What game theory can teach us about controlling pandemics

Although the daily number of new covid-19 infections globally is thankfully falling, there are still around 240,000 new infections worldwide each day, and there are many “hot spots” of uncontrolled transmission, including the UK, where daily case rates are rising. This situation presents an opportunity for the SARS-CoV-2 virus to mutate in ways that could render our current vaccines less effective. The best way to be sure that we avoid such a scenario is to go after the virus aggressively in every nation, to drive down the number of infections all around the world, and to do so as quickly as possible. 

Unfortunately, the global campaign to vaccinate the world is faltering. The COVID-19 Vaccines Global Access Facility (COVAX) was launched last year with the aim of distributing two billion vaccine doses to low- and lower-middle income countries by the end of 2021. However, because around three dozen rich countries hoarded the global vaccine supply, COVAX has only been able to distribute 365 million doses at the time of writing. At the current rate of global vaccination, poorer nations will have to wait until 2023 to see widespread vaccination. 

The good news is, it’s not too late to take a game-theory approach to “change the game” and steer the global pandemic in a safer direction—and in so doing, to lay a foundation of preparedness for the next pathogen with pandemic potential that emerges in the future.

From a game-theory point of view, the fundamental problem with COVAX and other aid-centric global health efforts is that everyone prefers for others to pay for the global public goods that result from these efforts—what’s known in economics as “the free-rider problem.” To overcome the free-rider problem, we need to find a way to reward the wealthy countries that help the rest of the world escape the pandemic, and to do so in a way that is so compelling that these wealthy countries view providing such help as essential to their own national interest. 

The military alliance NATO provides a model for how that could work. Countries that join NATO commit to provide troops, ships, aircraft, and funds to support NATO operations. In doing so, NATO members make the world a safer place, but there is no free-rider problem associated with this global public good because member countries know that being part of NATO is essential to their own national security. 

We envision a pandemic vaccine alliance (let’s call it “PANVAX”) that works similarly to NATO, one in which member nations work together to make the world a safer place, but also prepare and commit to one another’s mutual defense. But unlike NATO, this will be an alliance open to every country in the world willing to support it, and one in which low- and middle-income countries will play an essential, and eventually leading, role. 

We should state up front that our NATO analogy does not mean we support the “securitization” of global health—in fact, we believe there are dangers in conflating public health with military and intelligence concerns. But we feel there are valuable lessons to be learned from NATO as an alliance model that could be applied to pandemics, especially in terms of detection, containment, and response.

We initially proposed PANVAX on 7 May 2021 in a public meeting of the National Academies of Science, Engineering, and Medicine on global coordination and financing for influenza vaccine preparedness. PANVAX would aim to ensure the world’s biological security, by enabling the rapid development, testing, production, and distribution of enough vaccines for all people in any PANVAX member country against any novel pandemic pathogen. 

As we envision it, PANVAX will have three key functions. The first two (rapid pathogen detection and rapid vaccine development and trials) create global public goods. The third (rapid and robust vaccine production) exclusively benefits PANVAX members and hence will incentivize wealthy countries to support PANVAX, even if they only care about themselves.

The first function is rapid pathogen detection, to identify and characterise any new pandemic pathogen before it has infected a large number of people.

The notion of biological security, that people can be kept safe from unknown future biological threats, would have been unthinkable even a decade ago. Yet, with recent advances in genetic pathogen surveillance and vaccine technology, such security is now within reach. 

Consider the challenge of discovering when a pandemic pathogen has emerged. Disease-causing pathogens are discovered all the time, but most have no pandemic potential. Moreover, for pathogens like SARS-CoV-2 that only sometimes cause severe infection, a clear pattern of severe disease may not be evident until the pathogen is so widespread that containment is no longer possible. 

Fortunately, new genomic surveillance methods have been developed in recent years, including bioaerosol sampling on animal farms and detecting a pathogen’s genetic material in sewage. PANVAX can enable such rapid pathogen detection by funding and establishing a stronger global surveillance network and empowering low- and middle-income countries to develop and sustain their own surveillance capabilities. Funding and supporting such a global pathogen surveillance network, building on existing initiatives such as the newly launched WHO Hub for Pandemic and Epidemic Intelligence, should be a top priority right now for the United States and other high-income countries, as this will allow us to identify and prepare for new coronavirus variants as soon as possible, no matter where they emerge.

In our NATO analogy, rapid pathogen detection through genomic surveillance is akin to radar detection of a nuclear missile immediately after launch, and PANVAX’s global surveillance network is akin to NATO’s network of radar installations.

The second function that PANVAX would perform is rapid vaccine development and trials, to develop and test vaccines against the pandemic pathogen while it remains mostly regionally contained. 

Knowing the genetic code (the sequence) of a pandemic pathogen allows vaccines to be developed and produced extremely rapidly. For instance, just 66 days after the draft genome sequence for SARS-CoV-2 was published online on 10 January 2020, the first phase 1 clinical trial of a vaccine candidate began. Covid-19 vaccines were developed in under a year, the fastest vaccine development in history (the previous record was four years, for mumps vaccine). 

One reason why covid-19 vaccines were developed so quickly is that SARS-CoV-2 spread rapidly into high-income nations like the US and UK that have strong clinical trial sites. A future epidemic or pandemic may remain in just one or a few regions. In this situation, the vaccine development and testing process will require urgent trials in the “ground-zero countries” in which the pathogen has emerged (just as the Ebola vaccine, Ervebo, was trialed in Guinea during the 2014-2016 Ebola epidemic in west Africa). 

PANVAX could enable such rapid vaccine trials in several ways: by establishing ethical and scientifically-valid protocols for determining a vaccine candidate’s safety and effectiveness in the midst of an incipient outbreak; by partnering ahead of time with national health authorities to be ready to rapidly deploy those testing protocols; and by ensuring that ground-zero countries receive prioritized access to the first vaccine doses produced—and hence have a strong incentive to cooperate with PANVAX in the early days of the crisis. 

In our NATO analogy, rapid vaccine development and testing is akin to destroying a nuclear missile while it remains in space, whereas the traditional approach is like waiting until the missile lands and explodes before doing anything to mitigate its impact.

Thirdly, PANVAX would facilitate robust and reliable vaccine production on a massive scale.

The race to develop and produce covid vaccines has been a scientific and technological marvel, but also a sad spectacle of selfishness and fear. In the moment of greatest need, many rich countries took stock of what they had for themselves and then refused to share. Because of such “vaccine nationalism,” existing production supply chains that relied on international cooperation suddenly became unstable and unreliable, disrupting and slowing global vaccine production. 

PANVAX can avoid such production-coordination failures by establishing, ahead of time, a supply chain contained entirely within the PANVAX Alliance that is capable of rapidly, reliably, and robustly producing enough vaccine doses for everyone in the alliance. 

The electricity reliability systems that have been introduced in the United States over the past two decades provide a useful model for how PANVAX could establish and rapidly mobilize its vaccine supply chain. In the early 2000s, a series of unsettling price spikes and regional brownouts revealed the unreliability of our electricity supply. Most of the Independent System Operators (ISOs) who run America’s power grids (all except Texas) responded by introducing so-called “forward-capacity auctions” to ensure reliability. In a forward-capacity auction, the ISO specifies how much overall capacity will be needed in the future to reliably meet peak demand and takes bids from qualifying suppliers. 

Winners of the auction receive a subsidy in exchange for committing to be available in the future to provide electricity whenever the ISO calls for it, even on a moment’s notice. By design, “the future” in which suppliers commit to be ready to provide electricity is sufficiently far away (e.g. three years in New England) that someone can bid in the auction and then, after winning, have time to build a new facility from scratch to meet its reliability obligations. 

In much the same way, PANVAX could specify all the capabilities (detection, development, testing, and production) that it needs in order to be able to rapidly produce enough vaccine doses for everyone in the alliance, then identify and fund those best positioned to meet those needs. This includes funding, importantly, those in low-income countries that have traditionally lacked these capabilities, but that need to play a central role for PANVAX to be successful—initially in rapid pathogen detection and vaccine testing, but growing over time also into key centers for vaccine production and infectious-disease science. 

PANVAX as we envision it is a voluntary, opt-in mechanism. We believe that the incentives to join are so strong that most nations would participate but, as with any new voluntary programme, there is a risk that countries may take a “wait and see” approach, only joining the alliance once it is well-established. This is a potential weakness of our proposal, relative to a compulsory mechanism in which all nations are required to fund a pandemic vaccine facility according to their means. But it also means that there is a leadership opportunity here for high-income countries committed to the PANVAX vision. Once these “first-mover countries” develop the capabilities needed to rapidly mass-produce vaccines at a global scale, the rest of the world will naturally follow. Working together all across the world, we can then protect ourselves from the next pathogen with pandemic potential—whether it’s an even-more-dangerous SARS-CoV-2 variant or something entirely new.

David McAdams Professor of Business Administration and Economics, Fuqua School of Business, Duke University, Durham, NC, USA

Gavin Yamey Professor of Global Health and Public Policy, Center for Policy Impact in Global Health, Duke Global Health Institute, Duke University, Durham, NC, USA

Competing interests:

DM declares that he has no competing interests.GY declares that he is a signatory to the People’s Vaccine Campaign and is a funding member of Amnesty International, one of the members of the People’s Vaccine Alliance. He was a member of the COVID-19 Vaccine Development Taskforce, hosted by the World Bank, and participated as an academic unpaid adviser in the consultation process that led to the launch of COVAX, a global Covid-19 vaccine sharing mechanism. He has received grant funding from the WHO, Gavi, the Vaccine Alliance, and the Bill & Melinda Gates Foundation.