If you are like me, you spend a certain amount of time trying not to remember the pandemic experience. But COVID-19 pandemic did cause more than one million American deaths. In a world of sane and sensible prioritizing and policy-making, spending some time and effort focused on how to reduce the risks and costs of a future pandemic seems potentially productive. Alex Tabarrok discusses a few pragmatic possibilities in “Pandemic preparation without romance: insights from public choice” (Public Choice, published online April 16, 2025).
One metaphor for America’s level of unpreparedness for the COVID-19 pandemic is warehouses of rotting N95 masks. Tabarrok notes:
[C]onsider that The Strategic National Stockpile (SNS) of personal protective equipment (PPE) was severely inadequate to meet the demands of the COVID-19 pandemic. At the start of the pandemic, the stockpile had only about 35 million N95 masks on hand, far short of the estimated 3.5 billion that would have been needed to adequately protect healthcare workers and first responders. Moreover, much of the stockpile was rotting as the N95 masks were more than 10 years old by the time of the pandemic.
I should emphasize that even taken all together, Tabarrok is not claiming that his recommended policies can eliminate the costs of future pandemics. But if we could reduce the cost by say, just 10% , the US savings alone would have been more than 100,000 lives and more than $1 trillion in lost economic output. Here are four of his options:
#1: Testing for disease at sewage treatment plants
People infected with SARS-CoV-2 shed genetic material from the virus in their feces (Bivins et al. 2020). Wastewater surveillance can detect the presence, concentration and growth of this genetic material before people present clinically. Thus, wastewater surveillance gives public health officials an early warning which can be used to allocate scarce resources and to implement control measures. More generally, wastewater surveillance can detect a host of viral and bacterial pathogens including influenza viruses, poliovirus, norovirus, hepatitis A and E viruses and bacteria such as Escheric hia coli and Salmonella. Wastewater surveillance to monitor antibacterial resistance may be of special importance (Philo et al. 2023; Singer et al. 2023). As with surveillance for SARS-CoV-2, wastewater surveillance more generally can be used to predict disease outbreaks more quickly, track the spread and virulence of pathogens and novel variants of concern, and inform and provide feedback to public health decisions (Wu et al. 2020).
#2: Build a vaccine library, by doing the research in advance on vaccines for viruses most likely to cause an outbreak
In 2016, the WHO identified 11 viruses with the greatest potential to cause severe outbreaks. Gouglas et al. (2018) estimated that developing at least one vaccine candidate for each of these viruses up to phase 2a would cost approximately $2.8 to $3.7 billion in total (see also Krammer 2020). Bringing a vaccine candidate up to phase 2a means designing the vaccine and evaluating it for safety and essentially “proof of concept” in small trials. Prior to a significant outbreak, it would not be possible to run phase 3 efficacy trials. It should be clear that these costs are small, almost trivial, relative to the expected gains. It’s notable that SARS-CoV-1 was on the WHO’s list. The knowledge gained from studying SARS-CoV-1 helped to speed a vaccine for SARS-COV-II but had SARS-COV-I vaccines been developed to Phase 2a prior to the COVID pandemic, for example, we could have likely knocked months off the development process for SARS-COV-II, saving perhaps millions of lives and trillions of dollars worldwide.
#3: When a virus hits, test the vaccines with “human challenge trials”
COVID vaccines were tested through traditional randomized controlled trials (RCTs) in the field. In an RCT, participants are randomly assigned to either a vaccinated (treatment) group or an unvaccinated (control) group, and both groups resume their normal activities until enough participants contract COVID to establish a statistically significant difference in infection rates. A major drawback of RCTs in a pandemic is the unpredictability of reaching the infection threshold required for statistical significance. If infection rates are low or participants take steps to avoid exposure, trials can be prolonged, delaying vaccine rollout. While increasing the trial size can reduce these delays, it also increases the cost and complexity of the trials.
In contrast, in a human challenge trial (HCT), participants are randomly split into two groups and all of them are deliberately exposed to the virus, accelerating the timeline for obtaining results. Since participants are deliberately exposed the number of participants in a human challenge trial can be much smaller than in an RCT, perhaps on the order of 50–100. Most importantly, where an RCT might take years to produce results, a HCT can have results in a matter of months or weeks (Eyal and Lipsitch 2021; Nguyen et al. 2021). For a variety of reasons, HCT are not necessarily full substitutes for RCTs, but they are surely complements and should be used in emergencies.
#4: A Pandemic Trust Fund
As another example, some $60 billion was spent on special programs to pay furloughed pilots, flight attendants, and other airline staff as travel demand plummeted. Why? One factor was that the airlines were already well organized and politically active. The airlines, for example, spent over one hundred million dollars on lobbying in the year before the pandemic (Evers-Hillstrom 2020). During the pandemic, the airlines were also joined in their lobbying efforts by the airline unions making for a politically powerful team on both sides of the aisle. The lines of power were also well defined. The airlines knew, for example, which members of Congress sat on the requisite committees and what they needed.
In contrast, OWS [Operation Warp Speed, the program for developing vaccines] was a new program with few concentrated interest groups and no previous lobbying efforts. Although some of the vaccine manufacturers understood lobbying, there was no locus of support in Congress because committee responsibilities for a program like OWS had not been established. OWS was run primarily out of the executive and the DOD [Department of Defense]. The program was also controversial from the beginning and any lobbying at the time from the vaccine manufacturers would have been highly scrutinized.
The lesson from political economy is that we do not want emergency funds to be drawn, or to be perceived to be drawn, from other programs. Pre-approved legal authority to spend is necessary to quickly address a low-probability, high-cost emergency. One way to do this would be to establish a Pandemic Trust Fund (PTF) nominally composed of say $250 billion in US government bonds. The PTF would be something of an accounting fiction, similar to the Social Security Trust Fund, but accounting fictions can have real effects. … By clearly denoting pandemic spending rights, a pandemic trust fund would avoid budget battles in the event of a pandemic. At $250 billion and 3% interest, a PTF could also generate annual revenues of $7.5 billion for ongoing pandemic spending. Some of this spending would be wasted but sausages and legislation both require pork as an input.
In the context of total US federal spending, none of these steps are especially costly, but having them in place could make a real difference. As Tabarrok points out, there were plenty of well-publicized warnings in the decade or two about risks of pandemics, including high profile stories in outlets like TIME and CNN, a Bill Gates TED talk seen by millions, and even the 2011 movie Contagion. But when the crunch came, America was not well-prepared. There will be a next time.
