COVAX, Global Health Governance and the Failure of Multilateralism – Rebecca Brown

The COVID-19 pandemic has been a breakpoint for global health governance, necessitating unprecedented cooperation worldwide. However, the success of global initiatives has been hampered by lack of state interest. This piece investigates why one such initiative, the COVID19 Vaccine Global Access Facility (COVAX), was abandoned by its high-income state supporters, setting out its vision, development and failings. In doing so, this piece seeks to outline issues within the Global North’s approach to multilateral health governance more broadly.

In July 2020, when the development of an effective vaccine against COVID-19 was in sight, assuring timely access to this vaccine became the essential instrument in the Global North’s political toolbox. The UK secured a ‘portfolio’ of vaccines through privately contracting with a number of pharmaceutical manufacturers, while Australian Prime Minister Scott Morrison announced an agreement with AstraZeneca, before this in fact had been finalised. Now, in May 2021, a 40-year-old may be vaccinated tomorrow, by the end of the year, in some years, or has already been vaccinated – depending on their country of residence. Such disparity in access to vaccines is often attributed to the inadequacy of governmental agreements with private companies. However, this analysis presupposes contemporary knowledge of which vaccines indeed proved effective, and glosses over broader issues of vaccine inequity entrenched by global power differentials and healthcare’s commercialisation. 

To that end, the COVID-19 Vaccine Global Access Facility (COVAX) appeared a promising solution. Initially floated at the extraordinary G20 meeting in March 2020, and launched by a number of state leaders the following month, COVAX swore to maximise access to vaccines, ensuring populations in all participant countries could soon receive a vaccine. Yet, COVAX remains underfunded, and vaccine nationalism – with states competing to secure doses from manufacturers – stymied its ability to acquire and distribute vaccines. COVAX, set out as a multilateral solution to the COVID-19 pandemic, has been reduced to a vessel for financial contributions to lower-income states; its failure represents a broader failure of collective action, and a continuation of neo-colonialist attitudes towards health governance.

An overview of the COVAX initiative

COVAX is the vaccine ‘pillar’ of the Access to COVID-19 Tools (ACT) accelerator initiative, a joint project between the World Health Organization (WHO) and several non-governmental organisations, following the G20 appeal to develop a ‘sustained funding and coordination platform to accelerate the development and delivery of vaccines, diagnostics and treatments’. Co-led by the WHO; Gavi, the Vaccine Alliance; and CEPI, it intends to assist with vaccine access through two distinct operations: first, sourcing funding for their research, development and production, through facilitating advanced purchases of a putative suite of vaccines; and second, through an Advance Market Commitment (AMC), whereby states, organisations, and private organisations jointly fund the distribution of successful vaccines to middle- and low-income countries.

The Facility solicited state participation from July 2020, with interested states choosing to fund one or both of these operational limbs. For the former, states pledged to procure an agreed volume of vaccine doses by providing an upfront payment attached to the number of doses requested, and a purchase agreement guaranteeing the remainder of the price once the vaccines were produced. This commitment provided access to any successful vaccine from the portfolio of potential vaccines acquired by the Facility, the development of which were, in turn, to be funded by the Facility. 

Two funding schemes were available to states: a Committed Purchase Arrangement and an Optional Purchase Arrangement. Under the Committed Purchase Arrangement, a state provided US$1.60/dose upfront, with a financial guarantee of US$8.95/dose; however, it could only opt-out of the particular vaccine allocated to it if the sale price exceeded twice the estimated cost. Under the Optional Purchase Arrangement, a state could elect to not receive a certain vaccine from the portfolio, by paying the higher cost of US$3.10/dose up-front; this scheme was targeted at states who had concluded bilateral agreements with manufacturers. For either arrangement, if the ultimate price per dose – as negotiated between the Facility and manufacturers – was lower than the financial guarantee, the state would pay only for the quantity required, or could select to procure an additional amount. If the price exceeded the guarantee, the state was only committed to pay the agreed amount, albeit resulting in fewer doses purchased. 

The second limb sources funding to purchase and distribute vaccines to low- and middle-income states; states are still able contribute towards this fund, through finance or dose donation. While the AMC initially intended that its 92 recipient states would not need to purchase these vaccines, at a Gavi board meeting in earlier October 2020, the decision was made to charge states between US$1.60 and US2.00 per dose, likely a result of the Facility’s underfunding. Payment, however, will be ‘flexible’ until the end of 2021, with a possibility of further extension, and Gavi and the World Bank have sourced funding to assist with these new costs.

The Facility was envisaged to operate as follows. Through the provision of funds by the Facility, vaccine manufacturers are encouraged to expand production even prior to gaining regulatory approval, as to meet the agreed level of product; the assurance of funding removes the usual risks borne by the producer associated with that investment. On the other hand, ‘self-financing’ states are guaranteed access to any successful vaccine developed out of participating producers, rather than individually negotiating an agreement with each manufacturer, a challenging task both politically and practically. As purchase of the vaccine is centralised, each state receives sufficient doses to vaccinate the 20% of its population most in need, with further vaccines available for purchase once every state is able to reach this 20% threshold. Thus, with broad participation in the project, large-scale production of vaccines could have occurred at an early stage; the Facility’s collective purchasing power would have allowed it to negotiate lower prices with manufacturers; and vaccines could have reached the most vulnerable people in every country in a timely manner.

In actuality, the Facility remained underfunded: only 52 (primarily small) states finalised commitment agreements under the first limb, and most contributions to the AMC were provided long after the crucial stage of negotiating vaccine acquisition. The effects were compounded by the bilateral agreements states entered into with manufacturers, and the EU’s own facility through which vaccines were purchased. Altogether, high-income states, covering just 14% of the global population, purchased 53% of the potential vaccine supply. COVAX lacked the ability to compete. As such, while it has been able to provide some deliveries to both self-financing and AMC states, this has occurred at a far slower rate, and with far fewer doses available than originally envisaged. 

Of course, a number of factors contribute to these issues. The effectiveness of the Facility and the Accelerator are undermined by their failure to provide essential incidental elements to vaccine distribution – health facilities; know-how; training; and health education, among other issues in its design. An obvious flaw in COVAX’s rhetoric of global vaccine equity is the separation of the AMC from the remainder of the Facility, essentially creating a two-class system. However, the choice by high-income states to place all their eggs in one vaccine basket – negotiating individual, expensive agreements, rather than committing to an initiative that would not only help the global good, but those states themselves – represents a near-inexplicable failure of collective action.

COVAX and foreign aid

This failure was not inevitable. The ACT Accelerator’s launch on 24 April 2020 was headed by the European Commission and France, alongside the WHO and the Bill and Melinda Gates Foundation. Other state leaders provided political expressions of support, focusing on equitable access to COVID-19 antiviral drugs and vaccines, and the event’s official release committed to an unprecedented level of partnership in fulfilling the shared aim of equitable global access. This appeal was echoed at the 73rd World Health Assembly, wherein states called for ‘universal, timely and equitable access to, and fair distribution of … essential health technologies’, and requested that the WHO develop options for scaling up the development, manufacturing and distribution capacities needed for such equitable access and distribution. 

As time went by, high-income states started altering the initial scheme: the UK sought the flexible Optional Purchase Arrangement model; in September, the ‘COVAX Exchange’ was adopted, allowing participants to trade vaccines obtained through private agreements –  at the risk of high-income states swapping out their less effective vaccines. Ultimately, despite the initial support, very few high-income states were interested in contributing: while some – including Australia – expressed an intention to participate, this went no further.

However, even from an early stage, cracks were visible. Despite the European Commission’s pledge to raise EUR€7.5 billion to fund the Accelerator, it was never interested in promoting the Facility’s first limb, to the extent that it created its own purchasing scheme for EU member states. France, while launching the Accelerator and contributing funding, adopted the ‘clear’ position of purchasing vaccines through the EU’s mechanism alone. Indeed, many outward statements of support referred only to the AMC component: despite the Facility relying on the creation of a multilateral bargaining power through high-income participation, vaccine equity was reduced to the provision of foreign aid to lower-income nations, either in the form of financial contributions, or vaccine doses themselves.

The reduction of ‘foreign aid’ to something given by donor to donee; rich country to poor, is pernicious. As Gostin notes, it leads to unequal power relationships, with funding – deemed discretionary – provided only when politically expedient. COVAX, as a multilateral initiative, did not offer the heroic branding – the ‘blue flags with yellow stars’ on vaccine parcels – to warrant sufficient funding. This does not account for the whole COVAX story, however. The failure of high-income states to join the first limb betrays not only a keen distrust of multilateralism, but a harmful view of who healthcare is for. Health aid is viewed as something given, not received, and vaccine inequity as an issue faced by only the Global South. But such distinctions between ‘our disease’ and ‘their disease’ perpetuate paternalistic and unhelpful attitudes towards global health governance. Indeed, these resurface in the form of other hinderances to vaccine equity: most recently, the block to the proposal to waive intellectual property protections over COVID-19 vaccines by high-income states – including the largest supporters and funders of the COVAX AMC. 

High-income countries did not join COVAX because they did not view it as something for them – however much it would have benefitted their populations, and however much it would have benefitted the public good. As long as the Global North holds this view, limiting the role of global health governance to the provision of financial aid to states it views as lesser, multilateral solutions will not succeed.

Rebecca Brown is a Research Assistant at the University of Cambridge’s Faculty of Law, and the Convenor of the 2021 Cambridge International Law Conference.

Suggested citation: Rebecca Brown, ‘COVAX, Global Health Governance and the Failure of Multilateralism’ on ILA Reporter, 6 May 2021 <>