POLICY AND PRATICE
Global publicprivate partnerships: part II what are the health issues for global governance?
Partenariats mondiaux public-privé : partie II enjeux dune gestion mondiale
Fórmulas de asociación mundiales entre los sectores público y privado: parte II facetas de la gestión mundial
K. BuseI; G. WaltII
IDivision of International Health, Department of Epidemiology and Public Health, Yale University, PO Box 208034, New Haven, CT, 06520-8034, USA
IIHealth Policy Unit, Department of Public Health Policy, London School of Hygiene and Tropical Medicine, Keppel Street, London, England
ABSTRACT
This is the second of a two-part review of global publicprivate partnerships (GPPPs) for health development. Part I was published in the April issue of the Bulletin (Vol. 78, No.4). The recent emergence of GPPPs is rapidly reconfiguring the international health landscape. While most multilateral and bilateral agencies are currently grappling with how to proceed, there is little information in the public domain concerning how individual partnerships work and to date very little consideration of the many implications of this trend. This paper differentiates between product-based, product development-based and issues/ systems-based GPPPs and describes a number of examples of each type in the health sector. The benefits of these initiatives, not least the major resources which they harness for specific health problems, are identified. The final section of the paper explores the implications and dilemmas posed by GPPPs. It discusses whether or not shared goals can transcend conflicting values and mandates and how governance of partnership arrangements may transform and undermine certain attributes of multilateral organizations. The paper concludes that the current climate of goodwill between public and private sectors offers an opportunity that should not be missed: it can be used not only to foster new partnership but to ensure that partnership is truly in the interests of international public health.
Keywords: world health, trends; intersectoral cooperation, history; public sector; private sector; United Nations; developing countries
RÉSUMÉ
Cet article concerne la deuxième partie dune étude sur les partenariats mondiaux public-privé (PMPP) en faveur du développement sanitaire. La partie I a été publiée dans le numéro davril du Bulletin (Vol. 78, no 4). Lémergence récente de partenariats mondiaux public-privé est en train de remodeler rapidement le paysage sanitaire international. Si la plupart des organismes daide bilatérale et multilatérale sefforcent actuellement de déterminer la conduite à tenir, on ne sait généralement pas grand chose de la façon dont fonctionnent les partenariats et jusquici, on ne sest guère préoccupé des implications multiples dune tendance nouvelle. Les auteurs de larticle font une distinction entre les PMPP qui sont axés sur des produits, ceux qui sont axés sur le développement de produits et ceux qui concernent des problèmes/systèmes en citant des exemples de chacun de ces types de partenariats dans le secteur de la santé. Ils exposent les avantages de ces initiatives, dont le moindre nest pas la mobilisation de ressources considérables contre des problèmes de santé précis. Dans la dernière partie de larticle, les auteurs explorent les implications et les dilemmes posés par les partenariats public-privé en se demandant si des buts communs permettent de transcender des valeurs et des responsabilités contradictoires et si la gestion de ces partenariats ne risque pas de transformer et débranler certaines caractéristiques des organisations multilatérales. Ils concluent que le climat de bonne volonté qui règne actuellement entre les secteurs public et privé doit être mis à profit pour favoriser la multiplication des partenariats mais aussi pour obtenir que ces derniers soient réellement dans lintérêt de laction internationale de santé publique.
RESUMEN
Esta es la segunda parte de un análisis de las fórmulas de asociación mundiales entre los sectores público y privado (FAMPP) para el desarrollo sanitario. La primera parte apareció en el número de abril del Boletín (vol. 78, No 4). El reciente surgimiento de FAMPP está reconfigurando rápidamente el panorama internacional. Aunque la mayoría de los organismos multilaterales y bilaterales están luchando para hallar la manera de seguir adelante, apenas hay información de dominio público sobre el funcionamiento de fórmulas de asociación concretas, y hasta la fecha no se han analizado lo suficiente las numerosas repercusiones de esa tendencia. En este artículo se diferencian las FAMPP basadas en productos, las basadas en el desarrollo de productos y las basadas en problemas/sistemas, y se presentan varios ejemplos de cada uno de esos tipos en el sector sanitario. Se mencionan los beneficios conseguidos con esas iniciativas, uno de los cuales, y no el menor, son los abundantes recursos que manejan para abordar problemas de salud concretos. En la sección final del artículo se analizan las repercusiones de las FAMPP y los dilemas que plantean. Se examina si las metas compartidas permiten o no superar los conflictos de valores y mandatos y se analiza de qué manera la gestión de los arreglos de asociación puede transformar y minar algunas de las características de las organizaciones multilaterales. Se llega a la conclusión de que el actual clima de buena voluntad reinante entre los sectores público y privado brinda una oportunidad que no debe desperdiciarse, una oportunidad que puede aprovecharse no sólo para impulsar nuevas formulas de asociación, sino también para asegurar que esa colaboración redunde efectivamente en interés de la salud pública internacional.
Introduction
In Part I of this article, which appeared in the last issue of the Bulletin, we suggested that there have been a number of initiatives in which the corporate and public sectors sought collaboration in international public health through partnerships. That paper reviewed the concept of partnership and defined global publicprivate partnerships (GPPPs) as collaborative relationships which transcend national boundaries. Each partnership brings together at least three parties, among them a corporation (and/or industry association) and an intergovernmental organization, to achieve a shared health-creating goal on the basis of a mutually agreed division of labour. The paper described the context within which these partnerships are emerging, focusing particularly on changes confronting the United Nations and the corporate community during the 1990s.
While these partnerships are bringing major resources into international public health and have the potential to benefit large populations, they are also blurring traditional distinctions between public and private sector responsibilities and aims. The use of GPPPs in public health also raises a number of important questions of conflicts of interest, and implications for governance. This paper opens the debate on these issues, starting with a conceptual framework for understanding the different forms of global publicprivate partnership in the health sector, illustrated by a number of examples. It ends by exploring the implications of GPPPs for the 21st century, looking at issues of governance and equity.
What forms have GPPPs for health development taken?
GPPP categories
There are several ways to conceptualize and categorize partnerships. One is in terms of constituent membership, for example, donorrecipient or publicprivate. However, this is too broad to be very helpful for understanding GPPPs. Another categorization has been proposed by Mitchell-Weaver and Manning (1), who reason that as partnerships are primarily a set of institutional relations, they should be categorized by their organizational form. They differentiate between three institutional models on the basis of the degree to which private interests participate in the strategic-level decision-making in the public interest. The elite committee model (sometimes called a board or conference) is characterized by negotiation among relatively equal partners so as to arrive at decisions by consensus. The committee does not implement decisions, rather the individual members influence the behaviour of their respective organizations to achieve partnership goals and/or influence public policy through network associations. In the health sector, the Global Business Council on HIV/AIDS, which involves the leaders of 15 major companies, is one example of such a model (2).
The second institutional form is the NGO model, involving nongovernmental organizations. Mitchell-Weaver and Manning suggest that the relationship between parties is essentially one of delegation. The public side provides organizational, material or financial resources to enable a private partner to carry out the public programme. The NGO model links the public with the private sector through resource transfers and is exemplified by the Sexually Transmitted Diseases Diagnostics Initiative (3).
The third is the quasi-public authority model in which a hybrid organization with both public and private characteristics is created by public sector institutions. Acting in the public interest, this type of organization provides goods and services or enables the private sector to enter a market. In effect, the quasi-public authority model creates favourable conditions for private enterprise to provide public services or goods. The Medicines for Malaria Venture (MMV) (4) and the International AIDS Vaccine Initiative (IAVI) (5) could both be said to be examples of this organizational form at the global level. Although this characterization is promising, it fails to provide a model in which the private sector is a dominant partner. Consequently, it would be difficult to situate a number of GPPPs, including most of the drug donation programmes, within this framework.
Another approach to categorization would be to base it on the nature of activity undertaken by the partnerships. One might, for example, differentiate among those partnerships which focus on consultation between public and private actors (e.g., the WHO Working Group with Pharmaceutical Industry), and those which involve concertation of policy between public and private actors (e.g., collaboration on standard setting), and those with a mainly operational function (e.g., engaged together in research and development or a drug donation programme). These categories are, however, far too broad and reveal little about how partnerships are governed or function.
Kickbusch & Quick (6) have categorized global health partnerships as based on the following: (1) existing products (e.g., deworming drugs for children); (2) product development (e.g., designing a refrigerator for use in developing countries); (3) services; (4) systems and settings (e.g., healthy cities); (5) issues (e.g., polio eradication); (6) health messages (WHO/ UNESCO global malaria strategy); and (7) knowledge exchange (workplace health promotion). While the first two categories are distinct, the lack of specificity of the remaining categories results in some ambiguity. Although we feel that the approach to categorization based on institutional form (1) is worth further development, in this paper we use a goal-oriented, three-category classification of GPPPs: product-based, product development-based and issues/systems-based.
Examples of GPPPs
Some examples of these three types of partnership are provided in Tables 1, 2 and 3. The examples, mainly in the area of drugs and vaccines for communicable diseases, were chosen primarily on the basis of the availability of information, and include only a proportion of the better-known GPPPs currently operational. A number of GPPPs which deal with noncommunicable diseases, such as the WHO Partnership on Tobacco Dependence (7), and those which deal with the broader socioeconomic determinants of health (e.g. UNDPs PublicPrivate Partnership for Urban Environment) (8) are also emerging.
Product-based partnerships (Table 1) consist primarily of drug donation programmes, although partnerships also exist for the bulk purchase of products for public sector programmes in low-income countries, for example, female condoms (9) or AIDS medication (10). Drug donation programmes are generally established after the discovery that an existing drug (for animals or humans) is found to be effective in the treatment of some condition for which there is limited effective demand, due to lack of willingness and ability to pay, as was seen with AmBisome for the treatment of leishmaniasis. These types of partnership are usually initiated by the private sector. Pharmaceutical companies seek partnership with the multilateral sector to lower the cost and increase the chance of ensuring the drug reaches those who need it but cannot afford it. While private sector companies may seek short-term objectives through GPPPs such as the establishment of political contacts at global and country levels, it would appear that the longer-term objective is to establish their reputations as ethically oriented concerns. This end objective is not guaranteed, as product donation partnerships have been subject to controversy over dumping, dependency-creation and sustainability (11).
Product-development partnerships (Table 2) differ from product-donation partnerships in a number of respects. First, they are not targeted at specific countries. Second, these partnerships are generally initiated by the public sector (12). Third, the product-development partnerships are not based on ineffective demand so much as on market failure. Most of these products are perceived by the public sector as worthy of societal investment, but the market fails to allocate resources to their discovery and development because industry perceives that the potential returns do not justify the opportunity cost of investment. For example, although research on an AIDS vaccine is considered an important public good, industry is uncertain whether expenditure on research on it will yield a successful vaccine. Moreover, even if a vaccine is discovered, the private sector cannot be sure that a large enough market will exist to justify its development and commercialization. The potential cost of liability and regulation is another grey area (13).
Product-development GPPPs usually require the public sector to assume a number of risks associated with product discovery, development and/or commercialization (i.e., providing a public subsidy), thereby offsetting the opportunity cost of industrial involvement. A notable feature of a number of the product-development GPPPs is retention of the intellectual property rights by the partnership organization so as to retain leverage over eventual product pricing. Corporations may engage in product-development partnerships to mobilize a subsidy for research, to obtain assistance in carrying out clinical trials, or to pursue their own longer-term interests. Fundamentally, there is the certainty of some financial return (even if modest). Companies may also seek proximity or involvement in standard-setting and regulatory processes. Finally, companies may be seeking to portray themselves in a favourable light to help secure entry into emerging drug markets.
The systems / issues-based partnerships (Table 3) are a more eclectic group. Some have arisen to overcome market failures, such as the Malaria Vaccine Initiative (14). Some systems GPPPs have been established to complement the efforts of governments, such as the Secure the Future partnership (15), and others to tap non-medical private resources for disease control, such as theWorld Alliance for Research and Control of Communicable Diseases (16). A number of high profile issues-based GPPPs have recently been launched which seek to harmonize or bring strategic consistency to the approaches of various actors to single diseases, as well as to raise their profile on the health policy agenda. The Roll Back Malaria Global Partnership (17) and the Stop TB Initiative (18) are examples.
Are global publicprivate partnerships essential for health?
Shared goals and principles: can public and private be reconciled?
Definitions of partnership suggest a foundation of shared goals underpinned by agreement on key principles. The partnerships described in this paper have clear and uncontroversial goals. Of central importance to the global health agenda are the questions of who determines these goals, the processes by which they are determined, and to what extent the goals of GPPPs come to dominate the global health agenda. One might consider a continuum of partners interests. At one end of the continuum are the interests of the UN: our main stock in trade ... is to promote values: the universal values of equality, tolerance, freedom and justice that are found in the UN Charter (19). At this end of the continuum one also finds WHOs public health values and concern about inequities in health (20). The principles reflected in company policies may be at the other end of the spectrum with a concern to maximize profits so as to increase shareholder value. (This is not to suggest that these institutions and sectors are monolithic but instead to make a broad generalization concerning different interests and values. Naturally there are exceptions and employees with widely varying values within both private and public sectors). At the centre of this continuum are the GPPPs where, it is hoped, the interests of both can be met.
One critic of publicprivate partnerships argues, however, that the private sector has several mechanisms for maximizing profits which may conflict with the goal of better health (21). Among these is the reduction of costs by paying low wages and reducing the size of the workforce, thereby making people poorer. Is it possible that these private sector goals will ultimately dominate as the UN and industry move closer towards jointly defining their goals through GPPPs?
Alternatively, is it possible to ensure that core public and private identities and values are preserved in partnerships which limit themselves to specific win-win situations? Lenton of the International AIDS Vaccine Initiative (IAVI) has argued that shared goals are more important than shared values (22). In the short term, and with specific goals, it seems likely that goals can transcend disparate values and bind unlikely bedfellows together in a marriage of convenience. Literature is beginning to emerge on lessons learned on the effectiveness of health GPPPs (4, 23, 24). These emphasize the importance of: (1) clearly specified, realistic and shared goals; (2) clearly delineated and agreed roles and responsibilities; (3) distinct benefits for all parties; (4) the perception of transparency; (5) active maintenance of the partnership; (6) equality of participation; (7) meeting agreed obligations, inter alia. However, over the longer term the question arises of whether the values of the weaker partner are co-opted by the more powerful one. The answer to this question depends on the choice of private partners, as well as on the nature of GPPP governance.
Because of the potential clash between partners over principles and values, WHO, the World Bank and UNICEF all note the need to exercise caution over the selection of their private sector partners (6, 25, 26). In practice, given the short-term financial incentives that sometimes motivate UN organizations to enter into partnerships with the private sector, it may be difficult to refuse corporate offers which do not comply with all internal guidelines. For example, UNDP is alleged to be flouting its fundraising guidelines in pursuance of its Global Sustainable Development Facility initiative (27). In relation to health partnerships, Hancock urges sober second thoughts regarding the suitability of the pharmaceutical industry as a partner for WHO, at least in terms of health promotion, because of perceived or actual conflict of interests (21). He suggests instead that WHOs partners should be selected from industries which stand to profit economically from better health (e.g., life and health insurance, leisure and recreation, tourism and travel) and those that produce health (e.g., agri-food, housing). Hancock argues that at the global level, partnerships should be developed not with individual companies, who may wish to use the partnership for competitive gain, but with industry associations. He calls for a set of ethical criteria to guide the selection of partners.
Many multilateral and bilateral organizations are now aware of the need to give greater attention to this issue. WHOs proposed guidelines on partnership with the commercial sector single out tobacco producers and arms manufactures as incompatible partners (28), whereas other sections of the UN call for creative partnership with the arms industry (29).
Governance: representation, accountability and competence
Governance can be defined as the process whereby an organization or society steers itself (30). Broadly speaking, governance consists of the systems of rules, norms, processes and institutions through which power and decision-making are exercised. Good governance is thought to have four components: (1) representative legitimacy; (2) accountability; (3) competence and appropriateness; and (4) respect for due process (31). How is governance exercised in the global public private health partnerships?
Representation
The area of legitimate representation in publicprivate partnerships raises both normative and operational issues. Normative issues determine whose interests should be represented in the partnership and whose should not. Most UN organizations derive some of their legitimacy from near-universal membership in their governing bodies. For example, the World Health Assembly is attended by representatives of 191 member states (although its state-centric bias raises its own problems of legitimacy), all of which have equal voting rights irrespective of size of financial contribution. In contrast, representation in GPPPs is both narrower and more eclectic. For example, no health GPPP can claim near universal membership of nation states (which would in any case make them unwieldy), but, more importantly, few partnerships include representation by low-income countries. Furthermore, not all GPPPs include WHO on their governing boards and technical committees, and in some cases it appears that the private sector representation is ad hoc and based on personal contacts.
The legitimacy of GPPPs will depend to a great extent on the expert committees that are established to advise them. Whereas the specialized agencies of the UN, such as WHO, rely on extensive networks of technical experts and have established means for selecting and operating expert groupsa, there are concerns that GPPP expert groups may be chosen from exclusive communities of expertise. They may also suffer from a lack of independence due to the sources of funding (23) and may have circumscribed powers (for example the Technical Advisory Group of the International Trachoma Initiative did not have the opportunity to advise on the choice of recipient countries (32). Although many analysts have drawn attention to the extent to which international agenda-setting and the formulation of policy are controlled by transnational policy elites (33), the implications of the increasing role of the private sector in such policy networks have barely been explored.
Accountability
Accountability is broadly concerned with being held responsible for ones actions. Both the public and private sectors have well-established mechanisms of accountability. In the private sector, management is accountable to the companys shareholders. In the public sector, administrative structures report to political structures that are ultimately accountable to the ruled through the contestability of political power. However, accountability within publicprivate partnerships may be less straightforward, partly because of the distance between the global partners and the beneficiaries, and the length of time needed for any impact to be felt. Bain (34) quotes Fox & Browns study of transnational NGO networks which suggests that downward accountability is often weak, and is particularly limited when geographical distances are great or international as well as local organizations are involved in a project. Given that accountability is dependent upon the clear specification of objectives, activities, roles and responsibilities, it will be more easily achieved in formal partnerships where these are spelled out. In contrast, partnerships whose goals and division of labour are vaguely defined will lack accountability. Moreover, actually holding a partner accountable presents difficult challenges. At the moment, systems of sanctions that can be applied to negligent partners do not appear to have been developed.
Two mechanisms of accountability are emerging among health GPPPs. In one model, the management and scientific groups report to the corporate sponsor directly. For example, in the Mectizan® Donation Programme, the Mectizan® Expert Committee reports to Merck & Co., its corporate sponsor, through bi-annual meetings, while the Secretariat reports monthly on financial expenditure (23). Similarly the Program for Appropriate Technology in Health (PATH), which provides the secretariat for the Bill and Melinda Gates Childrens Vaccine Program, reports directly to Bill and Melinda Gates, its sponsors (35). Hence, in this model, the GPPPs are accountable first to their donors, and only indirectly to the public sector organizations and beneficiaries.
In another model, the management group reports to a governing body, whose members report back to their respective organizations. Hence, the secretariat of the International Trachoma Initiative will report on a six-monthly basis to its sponsors through the Governing Council. These reports will be shared with the wider trachoma network, notably the WHO Alliance for Global Elimination of Trachoma (36).
Competence
Partnerships raise difficult questions about competence and appropriateness. As global responsibility for specific health issues is transferred from WHO programmes to special GPPPs, there is some danger that WHO will fail to maintain expert groups on these issues as it tries to avoid duplicating the technical committees maintained under the aegis of the partnerships (whose membership is vetted by the corporate sponsors). Does this raise the spectre of the erosion of WHOs normative function? Where the private sector assumes a greater voice through partnership in WHO technical discussions, will global standards and norms increasingly reflect private interests, thereby jeopardizing their credibility? As an example, Muraskin (37) notes that the WHO Expert Committee on Biologicals (which established standards for vaccines) was criticized in the late 1980s on the level of standards set by the Organization. Many considered them unreasonably rigorous (thereby disadvantaging developing country industries) and too responsive to industry demands. Furthermore, will the proliferation of GPPPs exacerbate the fragmentation of international health organizations making it even more difficult to establish a coherent global health policy agenda?
The questioning of the component of governance concerned with due process is at a very early stage. Will due processes elaborated within the public sector be diluted to allow partnerships to proceed? There are some signs that this might be the case. For example, UNDP is alleged to have breached its own funding guidelines to accommodate specific corporate donations (27), and drug trials proposed under the Secure the Future partnership may erode global ethical norms governing clinical trials (38, 39). This points to the need for greater transparency and public disclosure of GPPP agreements and implementation.
Resources: who pays for partnership?
One of the notable features of a number of the partnerships is the volume of the resources at stake. For example, Pfizers contribution to the International Trachoma Initiative is valued at an estimated US$ 63 million over a two-year period. During the first half of 1999, the Bill and Melinda Gates Foundation (with assets of over US$ 18 billion) made major grant commitments to a number of health GPPPs. These included US$ 100 million to the Bill and Melinda Gates Childrens Vaccine Program, US$ 50 million to the Malaria Vaccine Initiative, US$ 25 million to the International AIDS Vaccine Initiative, and US$ 1 million to the International Trachoma Initiative. Meanwhile, Bristol-Myers Squibb donated US$ 100 million over five years to the Secure the Future partnership.
The total value of private funding to health GPPPs is difficult to estimate, but they clearly provide significant resources for specific health issues. Multilateral resources for disease control are paltry in comparison to the size of private funds for GPPPs. For example, the WHO 199899 total biennium budget for its Control of Tropical Diseases programme was US$ 29 million, while the TB programme received only US$ 7 million over the two-year period (40). In aggregate, private contributors to GPPPs are also significant in comparison to WHOs annual global budget, which is less than US$ 1 billion.
Notwithstanding the invaluable contribution of the private sector to health GPPPs, it appears that the costs (and risks) to the corporate sector may be relatively modest compared to the substantial gains in terms of public relations. First, many contributions are tax-deductible (hence the cost to the company is only approximately half the stated cost). Second, contributions may account for only a small fraction of the profits gained from a particular product. Regarding the Mectizan® Donation Program, for example, ivermectin is Merck & Co.s second largest selling drug (for veterinary purposes), and between 1984 and 1989, sales were greater than those of any other animal health product in the world (41). Bristol-Myers Squibbs annual contribution of US$ 20 million to Secure the Future is just over 0.1% of the companys US$ 18.3 billion in annual sales equivalent to less than one cent a share in each of the five years (42), although it would be more accurate to compare these figures against net profits rather than sales.
Third, the public sector contribution may account for the lions share of the cost of the partnerships activities. For example, the Medicines for Malaria Venture (MMV) aims to raise up to US$ 30 million per year, most of which will come from public sources. Although the private sector partners have agreed to making gifts-in-kind, they have not made more than a vague commitment worth millions per year (4). In other words, over a 10-year period the governing structure of MMV (including corporate representatives) will potentially control US$ 300 million of public funds, while the companies will contribute as they see fit. Even if it is argued that responsibility for the provision of health-promoting goods and services lies with the public sector, it is difficult to justify an equal involvement of the private sector in using these resources.
While GPPPs are clearly bringing new resources to international health, and currently play an essential role in drug and vaccine development, it is not axiomatic that all GPPPs are necessarily good for health. Where GPPPs are successful they can be spectacularly beneficial to health. However, it remains difficult to estimate the actual or potential health consequences of GPPPs (this is especially true for product-development partnerships) which are determined by the effectiveness of the initiative. For example, blinding onchocerciasis will soon cease to be a public health problem as a result of the Mectizan® Donation Program (43). Furthermore, product-development GPPPs represent only one form of GPPPs. In other types of partnership industry does not necessarily enjoy a natural comparative advantage. In these cases the public sector may be able to deal with the health issue at stake without involving industry (Table 1, Table 3). Furthermore, the chequered history of the pharmaceutical industrys marketing of its products in developing countries (44) indicates the need to proceed with caution and to examine each proposed partnership according to appropriate guidelines and criteria. Guidelines may help ensure that the public institutions retain their core characteristics such as integrity, legitimacy, authority and neutrality. Such guidelines are still under development, for example within WHO (28), and have yet to receive adequate scrutiny and public debate.
Conclusions
In this article we have charted the growth of global publicprivate partnerships in health, suggesting that they fall roughly into three categories: product, product development and systems/issues. We have demonstrated that for both the private and the public sector partnership offers major benefits, given the perception that existing and emerging health problems cannot be successfully tackled by one sector alone. For the UN multilateral organizations, partnership with the private sector is seen to have (1) bestowed more business credibility and authority; (2) extended the UNs ability to fulfil its mandates though increased resources; and (3) provided access to private sector skills and management talents.
For the private sector, partnerships have (1) increased corporate influence in global policy-making and at the national level; (2) brought direct financial returns, such as tax breaks and market penetration, as well as indirect financial benefits through brand and image promotion; and (3) enhanced corporate authority and legitimacy through association with UN and other bodies. Many of these benefits of publicprivate partnerships are also true for bilateral donor agencies, with the added advantage of increased authority at the national level. For communities suffering from high rates of HIV infection, trachoma or onchocerciasis, or which will eventually be the beneficiaries of new vaccines, there are clear advantages in partnerships that bring extra resources and targeted programmes.
However, a number of questions are also raised, especially for recipient countries, and should not be brushed aside. There are costs to aid, and some of the publicprivate partnerships have expected relatively high national inputs to their programmes. These costs include guaranteeing distribution networks, storing drugs at ports or airports, training health workers, and conducting trials using drugs that people may not be able to afford in the future. Such programmes do meet needs, but these needs are not determined on the basis of national priority or evidence-based assessment. In relation to drug donation programmes, Kale has argued for greater consultation with recipient countries (prior to launching), explicit cost-benefit analyses, and improved coordination between donation programmes and recipient programmes in the context of government leadership and ownership (45).
Differences in principles and values and issues of governance are therefore important, and need to be addressed. Limited representation of low-income countries in publicprivate partnerships raises questions of who is deciding the international policy agenda, and how much say is being given to recipient nations. The universality of multilateral institutions is diluted by partnerships, since some partners may represent a wide constituency of members, while others, such as the private sector or NGOs, may not. Furthermore, as the tables illustrate, many partnerships target specific countries. While the initiatives may need to be selective, they will primarily choose countries on the basis of the ability to get results. One value of multilateral action is its neutrality, which enables it to fill gaps left by the bilaterals which often support health services according to geopolitical considerations (46).
Accountability may also be interpreted in different ways by different partners. The use of private armies to protect oil pipelines by Shell in Ogoniland, Nigeria, and BP in Colombia was sharply criticized by activists both within and outside those countries. If either organization had had a partnership with a multilateral agency, the loss of perceived neutrality and legitimacy would have been significant. Many of the multilaterals have urged caution in selecting partners, but exercising that caution may be difficult. Moreover, where accountability is felt to be to shareholders or consumers rather than wider communities (or governments), conflicts of interest may occur. For example, the use of alliances within the pharmaceutical industry to fix prices may benefit shareholders but not consumers (47). It was this type of perceived conflict that aroused public concern over the UNDP Global Sustainable Development Facility partnership.
The costs for the private sector seem to be relatively small in relation to overall gains: a potential small loss of resources if programmes do not work, but huge benefits in public relations when they succeed. For bilaterals, there are many difficult questions. For example, how far should public money be supporting hugely profitable private sector firms, in the hope of potential future gains for poor countries or poor people? In one product-development partnership (LAPDAP), WHO and the Department for International Development (DFID) are both subsidizing the pharmaceutical company SmithKline Beecham to stimulate the development of an antimalarial tablet. Although this partnership is not yet governed by any formal agreement, the parties understand that the company will retain the prerogative to set the eventual price of its drug.
Without appropriate forethought, the costs of GPPPs to the United Nations may be even larger. It is conceivable that the profitable activities may be hived off to special partnerships, leaving the public organizations with the more difficult issues (e.g., supporting health systems and training personnel) for which it is harder to raise resources. It is also possible that GPPPs may serve to weaken systems of multilateral governance. This could happen as the control and authority presently vested in governing bodies is transferred to the steering groups of GPPPs, in which the private sector may have greater influence. There is an additional danger that the goals of the UN could be displaced as policies, strategies, resource allocation and activities are increasingly driven (or subject to approval) by industry instead of the organizations governing bodies. Similarly, if the perceived neutrality of the UN is compromised by private sector involvement in its normative activities, the UNs credibility, impartiality and integrity could be undermined. Finally, traditional support for the UN could be undermined as private sector partnerships erode the goodwill of those who believe in a multilateralism in which governments, not corporations, are the decision-makers (48).
Despite claims that social justice has been sought, for example through the Mectizan® donation (49), this review suggests that the promotion of social justice or equity through GPPPs may not be easy. Partnerships are highly selective in their choice of health problem. Resource allocation through partnerships may not be according to burden of disease calculations or need, but according to how a particular partnership reflects the views of its members. The donation of Zithromax® provides a case in point. Although the drug is effective against sexually transmitted infections, it was not until the discovery that Zithromax® could be used to treat trachoma that it was donated for use in some developing countries. It has been suggested that this is because public corporate involvement in the control of stigma-laden sexually transmitted infections would have caused discomfort to the shareholders, while preventing blindness had a more positive appeal.
Some GPPPs are also selective in terms of the countries in which they choose to operate. Mercks Mectizan® Donation Programme operates in all countries where onchocerciasis is endemic and, more recently, in countries where filariasis is co-endemic. Merck has also agreed to donate the drug until these diseases are eradicated. By contrast, the International Trachoma Initiative (ITI) has decided to donate Zithromax® to only five of the 16 WHO priority countries which have significant populations with trachoma, and has committed itself to only two years donation (although the initiative may be scaled-up and extended if the trial period proves successful). Although the diseases targeted and the drugs donated through these two programmes differ and therefore entail widely different risks for the pharmaceutical companies involved, it is clear that the ITI has chosen to work in countries which are perceived to be less difficult. Very poor countries, with large populations, unpopular governments or poor infrastructures may be excluded from global partnership programmes. Problems of exclusion have been raised in other forums too. Participants at a UNAIDS consultation on Bridging the Gap raised the issue of equity in terms of how countries were selected and how other countries not involved would benefit (50). The challenge to the development of GPPPs is to achieve a balance between harnessing the potential of partnerships while avoiding the potential negative effects.
In summary, while there are many positive aspects to the emergence of GPPPs, there remain many uncertainties and some cause for concern. Research is needed to learn more about what makes a partnership 'effective' and in particular, what organizational forms and management arrangements represent best practice for governance, accountability and representation, and what factors contribute to partnership effectiveness on the ground. Harnessing the potential and minimizing the risks of partnership relies on the systematic identification of the potential pitfalls associated with GPPPs and the use of these findings to develop appropriate guidelines, procedures and safeguards. Full advantage should be taken of the current climate of trust and goodwill between public and private sectors, not only to foster new partnership but to ensure that partnership is truly in the interests of international public health. 
Acknowledgements
We would like to thank Sissel Brinchman, Joseph Cook, Tim Evans, Allan Foster, Karin Holm, Adetokunbo O. Lucas, David Mabey, Anne Mills, Michael Reich, Trudie Stubbs, Derek Yach, and staff of the Health Policy Unit, London School of Hygiene and Tropical Medicine, for sharing their views and documents with us.
References
1. Mitchell-Weaver C, Manning B. Publicprivate partnerships in Third World development. The 20th Norma Wilkinson Memorial Lecture. Geographical Paper. Reading, Reading University, 1990.
2. Leading companies to mobilize against global AIDS epidemic. Press Release. Geneva, UNAIDS, 23 October 1997.
3. Chernesky MA. How can industry, academia, public health authorities and the Sexually Transmitted Diseases Diagnostics Initiative work together to help control sexually transmitted diseases in developing countries? Sexually Transmitted Diseases, 1997, 24 (2): 6163.
4. Ridley R, Gutteridge WE, Currat LJ. New Medicines for Malaria Venture: a case study of the establishment of a public sector private sector partnership. Paper presented at the Third Global Forum for Health Research, 811 June 1999, Geneva.
5. Berkley S, Lenton C. The International AIDS Vaccine Initiative. Paper presented at the Third Global Forum for Health Research, 811 June 1999, Geneva.
6. Kickbusch I, Quick J. Partnerships for health in the 21st century. World Health Statistics Quarterly, 1998, 51: 69.
7. WHO European partnership project on tobacco dependence. Draft document. Geneva, World Health Organization, 22 December 1998.
8. Public-Private Partnership Programme of UNDP. Internet communication on November 12 1999 at http:/www.undp. org/info21/business/annex3.html
9. Letter from SG Cowal, Director of External Relations, UNAIDS, to undisclosed recipient dated 19 January 1999. Geneva, UNAIDS, 1999.
10. UNAIDS launches initiative to help bridge gap in access to HIV/ AIDS-related drugs in developing world. Press Release. Geneva, UNAIDS, 5 November 1997.
11. Guidelines for drug donations.Geneva, World Health Organization, May 1996.
12. Godal T. Fighting parasites of poverty: public research, private industry and tropical disease. Science, 1994, 264: 1864 1866.
13. Finance report of the International AIDS Vaccine Initiative: accelerating the development of preventive HIV vaccines for the world. New York, International AIDS Vaccine Initiative, 1995.
14. Malaria Vaccine Initiative: statement of purpose. Seattle, WA, Program for Appropriate Technology in Health, 1999.
15. Bristol-Myers Squibb commits $ 100 million for HIV/AIDS research and community outreach in five African countries. Press release, Washington, DC, 6 May 1999.
16. Innovative private sector venture under way. TDR News, 1999, 58: 3.
17. Roll Back Malaria: report by the Director-General to the Fifty Second World Health Assembly. Provisional agenda item, 13 May 1999. Geneva, World Health Organization.
18. Nunn P. Personal communication with Head, Stop TB Initiative. London, 18 June 1999.
19. Annan K. Address to the US Chambers of Commerce, 8 June 1999, Washington, DC.
20. Walt G. International organisations in health: the problem of leadership. In: Pocantico Retreat: Enhancing the performance of international health institutions. Rockefeller Foundation, Social Science Research Council, Harvard School of Health, 1999.
21. Hancock T. Caveat partner: reflections on partnership with the private sector. Health Promotion International, 1998, 13 (3): 193.
22. Lenton C. Remarks made at parallel session number 7.1 at the Global Forum for Health Research, 8 June 1999, Geneva.
23. Frost L, Reich M. Mectizan® Donation Program: origins, experiences, and relationships with co-ordinating bodies for onchocerciasis control. Department of Population and International Health. Boston, Harvard School of Public Health, 1998.
24. Widdus R, Evans P. Lessons learned from the Childrens Vaccine Initiative 1990-1999. Paper presented at the Third Global Forum for Health Research, Geneva, 9 June 1999.
25. Bellamy C. Public, private and civil society. Statement of UNICEF Executive Director to Harvard International Development Conference on Sharing responsibilities: public, private and civil society. Cambridge, MA, 16 April 1999.
26. Dukes G. The contribution of the private sector: an introduction. Australian Prescriber, 1997, 20 (Suppl 1): 7475.
27. Klein N. UN pact with business masks real dangers. Toronto Star, 19 March 1999.
28. WHO guidelines on collaboration and partnerships with commercial enterprises. Draft discussion document dated 24 July 1999. Geneva, World Health Organization, 1999.
29. Deen T. UN calls for new partnership with arms industry. InterPress Service Daily Journal,7 (129), 7 July 1999. Internet communication at http:/www.globalpolicy.org/reform/ armsindy.htm
30. Rosenau JN. Governance in the twenty-first century. Global Governance, 1995, 1 (1): 1343.
31. World Bank. Governance: the World Banks experience. Washington, DC, World Bank, 1994.
32. Mabey D. Personal communication, Member of Technical Expert Committee, International Trachoma Initiative, London, 28 June 1999.
33. Haas PM. Do regimes matter? Epistemic communities and Mediterranean pollution control. International Organisation, 1989, 43 (3).
34. Bain K. Building or burning bridges? The accountability of transnational NGO networks in policy alliances with the World Bank. Paper prepared for the Conference on NGOs in a Global Future, Birmingham 1999.
35. Bill and Melinda Gates Childrens Vaccine Program: Outline for Action. Seattle, WA, Program for Appropriate Technology in Health, 1998.
36. Cook J. Personal correspondence with J Cook, Executive Director, International Trachoma Initiative, 20 May 1999.
37. Muraskin W. The war against hepatitis B: A history of the International Task Force on Hepatitis B Immunization. Philadelphia, Pennsylvania University Press, 1995.
38. International guidelines for ethical review of epidemiological studies. Geneva, Council for International Organizations of Medical Sciences, 1991.
39. Khan P. UNAIDS to publish guidelines on ethics of vaccine trials. International AIDS Vaccine Initiative Report, Spring 1999. New York, IAVI, 1999.
40. Proposed programme budget for the financial period 19981999. Geneva, World Health Organization, 1996.
41. Eckholm E. River blindness Conquering an ancient scourge. The New York Times Magazine, 8 January 1989.
42. Bristol-Myers Squibb Secure the future announcement and media reaction. Internet posting from the Treatment Access Forum. Internet communication, 11 May 1999 at http://www.hivnet.ch: 8000/treatment-access/tdm
43. Chetley A. A healthy business? World health and the pharmaceutical industry. London, Zed Books, 1990.
44. Lucas A. Personal communication, 13 July 1999.
45. Kale OO. Review of disease-specific corporate drug donation programmes for the control of communicable diseases. Paper presented at the Symposium: Drugs for Communicable Diseases Stimulating development and availability, Paris, 15 October 1999.
46. Drager N et al. What determines aid for health: an empirical analysis of bilateral aid flows. International Conference on Macro-Economics and Health in Countries in Greatest Need. Geneva, World Health Organization, 1992.
47. Weber J, Barrett A. Volatile combos: pharmaceutical alliances can boost both players health or drag them down. Business Week, 25 October 1999.
48. Korten D. The United Nations and the corporate agenda. Text circulated on the internet, July 1997. Internet communication at http://www.igc.org/globalpolicy/reform/korten.htm
49. Foege WH. Ten years of Mectizan. Annals of Tropical Medicine and Parasitology, 1998, 92 (1): 710.
50. Towards the creation of strategic partnerships: improving access to drugs for HIV/AIDS: report of a consultative meeting, 30 June2 July 1997, World Health Organization, Geneva. Geneva, UNAIDS, 1997.
Correspondence
K. Buse
Department of Epidemiology and Public Health, Yale University
PO Box 208034, New Haven, CT, 06520-8034, USA
E-mail: kent.buse@yale.edu
a This is not to suggest that the selection of WHO experts is perfect, but to emphasize that there are checks and balances which, for example, ensure attention to questions of representation of, say, developing countries or women