Debates over the legal interpretation of trade treaty (WTO and NAFTA) exemption clauses for public services display a common pattern. Critics of trade agreements argue that these clauses are likely to be narrowly interpreted, providing scant protection from international trade rules to public health care. Defenders usually argue that they will be given a reasonably expansive definition and that trade obligations (at least the more onerous WTO national treatment obligations) will generally not apply to public health care services. This paper argues that while this WTO optimism is well-founded when viewed from a static perspective, the protection afforded by exemption clauses shrinks with the expansion of market elements in health care. Hence, the major implication of such “carve-outs” for health policy makers will not be the liberty to engage in “business as usual”, but rather the need to assess the trade-related risks associated with market-based reform in the future. This paper analyzes the WTO and NAFTA provisions limiting the application of these trade agreements to the health care sector.
It is now commonly recognised that international economic agreements – whether multilateral agreements under the auspices of the World Trade Organisation (WTO), or bilateral or regional trade agreements, such as the North American Free Trade Agreement (NAFTA) – are rapidly evolving into instruments that operate “inside the borders” of countries, affecting domestic policy through the extension of trading principles to the new areas of investment, intellectual property and services. Indeed, if there is one factor that most clearly distinguishes the “new wave” of globalisation in the early 21st century from earlier epochs, it may well be the pressures for reciprocity, harmonisation and governance between advanced administrative and regulatory regimes that did not exist in the simpler world of “small” states engaged in “free trade” (in goods) a century or more ago.
The Uruguay round of international trade negotiations that created the WTO and the General Agreement on Trade in Services (GATS) not only extended international trading rules to services, an area that was mostly excluded from the General Agreement on Trade and Tariffs (GATT), but in so doing, actually redefined the scope of the international trade regime in a novel way to include not just traditional cross-border trade, but also three non-traditional modes of supply. The breadth of these categories explains why the GATS has been referred to simultaneously as a trade agreement, a multilateral investment agreement, and a labour mobility agreement. It also suggests the potential for liberalisation of health insurance and other health services, as well as escalating transformations of the broader economic, political, and regulatory context of health care. It is, therefore, unsurprising that the launch of the WTO’s initial round of negotiations under the GATS in February 2000 occasioned a good deal of critical commentary from Non-Governmental Organisations (NGOs) and others concerning the possible deleterious impacts of multilateral trade liberalisation on public health systems and other public services. The fear is that international trade agreements constrain government action in a way that could adversely affect either the quality, universality or equity of access to essential medical services – by the entry of foreign for-profit insurers, hospitals, physicians and nurses; or by rules that make the future expansion of public health care more difficult or expensive because of obligations to foreign suppliers. This can be seen as part of a larger analysis of trade agreements serving as “a conditioning institutional framework that promotes and consolidates neo-liberal restructuring” (Grinspun and Kreklewich 1994, 33; McBride 2001, 103).
The NAFTA, a regional trading agreement built upon the foundations of the earlier bilateral Canada-US Free Trade Agreement (FTA), pioneered the inclusion of service, investment and intellectual property provisions in major international economic agreements. One particularly important innovation of the NAFTA was its controversial Chapter 11 on investor’s rights, which enables corporations based in one NAFTA country to directly challenge the actions of foreign governments without “their” government acting as an intermediary. Critics have worried that, even if trade and investment agreements can play a constructive role in reducing poverty and fostering sustainable development around the world, legal challenges from private actors are less likely to be restrained by reciprocal concerns for environmental, labour, or social policy standards. The early results in such cases as Ethyl, Myers, Methanex and Metalclad lend credence to these worries. Foreign investors and their companies operating in host states are given privileged access to government decision-makers on actual or proposed measures that have potential impacts on investment, along with a potential hammer with which to influence policy decisions.
One way to reconcile the demand for international trade liberalisation and its extension to services with the public concern to preserve policy autonomy is to include within trade agreements sectoral exemptions that would shield essential public services from application of free trade principles. The two most important principles are the most favoured nation principle (MFN), or external non-discrimination by a Member country among its foreign trading partners, and the national treatment principle (NT), or non-discrimination between foreign and domestic interests inside of a Member country. In the FTA and NAFTA, the direct provision of health, educational and social services were expressly exempted from both types of obligation, but future rules governing management services (including healthcare management, to the extent that the government allowed them to be privatised) were not (Doern and Tomlin 1991). In the GATS/WTO, public services are exempt by virtue of either not being covered by the GATS general obligations (i.e. MFN or transparency rules), or not being chosen by individual Member countries for specific NT or market access obligations.
This article argues that several factors limit the value of sectoral exemptions as a strategy for insulating public health care from international trade agreements. Part One demonstrates that, even if health care systems are viewed from a relatively static perspective and legal opinion favours a fairly broad interpretation of exemption clauses, the uncertainties of legal interpretation and the vagaries of international dispute resolution serve to transform many purely domestic policy issues into essentially contestable “legal” ones under international agreements. Part Two shows that the dynamics of international health care reform are likely to increase the exposure of domestic health care systems to international trade rules over time. That is because current and projected trends in healthcare policy are in the direction of market-based reforms, and the applicability of sectoral exemptions generally depends on the degree to which supply of services is market-oriented. It follows from these considerations that the essential function of carve-outs is not to preserve the historical insulation of health care from international trade, but to structure the linkage between domestic and international systems in this sector in a way that makes the degree of exposure to international rules a function of domestic policy. The nature of this linkage can be expected to vary between countries and between different welfare state traditions, but it is largely unavoidable.
It therefore behooves health policy communities and policy makers in all advanced welfare states to not rely upon optimistic interpretations of trade treaty exemption clauses as a basis for taking a “business as usual” approach to health policy. It is necessary instead to recognise the need for systematic integration of health and trade policies, and for the strategic management of trade-related risks. The aspect of domestic policy autonomy most clearly affected by international trade agreements is policy reversibility – a very important criterion for health care reform. The areas most vulnerable to being removed from domestic control by a trade action include the extension of public financing to new services that are currently provided by private firms (e.g. home care, drug plans, and telehealth in many countries), and the restoration of public financing or provision in areas that have been turned over to private markets as part of health care reform.
I. Legal Interpretations of the GATS and the NAFTA
A. The Structure of Trade Agreements: Two Basic Models
As several studies have pointed out, measures included in the FTA and NAFTA to enable parties to protect or exempt public services were necessary in order to accommodate domestic interests worried about the effects of “free trade” on the welfare state. This was of course primarily a concern in Canada, where it turned out to be a central issue in the 1988 federal election (Doern and Tomlin 1991; Hart 2002). The creation of express exemptions for social services was also necessitated by the general architecture of the NAFTA: it is primarily a “top down” agreement, which binds parties unless they are expressly exempted. (Sanger and Sinclair 2004, 30-33). While NAFTA critics in Canada have sometimes warned that the Agreement could force Canada to open its health care “market”, and the federal government has repeatedly averred that health care is “protected” , academic commentators have pointed out that both positions are overly simplistic. While the NAFTA explicitly protected those healthcare measures in effect in Canadian provinces at the time of the Agreement’s coming into force in 1994, it did nothing to protect future reforms that might be needed to modernise medicare, such as the expansion of pharmacare and home care. While there is little or nothing in the NAFTA to cause US-style “privatisation” (i.e. the replacement of non-profit hospitals with for-profit firms, or the extension of private insurance into areas now subject to a public insurance monopoly) there is much in the Agreement to make the reversal of such privatisation more costly and difficult. The NAFTA has been accused of failing to recognise the mixed public/private nature of the health care system, resulting in uncertainty as to what services are protected from application of some of the NAFTA's key provisions. The Chapter 11 expropriation provision and somewhat veiled dispute settlement process also presented particular concerns for some commentators, as the reservations provide no direct protection from litigation by private interests (Epps and Flood 2002; Van Duzer 2004a).
The early years of the WTO have also seen no shortage of legal opinions about the scope of the GATS, ranging from the view that any departure from statutory monopoly in health care will necessarily cause the GATS to apply, to the view that all GATS clauses will be interpreted so as to give priority to the regulatory autonomy of WTO members. Since 2002, legal academics have developed comprehensive legal strategies for assessing the scope of the Article I: 3 (b) and (c) governmental authority exclusion; the Article XVI and XVII market access and national treatment obligations flowing from specific commitments; and the Article VI domestic regulation provision, with regard to their implications for public health care. As a result, we now have a detailed analysis of these GATS provisions comparable to that already accorded to the NAFTA Chapter 11 and Social Service Reservation; note, however, that the different structure of the GATS limits the relevance of the NAFTA as a model. The GATS/WTO is primarily a “bottom up” agreement, at least when it comes to the most serious obligations. That is, the GATS only imposes full national treatment obligations with respect to particular sectors if a Member has explicitly chosen to be bound in those sectors.
Early indications are that the GATS/WTO model, with an exemption from the most general obligations for public services and a more flexible “bottom up” or “positive list” approach is the preferred model for other regional trade areas. This is even true of the latest drafts of the Free Trade Area of the Americas (FTAA), which one might expect to be more influenced by the NAFTA model. The greater flexibility afforded by the scheduling methodology is seen as more desirable by Member states, although it is far from free of difficulty or hazard, as Canada’s experience with split-run magazines illustrates.
1. Canadian Health Care Under the NAFTA Social Service Reservations
The view that the scope of NAFTA reservations in relation to Health Services is sufficient to protect publicly funded healthcare in Canada from any NAFTA challenge is a reasonable interpretation from a static perspective, based upon the accepted definitions of public and private health services at the time of NAFTA’s inception (VanDuzer 2004a). Canada’s Annex I Reservation states that all provincial government measures that were in force as of January 1, 1994 are outside the NAFTA rules relating to national treatment, MFN, and some other disciplines relating to local presence requirements for cross-border services and nationality requirements for senior managers. Thus the most important traditional features of Canadian health law – “medically necessary” hospital services and “medically required” physician services – are covered, but there is a current shift toward drug therapy, home care and new technologies such as telehealth that largely fall outside this traditional definition of medicare and may not therefore be covered by the Canada Health Act (CHA), forcing an increased reliance upon private financing. It is an anomalous feature of the Canadian system that drug treatment is only considered “medically necessary” under the CHA when provided in hospitals, and that coverage for drugs used outside of hospitals varies between provinces and generally involves considerable out-of-pocket expenses for most categories of patients. While a number of proposals have been put forward to provide universal drug coverage and/or home care, such new policies will not be covered by Annex I; and any future measures that exclude or otherwise discriminate against US and Mexican providers of services will be found to be contrary to the NAFTA, unless they are saved by the Annex II Social Service Reservation. There are also a growing variety of public/private arrangements (contracting out, licensing, public/private joint ventures, and so on) that will make it more difficult to draw a clear legal line between “public” and “private” in the future.
Under Annex II of the NAFTA, each Party reserved the right to adopt or maintain any measure relating to health services that may be characterised as being with respect to a “social service established or maintained for a public purpose”. The precise scope of this Social Service Reservation is the subject of much debate and speculation. The US Trade Representative in 1995 suggested that “where commercial services existed that sector no longer constituted a social service for a public purpose” (Appleton 1996, 96). Were this interpretation to apply, it would leave very little to be protected by reservation, since most health care services in Canada are supplied by private entities, including both (not-for-profit) hospitals and physicians. The Canadian government has claimed that, to the contrary, the broad ordinary meaning of “service established or maintained for a public purpose” reflects an intention to permit each Party to NAFTA to decide for itself whether it views a particular service as falling within the reservation.
This is a difficult issue to settle definitively, since the NAFTA leaves both the term “social service” and “public purpose” undefined, and the scope of the Annexes still have not been tested before a NAFTA dispute settlement tribunal. Epps and Flood have argued that the definition of “social service for a public purpose” is satisfied where the government funds the service “for the benefit of all those who require them on the basis that everyone ought to have access to such care”. They also find that the meaning of “public purpose” is “arguably wide enough to include services that the government wishes to fund for the public benefit” (2002, 778). Perhaps they are being a little too sanguine about the superiority of the Canadian interpretation, in view of the fact that the NAFTA’s objectives (i.e. trade liberalisation) are to be used to assist in interpreting the text. Nevertheless, the Canadian position does appear to be more reasonable than the American position, which would render the Annex II Reservation little more than superfluous. Legal academics generally agree that an objective test based on general criteria for what constitutes a public service is probably needed in addition to a statement of public purpose. (Appleton 1996: 96; Epps and Flood 2002; Van Duzer 2004a; Epps and Schneiderman 2005). Where full state funding is combined with extensive government control over delivery, then there is a very strong case for the application of the reservation. It is probable that full state funding alone, or in combination with a statement of public purpose, is sufficient even where governments permit competition and for-profit delivery in the interests of efficiency (Epps and Flood 2002, 777-780).
Nevertheless, it is already apparent that the Social Services Reservation does not protect measures related to for-profit privately funded services of physicians and other healthcare professionals; or privately-funded home care or nursing home services (VanDuzer 2004a). Allowing private insurance for services designated as “medically necessary” would further reduce the scope of this NAFTA reservation. Furthermore, regardless of the reservation’s scope, it does not provide protection from the Chapter Eleven (article 1110) expropriation provision – which does apply to any services currently being privately financed that could lead to foreign companies bringing compensation claims for “indirect expropriation” of their investments in violation of Canada’s national treatment obligations. How significant this exposure is depends upon the degree of market penetration by foreign private insurers from NAFTA partners.
2. The WTO/GATS and the Article I (3) “Governmental Authority” Exclusion Clause
A service “supplied in the exercise of governmental authority” is defined in Article I: 3 as any service, which is “supplied neither on a commercial basis, nor in competition with one or more service suppliers” (WTO 1994, 4). The Vienna Convention on the Law of Treaties codifies the most important customary rules of treaty interpretation. Article 31: 1 of the Convention states that a treaty shall be interpreted “in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its objective and purpose”. International lawyers generally give pride of place to the text and context of relevant treaty provisions, using the intentions of parties and the objects and purposes of treaties as supplemental means. (Sinclair 1984, 114-118; Marceau 1999) WTO dispute panels and appellate bodies have followed this hierarchy of principles, including the few panels that have dealt with the GATS to date. The same approach would be applied to the interpretation of the governmental authority clause. (VanDuzer 2004b, n.233) Other principles can be applied on a supplemental basis if circumstances or the treaty text call for it. Of these, the principle of in dubio mitius, or deference to the sovereignty of states, can be expected to be used in legal arguments over the scope of governmental authority: “[i]f the meaning of a term is ambiguous, that meaning is to be preferred which is less onerous to the party assuming an obligation, or which interferes less with the territorial and personal supremacy of a party, or involves less general restrictions upon the parties”. (WTO Appellate Body Report 1998, para 167, n 154)
Commentators generally assume that a pure government monopoly that does not charge for its service would meet this definition. Several writers critical of the GATS and worried about its potential impacts on public services have alleged that little else would be covered by the clause. According to Sinclair, these services “are defined so narrowly that the exclusion has very little practical value”. (Sinclair 2000, 5) Sanger has added that claims by WTO and Canadian officials that health services were absolutely protected by Article I: 3 were “at odds with a plain reading of the text and the advice the WTO Secretariat provides member nations”. (Sanger 2001, 113) Krajewski’s detailed legal analysis concluded that “[t]he narrow meaning of ‘governmental authority’ is caused by the dependence of the clause on the circumstances of supply and not on the nature of the service. The notion of competitiveness especially makes it difficult to exclude any service sectors per se” (Krajewski 2003, 354).
Although these lines of argument represent real possibilities for the future development of the law, they depend in large part on how three more specific issues are resolved. First, it is claimed that statements from WTO officials indicate an intended or expected narrow interpretation. (Pollock &amp;amp;amp;amp;amp;amp; Price 2000; Sanger 2001, 113) Second, statements by Members themselves are held to have weight, such as the statement by the EU and some associated states that the governmental authority exclusion is similar to Article 55 of the Treaty of Rome (Krajewski 2003, 363). Third, it has been argued that, like Article 55, Article I: 3 should be considered an exception or exemption clause, which, according to the European Court of Justice, is to be interpreted narrowly. (Krajewski 2003, 365-366). Statements from the WTO Secretariat and Member states about the scope of Article I: 3 may be relevant to its interpretation, but they are so varied and imprecise that their implications for the scope of government authority are by no means uniformly restrictive. For example, the infamous WTO documents referred to above may actually have been an attempt to make it clear that a carve-out does not exempt private health services, without necessarily implying that public provision would be affected. (Adlung 2001, 3) As both Johnson and VanDuzer have pointed out, the interpretive rule that exceptions should be narrowly interpreted is increasingly contested, and, in any case, should have no application to the interpretation of GATS Article I: 3 (b). “Exclusion” for the purpose of defining the scope of Members’ obligations toward one another is not the same as an “exception” (Johnson 2002, 18-20, 25-26; VanDuzer 2004b, n.256-61).
Accordingly, the potential scope of Article I:3 may be broader than previously thought. While the legal meaning of “competition” in services is unclear, it probably involves consumers being able to choose between “like” services offered by different suppliers on the basis of quality or price. Similarly, any finding of supply to be on a “commercial” basis would need to consider a range of criteria: whether a service is supplied on a for-profit basis; whether user fees are charged; whether any revenues earned in excess of cost are not devoted to fulfilment of a not-for–profit purpose; and the degree of government involvement and control over conditions of service delivery. Most of these criteria, when applied to core medical services that are publicly financed and supplied on the basis of need, would not necessarily indicate their classification as being supplied on a commercial basis. If more aspects of the public health system are found to lie outside of the Article I: 3 (b) (c) exclusion in the future, a likely cause will be the creation of health care markets that can trigger the application of the “competition” and “commercial” criteria (Crawford 2005).
ii. Legal Uncertainty, Differential Impacts, and Risk Analysis in Domestic Health Policy Regimes
If the only public services that are certain to be exempt from GATS obligations or to be protected by NAFTA Annex II-style reservations are those that are “publicly funded”, “wholly managed by government” and supplied for “free” at the point of consumption, then a lot may hinge upon the variety of ways that health care services are organised in different countries and the general direction of international health care reforms in welfare-state democracies. “The domestic neo-liberal agenda of privatisation therefore could place all that is privatized or partly privatized onto the international agenda – and this is no less true of privatized services in the developed states than of the IMF and World Bank-mandated privatisation schemes in developing countries” (Wiener 2005, 158). All real-world departures from the limiting case of “pure” public services entail some degree of risk that they are not protected by trade treaty “carve-outs”.
The more that some states involve elements of competition and commercial provision of health services, the more significant these risks are. Those countries which retain more elements of a traditional publicly-operated delivery system that reimburses hospitals and physicians on a fee-for-service basis (such as the original NHS system in the UK or the present system in Canada, where escalating costs and recent legal and political developments have also generated pressure for market-based reforms) are the most insulated from trade treaty obligations. Those countries that have adopted elements of an internal market (government-appointed purchasers to bargain and enter into contracts with competing public and/or private health service providers, on the basis of a “purchaser/provider split, as in the UK or New Zealand, while retaining heavy reliance on the single government payer); managed competition (which requires private insurers to compete for customer allegiance on the basis of cost and quality instead of risk avoidance, as in the Netherlands); or managed care (i.e. where an insurer/purchaser attempts to influence the cost, volume, and quality of health services supplied, which is consistent with internal market and managed competition reforms but can exist in an ad hoc form resulting in competition between insurers on the basis of risk avoidance, as in the US) are generally more likely to be subject to trade treaty obligations. (Flood 2000, 4-13).
Uncertainty in legal interpretation therefore needs to be viewed in the dynamic contexts of the near-universal pressures for market-based health care reform and the evolution of international trade dispute settlement. Although there is little sign as yet that trade agreements are seriously constraining domestic health policy making or driving international health care reform, there are at least two senses in which civil society and academic concerns are warranted. First, while it appears that trade liberalisation does not imply “privatisation” (in the sense of for-profit health care and competitive and commercial delivery), to a considerable extent the reverse may be true – specifically when market reforms cause health care measures to fall outside of sectoral exemptions for public services. Second, we can also see that much of the potential policy constraint that derives from trade rules comes from the reduced reversibility of market-based policy options, rather than from the operation of market-based instruments themselves.
A rough estimation of the potential exposure to trade treaty obligations of various market reform models is depicted in Figure 1. The accuracy of these preliminary assessments is perhaps less important than the methodology of undertaking assessments associated with health care reform. The most ubiquitous model is internal market reform, which can be applied piecemeal to specific services (e.g. physicians, hospitals, pharmaceuticals) and incrementally to any public health care system as a response to rising costs or inefficiencies. The principal worry here is that a very restrictive interpretation of exclusion clauses will cause even limited experiments with markets to fall outside of the GATS Article I:3 exclusion or the NAFTA Annex II Social Service Reservation. This risk is characterised as low to medium because legal analysis has challenged some of the assumptions underlying the more pessimistic predictions of trade critics, and because legitimacy concerns are arguably pushing WTO Dispute Settlement toward what G.R. Shell describes as a “Trade Stakeholders” model, in which the WTO dispute resolution process is open to all groups with a stake in the outcomes of trade decisions, at least in the sense of transparency and possibility of intervention through oral or written submissions (Shell 1995). It is more difficult to generalise about the degree of trade-related risk associated with managed care reform, since it may form a part of planned and integrated health care systems aimed at universal coverage. Nevertheless, since all managed care schemes refuse to simply reimburse hospitals and physicians for their services, they are likely to include more competitive and commercial elements at the point of delivery (Flood 2000, 8-9). Where managed care is combined with wide open competition between private for-profit insurers on the basis of risk avoidance as well as other factors, without government-mandated fee schedules and obligations for universal coverage, then the risk of both NAFTA and GATS obligations are very high. Managed competition has evolved in continental Europe, initially to give the conservative employment-based systems greater universality and then, more recently, to achieve some of the administrative efficiencies associated with single-payer systems on the Beveridge model. Reliance on competition between private insurers implies exposure to GATS obligations, but this risk is mitigated somewhat by the highly regulated and mandated nature of the competition, which could be described as being de facto no closer to an open market than fee-for service or internal market reform in Beveridge systems.
Figure 1 (a properly formatted table appears on p. 15 of the article
Health Care Reform and Associated Risk of Trade-Treaty Obligations
· UK Health Authorities buy hospital services from NHS Trust and physician services from GP Fundholders
· Contractors bidding for contracts from health authorities (e.g. as proposed in Canada by Senate Committee) in lieu of passively reimbursing hospitals and physicians for services
· Most applicable to “Beveridge” systems and social democracies seeking cost control
Potential Exposure to Trade Risks : · Low-medium risk of service falling outside GATS Article I:3 exclusion · Low risk of falling outside of NAFTA Annex II Social Service Reservation · Medium risk that NAFTA contractors will have opportunity to exercise Ch.11 investor rights
· Dutch private insurers must insure everyone and charge same flat fee for all of their clients. Competition on basis of flat fee and quality of services regulated and largely funded by government Sponsor
· Government mandates universal coverage and prevents competition between insurers on risk avoidance
· Most applicable to “Bismarck” systems that have contributory and employment-based social insurance
Potential Exposure to Trade Risks : Medium risk of service falling outside of GATS Article I:3 (competitive for-profit insurers are nonetheless highly regulated and mandated) · Medium risk of falling outside of NAFTA Annex II · High risk that countries with commitments in area of private health insurance will incur GATS NT obligations
· Insurer/purchaser (e.g. Health Maintenance Organisation in the US) contracts with hospitals, doctors, etc. to provide care to more patients in return for a discount (as compared to indemnity insurance)
· Physicians may be paid on a capitation basis (per enrolee)
· Not necessarily universal coverage; competition may be allowed on risk avoidance
· Most applicable to “liberal” systems
Potential Exposure to Trade Risks:· High risk of falling outside of Article I:3 and incurring general GATS MFN and transparency obligations · High risk of GATS NT obligations where commitments made in area of private insurance · High NAFTA risk, esp. in absence of government clearly mandating coverage and premiums
A. Trade-Related Uncertainty and the Reform of “Beveridge” Health Care systems: UK, New Zealand, Australia, Canada, Sweden
Welfare state reform has been on the public agenda of most industrial democracies at least since the mid-1980s, and for broadly similar reasons: the growth in the health care component of budgets has prompted calls for review in times of burgeoning public debt; concerns over inefficiency, reduced access and quality of services; and the political wearing down of the post-war Keynesian welfare state consensus and consequent rise of conservative or neo-liberal policy, as most boldly represented by the governments of Margaret Thatcher and Ronald Reagan (Marmor 1999, 265). Nevertheless, it is useful to group welfare states into two broad categories in terms of financing structure, since these different structures influence how health services are organised (and hence how they are reformed). The first is the group of systems financed out of general tax revenues named after the creator of the National Health Service (NHS) after the Second World War, Lord Beveridge. The second is a social insurance model called the Bismarck model after Otto von Bismarck, the architect of the late-nineteenth century German welfare state.
The Beveridge model, in its original form, was highly integrated, with the government acting as both the payer and the provider of health care services. This is unsurprising given the socialist vision of the Labour Government and the UK’s status as a unitary state governed according to parliamentary supremacy. Tax-based systems typically place a greater emphasis on primary care, prevention, home care, and pay doctors on a salaried (fee-for-service) or capitated (pre-paid care, per-capita, for a defined group of patients) basis. The greater administrative efficiency of an integrated public payment-and-delivery system (since governments do not expend resources in avoiding risk and risk-rating of premiums, and typically experience lower transaction costs) has arguably been responsible for the Beveridge countries historically absorbing two-to three percent less of GDP than the Bismarck systems (Fuller 1998). Canadian medicare, which was adopted in all provinces in the 1960s, deviated from the Beveridge model insofar as it relied on private sector delivery by physicians and not-for-profit hospitals. It also fails to cover many core health services when they are provided outside of hospitals.
In the late 1980s and early 1990s both the UK and New Zealand released proposals for “internal market reform” in response to what were perceived as inefficiencies caused by health authorities being both purchasers and providers of public hospital services. Both countries split the purchaser and provider roles of regional or area health authorities, which were no longer permitted to provide health services directly. Public hospitals are now managed by “NHS Trusts” in the UK and “Crown Health Enterprises” in New Zealand, which are meant to act like private firms and compete with each other and with private firms for supply contracts with the new Health Authorities. Exceptions to the purchaser/provider split came in the form of such managed care arrangements as the “GP Fundholders” in the UK (abolished in the New Labour reforms of 1997 and replaced by larger groups of doctors and nurses called “Primary Care trusts”) and “Independent Practice Associations” in New Zealand (or “Budget-holders”). New Zealand has also seen out-of-pocket expenditures increase, with a new array of user charges and patient copayments at the point of service (Donelan, Blendon, Schoen, Davis and Binns 1999, 214).
While these “internal market” reforms are criticised by advocates of managed competition because they still force citizens to rely upon government-appointed health authorities to purchase services on their behalf, and patients cannot switch their share of public funding to competing purchasers, this feature probably yields a significantly reduced likelihood that a particular health care service falls outside of the Article I:3 “governmental authority” clause. As discussed above, legal analysts have found some good grounds for believing that this clause will be given meaningful scope, since its “ordinary meaning” in the light of the “objective and purpose” of the GATS will mean balancing the goal of trade liberalisation against respect for domestic policy objectives and sovereignty. The mere splitting of purchaser/provider roles in order to facilitate competitive bidding for contracts from Health Authorities should only result in a low risk to trigger the GATS obligations, because being “supplied neither on a commercial basis, nor in competition with one or more service suppliers” can be interpreted as an over-all characterization of the service as it is provided to the patients. (A narrower interpretation that seizes upon the existence of any competitive or commercial aspects to trigger an obligation would likely render the clause largely superfluous in several countries, as where private physicians and hospitals are publicly funded.) In particular cases, however, much may turn on whether the “internal market” really is internal; the more that for-profit contractors are competing on the basis of price and quality and the less that Government is directly regulating the provision of services, the stronger the likelihood that the case will be characterised as being supplied “in competition”.
Another prominent feature of British NHS reforms, however, has been the reliance upon “Private Finance Initiatives” (i.e. Public-Private Partnerships or P3s) to finance and run hospitals. 94 percent of new hospitals built since 1997 have relied on PFIs, which have proven to be highly controversial. Whether improvements in private investment and patient choice have actually outweighed costs in terms of diverted resources from integrated front-line care is hotly debated; it is claimed by many that price competition in a balkanised NHS has actually harmed quality of service (Pollock, Shaoul and Vickers, 2002). These reforms could also be controversial for their trade treaty implications: the more that revenues earned by hospitals are devoted to for-profit purposes, the more that user fees are charged, and the more that competition between hospitals or PFIs is directed toward the ultimate consumers of services (i.e. patients) and not just to Health Authorities, then the more likely it is that NHS reforms will be characterised as “commercial” services and no longer as “in the exercise of governmental authority”.
The issue of reversibility of privatised health services illustrates what is at stake for democratic policy-making in this context. A strength of parliamentary systems is that the fusion of executive and legislature not only can make innovation easier, but it is difficult to bind future parliaments: mistakes can be undone. It is reasonable to hypothesise that Beveridge-style systems that experiment with for–profit delivery of health care essential health care services are strong candidates for policy reversal in the future. Although such innovations are frequently attractive in the short-run, since private investors in hospitals and clinics bring additional supply to market more quickly and at a lower direct cost to taxpayers, the steady flow of public funds to private shareholders and the mixed results in terms of cost efficiency and value for money is likely to cause public enthusiasm to wane. Sweden is a case in point: it has backed out of its experiment with for-profit hospitals, begun in 2002 and suspended in 2006 after an election largely fought on the issue (Sandborn 2006).
In Canada, analysts on the political right blame public monopoly of financing and lack of competition in delivery for the fact that Canada’s medical system has cost more than other Beveridge systems (e.g. Premiers Advisory Council on Medicare 2002; Esmail, Walker, and Yeudall 2004). Analysts on the left are more likely to blame the high costs associated with fee-for-service and the public purchaser/private provider split (e.g. Gibson and Fuller 2006; Sanger and Sinclair 2004). Each point of view has very different implications for reform that are difficult to fully evaluate. It is nevertheless quite easy to compare them in terms of the risks they pose of incurring trade treaty obligations. A pure public service for health care, as in the original NHS Beveridge model, is certainly outside of both GATS general obligations and NAFTA obligations. Canada’s existing system of single public payer and private delivery of hospital and physician services is almost certainly also free of such trade treaty obligations but, as market-based reforms are introduced and the nature of medically necessary services changes, the risk of exposure to the GATS and the NAFTA will rise.
It also appears likely that the growth of private health insurance for essential medical services, such as is allowed for in the recent Supreme Court of Canada decision striking down the Quebec Government’s ban on private insurance where waiting lists threaten “life, liberty and security of the person” could attract not only MFN obligations under the GATS, but also National Treatment and Chapter 11 obligations under the NAFTA (Epps and Schneiderman 2005). In addition, Canada’s earlier commitments at the WTO to liberalise financial services included private health insurance, because the government had mistakenly assumed that these would not include medically necessary services in the foreseeable future (Crawford 2006). The Government of Quebec has responded with a strategy of minimal compliance that will result in little or no market penetration by foreign private insurers, rendering the issue moot for the time being. The Government of Alberta had signalled its interest in experimenting with private insurance, although it later backed down from this position. The adoption in a Canadian province of an Australian-style two-tier “medicare–plus optional private insurance” would probably attract all of the aforementioned GATS and NAFTA obligations, at least for the private portion of the service.
B. Trade-Related Uncertainty and the Reform of Insurance-based “Bismarck” Health Care Systems: Netherlands, France, Germany
European countries continue to retain remarkably distinctive national health care systems, notwithstanding pressures to harmonise regulations and loosen restrictions on the mobility of health care providers, and the adoption of common external trade policies. Of the 102 signatories to the 1997 Financial Services Agreement, 80 made commitments in the health insurance sub-sector, including the Member states of the EU. The potential impacts of these commitments can vary significantly between Members, however. While the effect is to create a more level global playing field for private service providers, the impacts for citizen/consumers are likely to prove more uneven.
Given the importance attached to public funding as a criterion for finding health care services to be covered by the GATS Article I:3 and the NAFTA Annex II Social Service Reservation, what can be said about insurance-based or “Bismarck” systems that have been financed by a mixture of public and private sources? For example, the Netherlands prior to its 1987 “managed competition” reform proposals had a system that was financed 14 percent from general taxation, 60 percent from compulsory health insurance premiums, 16 percent from voluntary health insurance premiums, and 11 percent from patient user fees (Flood 2000, 62). This fragmentation of funding between different insurers led to higher costs than the Beveridge “single-payer” model, owing to the inability of the government to control cost-shifting and co-ordination problems. Nevertheless, the Dutch government did not try to overturn the entire Bismarck model; instead, a system of private insurers and non-profit Sickness Funds contracting with private health care providers was brought under tighter government regulation in order to ensure that competition occurs between insurers on the basis of cost and quality rather than risk avoidance (Flood 2002, 6-7). As a system financed progressively with little direct relation between individual contributions and entitlements (the Government Sponsor pools the funds and regulates the competition between insurer/purchasers for enrolees), government-mandated insurance is able to insure nearly 100 percent of the Dutch population.
Is the health care system of the Netherlands subject to a GATS Article I:3 obligation to meet WTO transparency standards and to not discriminate between foreign insurers and service providers? The rebuttal to Markus Krajewski’s analysis of the GATS “government authority” clause, which finds that it is narrow because it is “defined in terms of the economic basis and circumstances of the supply and not in respect to the public interest of the service”, is that public funding is the pre-eminent criterion for triggering an Article I:3 obligation. In the Netherlands, the post-1988 managed competition reforms do not fundamentally alter the plural/private sources of funding. Although the Government’s specification of insurance plans and regulation of prices attenuates the “commercial” and “competitive” aspects of the system, it is difficult to see how the system as a whole could fall within the GATS exclusion. In particular, the preservation of consumer/patient choice of insurer would appear to meet even the narrowest definition of a service supplied “in competition” (Krajewski 2003, 352-353).
The French Public Health Insurance System (PHIS) evolved, like the German and Dutch systems, from a variegated collection of funds originally based on professional activity with disparate reimbursement rates, into a more standardised system characterised by uniform rates and universal entitlement for all legal residents. Most funds are still private entities, however, jointly managed by employers’ and union federations under State supervision (Couffinhal 2001). One reason that waiting lists are not the problem in France that they are elsewhere is that there is no physician gatekeeping in France, but there is a tariff (“conventioneee”) at the point of service, regardless of whether the service provider is “public” or “private”. Sometimes additional charges (“depassements”) are also levied. Since 100% reimbursement is not always provided, and there are almost always tariffs to pay, most French residents purchase a “top up” insurance policy from a variety of providers, including some foreign private insurance companies. Again, the elements of patient choice and payment to competing (and sometimes for-profit) firms at the point of service would make it difficult to place French health care entirely within the “governmental authority” exemption, although it is possible that trade treaty obligations may only apply to the “top up” portion of the service.
In Germany, residents earning below a stipulated income level are obligated to belong to the statutory health insurance system based on some 600 “Sickness Funds” (Krankenkassen), which are billed directly for most medically necessary services, although co-payment from patients is required for pharmaceuticals and dental work. Patients at higher income levels have the option of staying in the state plan at a higher premium or opting out in favour of private insurance. Here, the high proportion of public funding and government regulation coupled with a fairly clear separation from the private system would make at least the basic (obligatory) system a good candidate for coverage by the Article I:3 “governmental authority” exemption. Remember, however, that the EC’s commitments in private health insurance mean that, as in France, the private system is subject to National Treatment and Market Access obligations of the GATS, and not just the general MFN and transparency requirements.
iii. Conclusion: The Need to Incorporate Trade Law Considerations into Health Care Policy-Making and Risk Management
It is an illusion to think that a reasonably broad interpretation of the GATS Article I:3 “Governmental Authority” clause, the NAFTA Annex II Social Service Reservation, or any other trade treaty provisions will in itself insulate health care policy from constraint by international trade rules. The difficulty is that (1) estimating health care’s insulation from largely untested and uninterpreted parts of trade treaties is a probabilistic, not a categorical exercise, reflecting a risk and uncertainty that is simply too great to be ignored by domestic decision-makers; and (2) the general direction of health care reform in most advanced industrial democracies is raising the risk of triggering trade treaty obligations, regardless of other trends such as those in dispute settlement. Since trade-related risks are unavoidable as well as important, it is time to acknowledge the mutual constitution of trade and health law, and to more fully coordinate and integrate trade and social policy.
As the case of Sweden illustrates, significant moves toward competitive for-profit delivery of health services are likely candidates for policy reversal in the future. Policy-makers will need to assess: what are the costs, in terms of trade concessions to WTO Members and NAFTA compensation to firms, of future re-socialisation of medicine? What value is to be placed upon policy reversibility as a criterion of public policy and policy autonomy? What are the costs in terms of trade treaty obligations of expanding the public sector into new areas that currently have foreign private sector suppliers? Even on an optimistic view of the scope of public sector exemptions and the future evolution of WTO law, it is plain that there is no going back to the watertight compartments of the past.
. Thus, the GATS covers not just cross-border trade, such as (in the case of health services) provision of diagnosis or treatment planning services in country A by suppliers in Country B via telemedicine (mode 1), but also consumption abroad, e.g., movement of patients from Country A to Country B for treatment (mode 2); commercial presence in a country, e.g., establishment of, or investment in, hospitals in a country by foreigners (mode 3); and temporary presence of natural persons in a client/host country, e.g., service provision in Country A by health professionals from Country B (mode 4).
. See, for example, Grady and McMillan 1999, 93; McBride 2001, 109-113.
. See e.g. Sinclair 2000, 54-55, 81. The WTO panel found that Canadian and US magazines were “like goods” and that the Canadian measure to subject foreign split-run magazines to a tax on advertising designed to protect Canada’s culturally distinctive magazine industry was ruled to violate national treatment under NAFTA. Canadian trade officials had argued that advertising was a “service” and that Canada had scrupulously avoided commitments in this area in the GATS negotiations. The panel reasoned that goods and services were “overlapping” and that the measure had to comply with both the GATT and the GATS.
. EU – Bananas, Canada – Periodicals, Canada – Autopact, and Mexico – Telecommunications. Disputes, chronologically available on the WTO website.
. In one study conducted at McMaster University, a team of 20 researchers reviewed 788 medical articles on hospital care, eventually honing in on the 8 highest quality and most relevant studies – which included a total of 350,000 patients and a median of 324 hospitals in each study. They contacted the original authors to verify the findings, then used advanced statistical techniques to combine the 8 studies. Of the 8 studies, only one showed that for-profits had lower costs. See Devereaux, Schunemann, et al. 2004, 1817-1824.
. Chaoulli v. Quebec (Attorney-General), 2005 SCC 35.
. On April 20, 2006, the Government of Alberta announced that it was not proceeding with plans to expand private health insurance and to let doctors work in both public and private systems, due to public opposition and concerns about non-compliance with the Canada Health Act, which could lead to the federal government withholding transfer funds for health care.
Adlung, Rudolph. 2001. ‘Health Care Systems and the WTO: No Grounds for Panic’. 27 March. Available at http://www.wto.org/english/tratop_e/serv_e/comments_lancet_e.doc, accessed 14 May 2001.
Appleton, Barry. 1996. ‘International Agreements and national Health Plans: NAFTA’. In Market Limits in Health Reform: Public Success, Private Failure, eds., Daniel Drache and Terry Sullivan. London: Routledge.
Chaoulli v. Québec (Attorney General), 2005 SCC 35.
Couffinhal, Agnes. 2001. ‘The French Health Care System: A Brief Overview’. IRDES October 2001, 2. Available at http://www.medicalnewstoday.com/medicalnews.
Crawford, Mark. 2005. ‘Truth or Consequences: The Law and Politics of the GATS Health Care Debate’ Canadian Foreign Policy 12 (2): 97-133.
_______. 2006.’Interactions: Trade Policy and Healthcare Reform After Chaoulli v. Quebec’. Healthcare Policy 1 (2): 90-102.
Devereaux, P.J., Holger J. Schunemann, et al. 2004. ‘Payments for Care at Private and For-Profit and Private Not-for-Profit Hospitals: A Review and Meta–Analysis’. Canadian Medical Association Journal 170 (12): 1817-1824.
Donelan, Karen, Robert J. Blendon, Cathy Schoen, Karen Davis, and Katherine Binns. 1999. ‘The Cost of Health System Change: Public Discontent in Five Nations’. Health Affairs 18 (3): 206-216.
Doern, G. Bruce, and Brian W. Tomlin. 1991. Faith and Fear: The Free Trade Story. Toronto, ON: Stoddart Publishing.
Epps, Tracey, and Colleen M. Flood. 2002. ‘Have We Traded Away the Opportunity for Innovative Health Care Reform? The Implications of the NAFTA for Medicare’. McGill Law Journal 47 (4): 747-790.
Epps, Tracey, and David Schneiderman. 2005. ‘Opening Medicare to Our Neighbours or Closing the Door on a Public System? International Trade Law Implications of Chaoulli v. Quebec’. In Access to Care, Access to Justice: The Legal Debate Over Private Health Insurance, eds., Colleen M. Flood, Kent Roach, and Lorne Sossin. Toronto, ON: University of Toronto Press.
Esmail, Nadeem, Michael Walker, and Sabrina Yeudall. 2004. How Good is Canadian Health Care? Vancouver, BC: The Fraser Institute.
Flood, Colleen M. 2000. International Health Care Reform: A Legal, Economic and Political Analysis. New York: Routledge.
Fuller, Colleen. 1998. ‘Public Health Insurance: Imagine Something Better’. The Parkland Post 2 (1): 1.
Gibson, Diana, and Colleen Fuller. 2006. The Bottom Line: The Truth Behind Private Health Insurance in Canada. Edmonton, AB: NeWest Press and the Parkland Institute.
Grady, Patrick, and Kathleen MacMillan. 1999. ‘Canada in the Global Economy: An Overview’. Canadian Business Economics 7(3): 1-10.
Hart, Michael. 2002. A Trading Nation: Canadian Trade Policy from Colonialism to Globalization. Vancouver, BC: UBC Press.
Johnson, Jon R. 2002. ‘How Will International Trade Agreements Affect Canadian Health Care?’ Discussion Paper No.22. Ottawa, ON: Commission on the Future of Health Care in Canada.
Krajewski, Markus. 2003. ‘Legal Services and Trade Liberalization: Mapping the Legal Framework’. Journal of International Economic Law 6 (2): 341-367.
Kreklewich, Robert. 1994. ‘Consolidating the Neoliberal State: Free Trade as a Conditioning Framework’. Studies in Political Economy 43 (1): 33-61.
Marceau, Gabrielle 1999. ‘A Call for Coherence in International Law – Praises for the Prohibition Against “Clinical Isolation” in WTO Dispute Settlement’. Journal of World Trade Law 33 (5): 87-152.
Marmor, Ted. 1999. ‘The Rage for Reform: Sense and Nonsense in Health Policy’. In Market Limits in Health Reform: Public Success, Private Failure eds., Daniel Drache and Terry Sullivan. London: Routledge.
McBride, Stephen. 2001. Paradigm Shift: Globalization and the Canadian Welfare State. Halifax, NS: Fernwood Publishing.
Pollock, Allyson M., Jean Schaoul, and Neil Vickers. 2002. ‘Private Finance and “Value for Money” in NHS Hospitals: A Policy in Search of a Rationale?’ British Medical Journal 324 (7347): 1205-1209.
Pollock, Allyson, and David Price. 2000. ‘Rewriting the Regulations: How the WTO Could Accelerate Privatization in Health Systems’. The Lancet 356 (9246): 1995-2000.
Premier’s Advisory Council on Health for Alberta. (Rt. Hon. Don Mazankowski, Chair). (December 2001). Report of the Premier’s Advisory Council on Health. A Framework for Reform. Edmonton, AB: Premier’s Advisory Council on Health for Alberta. Available at http://www.gov.ab.ca/home/health_first/documents/PACH_report.pdf.
Sandborn, Thomas. 2006. ‘Europe’s Health Reforms: Hard Lessons’. The Tyee.ca. Available at http://thetyee.ca/Series/2006/03/08/EuropeHealthReforms/.
Sanger, M. 2001. Reckless Abandon: Canada, the GATS and the Future of Health Care. Ottawa: Canadian Centre for Policy Alternatives.
Sanger, Matthew and Scott Sinclair (2004). Putting Health First: Canadian Health Care Reform in a Globalizing World. Ottawa: Canadian Centre for Policy Alternatives.
Shell, G. Richard . 1995. ‘Trade Legalism and International Relations Theory: An Analysis of the World Trade Organization’. Duke Law Journal 44 (5): 829-927.
Sinclair, Ian. 1984. The Vienna Convention on the Law of Treaties. 2nd ed. Manchester: Manchester University Press.
Sinclair, Scott. 2000. GATS: How the World Trade Organization’s New ‘Services’ Negotiations Threaten Democracy. Ottawa: Canadian Centre for Policy Alternatives.
VanDuzer J. Anthony. 2004a. ‘The Canadian Preoccupation with NAFTA’s Impact on Health Services: A Serious Issue, A Non-Issue or Something In Between?’ In The First Decade of the NAFTA: The Future of Free Trade in North America , ed., Kevin C. Kennedy, New York, NY: Transnational Publishers.
_____________. 2004b. Health, Education and Social Services in Canada: The Impact of the GATS. Trade Policy Research 2004. Retrieved September 20, 2005. http://www.dfait-maeci.gc.ca/eet/research/TPR-2004/TPR-2004-10-en.asp#*d
Wiener Jarrod. 2005. ‘GATS and The Politics of “Trade in Services”’. In The Politics of International Trade in the 21st Century: Actors, Issues and Regional Dynamics, eds., Dominic Kelly and Wyn Grant, London: Palgrave Press.
WTO. 1998. Appellate Body Report: EC-Measures Concerning Meat and Meat Products (Hormones), WT/DS26/AB/R, adopted 13 February.
____ . 2001. GATS – Fact and Fiction. Geneva: World Trade Organization.