Re-evaluating the benefits of non-pharmaceutical interventions during the 1918 pandemic
The current COVID-19 crisis raises many questions for economists and policymakers (e.g. Baldwin and Weder di Mauro 2020a, 2020b). Past pandemics such as the 1918 influenza offer an interesting opportunity to evaluate the potential impact of pandemics on economic activity (Barro et al. 2020a, 2020b) and the effect of non-pharmaceutical interventions (NPIs) such as school closures and social distancing on the spread of the disease and on economic activity.
The medium-term impact of non-pharmaceutical interventions on the death rate
In the US, the 1918 influenza was probably spread by troops returning from Europe in the autumn of 1918. The epidemiological literature shows that all large cities implemented some kind of non-pharmaceutical intervention, such as quarantines, social distancing, or school closures. However, the degree and the rapidity of the responses varied even within regions.
For example, New York responded rapidly to the pandemic and managed to flatten the epidemic curve through strictly enforced isolation and quarantine. According to Markel et al. (2007), this allowed the city to experience the lowest death rate on the East Coast. In contrast, Pittsburgh took action later, which resulted in the highest excess mortality burden in the sample studied.
Epidemiological studies of the 1918 pandemic mostly focus on the capacity of non-pharmaceutical interventions to lower the death rate during the peak of the epidemic, i.e. to ‘flatten the epidemic curve’. In a recent paper (Chapelle 2020), I examine the medium-run impact of non-pharmaceutical interventions on the 1918 pandemic, looking at their consequences once the main wave was over.
Using an event study, I compared how death rates evolved from 1910 through 1924 in cities that implemented non-pharmaceutical interventions for longer versus shorter periods. Figure 1 summarises the main result. Cities that implemented longer non-pharmaceutical interventions in 1918 had no significant variation from their pre-1918 annual death rate and managed to decrease their death rate in 1918 compared with cities that adopted shorter non-pharmaceutical interventions. However, they also ended with relatively higher mortality levels in the subsequent years.
Robustness checks reject the possibility that the phenomenon might be explained by changes in the demographic structure or the mortality in 1918. Furthermore, they tend to support the hypothesis that non-pharmaceutical interventions lower the herd immunity, thus making the population more susceptible to ensuing strains of influenza. Indeed, the subsequent influenza epidemics, with the exception of avian influenza, have been caused by descendants of the 1918 virus.
The estimated number of deaths after five years caused by longer non-pharmaceutical interventions is greater than the estimated number of lives saved in 1918. These results suggest that optimal policy responses should account for herd immunity and include an exit strategy that can compensate for the lower herd immunity.
Figure 1 Event study: Estimates of the aggregate impact of non-pharmaceutical intervention implementation duration on death rates
Non-pharmaceutical interventions and economic growth
Caution must be exercised in inferring any causal links between economic activity and non-pharmaceutical interventions in US cities, as discussed in Correa et al. (2020), for two main reasons. First, the manufacturing sector in cities that had high mortality rates and longer non-pharmaceutical interventions in 1918 had historically different economic dynamics, making any causal inferences hard to establish. This finding was later confirmed by Lilley et al. (2020).
Second, given the negative effect of the 1918 non-pharmaceutical interventions on the mortality rate in subsequent years, such interventions did not preserve human capital in the medium run. This casts doubts on the channels likely to explain the positive relationship between health policy and economic growth.
My findings tend to support the intuitions of Toda (2020a,b) that draconian non-pharmaceutical interventions might be suboptimal because they prevent the building of herd immunity, and of Benmelech and Frydman (2020) that large US cities recovered rapidly from the 1918 influenza pandemic.
This does not mean that epidemics do not have any economic consequences. Several papers have used other data and contexts to document the negative impact of the 1918 pandemic at the aggregate level (Barro et al. 2020a) or at the local level (Dahl et al. 2020).
It would be difficult to draw any inferences for the predicted impact of non-pharmaceutical interventions on the COVID-19 crisis, not least because the magnitude and scale of the interventions from the two pandemics are different. Today, non-pharmaceutical interventions are mainly implemented on a national or state scale, rather than at the city level. Moreover, pharmaceutical technologies were less developed in 1918 than today, and the capacity to produce a new vaccine within a reasonable time was much lower a century ago.
COVID-19 and the 1918 influenza are two different viruses, and evidence on the immune response after contracting COVID-19 remains scarce. We can hope that the development of a treatment or a vaccine will limit the spread of the disease in the next years (Bethune and Korinek 2020).
Baldwin, R, and B Weder di Mauro (2020a), Economics in the time of COVID-19, a VoxEU.org eBook, CEPR Press.
Baldwin, R, and B Weder di Mauro (2020b), Mitigating the COVID economic crisis: Act fast and do whatever it takes, a VoxEU.org eBook, CEPR Press.
Barro, Robert J, José F Ursùa and Joanna Weng (2020a), “The coronavirus and the great influenza pandemic: Lessons from the ‘Spanish flu’ for the coronavirus’s potential effects on mortality and economic activity”, NBER Working Paper 26866.
Barro, Robert J, José F Ursùa and Joanna Weng (2020b), “The coronavirus meets the Great Influenza Pandemic”, VoxEU.org, 25 April.
Benmelech, Efraim, and Carola Frydman (2020), “The 1918 influenza did not kill the US economy”, VoxEU.org, 29 April.
Bethune, Zachary, and Anton Korinek (2020), “COVID-19 infection externalities : Herd immunity versus containment strategies”, VoxEU.org, 3 May.
Chapelle, Guillaume (2020), “The medium term impact of nonpharmaceutical interventions”, Covid Economics, Vetted and Real-Time Papers 18.
Correia, Sergio, Stephan Luck and Emil Verner (2020), “Pandemics depress the economy, public health interventions do not: Evidence from the 1918 flu”, SSRN 3561560.
Dahl, Christian Møller, Casper Worm Hansen and Peter Sandholt Jensen (2020), “The 1918 epidemic and a V-shaped recession: Evidence from municipal income data”, Covid Economics: Vetted and Real-Time Papers 6.
Lilley, Andrew, Matthew Lillley and Gianluca Rinaldi. (2020), “Public health interventions and economic growth: Revisiting the 1918 flu evidence”, working paper.
Markel, Howard, Harvey B Lipman, J Alexander Navarro, Alexandra Sloan, Joseph R Michalsen, Alexandra Minna Stern and Martin S Cetron (2007), “Nonpharmaceutical interventions implemented by US cities during the 1918-1919 influenza pandemic”, JAMA 298(6): 644–54.
Toda, Alexis Akira (2020a), “Early draconian social distancing may be suboptimal for fighting the COVID-19 epidemic”, VoxEU.org, 21 April.
Toda, Alexis Akira (2020b), “Susceptible-infected-recovered (SIR) dynamics of COVID-19 and economic impact”, Covid Economics, Vetted and Real-Time Papers 1.