; 2014. [30] BKartA. Leitfaden zur Marktbeherrschug in der Fusionskontrolle: Stand 29. März 2012. Bonn; 2012. [31] Varkevisser M, Schut FT. The impact of geographic market definition on the stringency of hospital merger control in Germany and the Netherlands. Health Economics, Policy and Law 2012;7: 363–81. [32] Kuchinke BA, Kallfass HH. Die Praxis der räumlichen Marktabgrenzung bei Krankenhauszusammenschlüssen in den USA und in Deutschland: Eine wettbewerbsökonomische Analyse. Zeitschrift für Wettbewerbsrecht 2007;07:319–37. [33] CC, OFT. Merger Assessment Guidelines: September 2010, CC2 (Revised), OFT1254, A joint publication of the Competition Commission and the Office of Fair Trading; 2010. [34] CMA. CMA guidance on the review of NHS mergers: 13 July 2014; 2014. [35] ACM. ACM prohibits proposed merger between two Dutch hospital groups: press release 15, July 2015; 2015. [36] Loozen E, Varkevisser M, Schut E. Dutch Authority for Consumers and Markets fails to meet the standard of proof in recent hospital merger decisions. European Competition Law Review 2014;35: 16–22. [37] ACM. Assessing mergers and collaborations in hospital care. Available from: https://www.acm.nl/en/download/publication/ ?id=12304 [accessed 23.02.15].
A. Schmid, M. Varkevisser / Health Policy 120 (2016) 16–25 [38] KHSG. Gesetzentwurf der Fraktionen der CDU/CSU und SPD: Entwurf eines Gesetzes zur Reform der Strukturen der Krankenhausversorgung (Krankenhausstrukturgesetz – KHSG). Deutscher Bundestag, Drucksache 2015;18/5372. [39] Bund-Länder-AG. Eckpunkte der Bund-Länder-AG zur Krankenhausreform 2015. Berlin; 2014. Available from: http://www.bmg.bund.de/fileadmin/dateien/Downloads/B/Bund Laender Krankenhaus/Eckpunkte Bund Laender Krankenhaus.pdf [accessed 16.01.15]. [40] Haucap J. Wo die Anwendung des Kartellrechts reformbedürftig ist: FAZ; 12.04.2015. Available from: http://www.faz.net/-gqi-81zoj [accessed 22.04.15]. [41] Ashenfelter O, Hosken D, Vita M, Weinberg M. Retrospective analysis of hospital mergers. International Journal of the Economics of Business 2011;18:5–16. [42] Rice EL. Evanston’s Legacy: a prescription for addressing two-stage competition in hospital merger antitrust analysis. Boston University Law Review 2010;90:431–63. [43] FTC. ProMedica/St. Luke’s: FTC Matter 101 0167, Docket Number 9346; 2012.
25
[44] Steinhauser C. Patientengewinnung im Krankenhaus: Determinanten der Krankenhauswahl und ihre Beeinflussung – Eine theoretische und empirische Analyse. Hamburg: Dr. Kovac; 2015. [45] Varkevisser M, van der Geest SA, Schut FT. Assessing hospital competition when prices don’t matter to patients: the use of time-elasticities. International Journal of Health Care Financing and Economcis 2010;2010:43–60. [46] Tenn S. The price effects of hospital mergers: a case study of the Sutter–Summit transaction. International Journal of the Economics of Business 2011;18:65–82. [47] Haas-Wilson D, Garmon C. Hospital mergers and competitive effects: two retrospective analyses. International Journal of the Economics of Business 2011;18:17–32. [48] Thompson A. The effect of hospital mergers on inpatient prices: a case study of the New Hanover–Cape Fear transaction. International Journal of the Economics of Business 2011;18:91–101. [49] Gowrisankaran G, Nevo A, Town R. Mergers when prices are negotiated: evidence from the hospital industry. American Economic Review 2015;105:172–203.