BPEA | Spring 2021

Behavior and the dynamic of epidemics

Doctor giving a shot to a patient, both wearing surgical masks.
Editor's note:

The paper summarized here is part of the spring 2021 edition of the Brookings Papers on Economic Activity, the leading conference series and journal in economics for timely, cutting-edge research about real-world policy issues. Research findings are presented in a clear and accessible style to maximize their impact on economic understanding and policymaking. The editors are Brookings Nonresident Senior Fellow and Northwestern University Professor of Economics Janice Eberly and Brookings Nonresident Senior Fellow and Harvard University Professor of Economics James StockSee the spring 2021 BPEA event page to watch paper presentations and read summaries of all the papers from this edition.

Behavior importantly shaped the course of the COVID-19 pandemic, suggesting lessons for mitigating the impact of future infectious disease outbreaks, according to a paper presented at the Brookings Papers on Economic Activity conference on March 25.

In Behavior and the dynamics of epidemics, Andrew Atkeson of the University of California-Los Angeles uses a simple model to project U.S. deaths under various scenarios. The exercise suggests epidemiologists likely were correct early on when they forecast the pandemic “will not resolve until high proportions of the population have acquired immunity either through infection or vaccination,” he writes.

Without a vaccine or cure for the virus and without disease-control measures such as masking, social distancing, testing, and contact tracing, the model predicts deaths would have soared in the first six to nine months of the pandemic, cumulating to 1.5 million.

Adding disease-control measures to the model slows the spread of the virus. However, behavior offsets their effect. As deaths and cases decline, authorities relax economically costly measures such as business lockdowns, and “pandemic fatigue” causes many people to become less cautious. That produces a more drawn-out pattern with waves of deaths, similar to what has actually occurred. Long-run deaths in this scenario, absent a vaccine or cure, total 1.27 million over two-and-a-half years.

Factoring in the introduction in January 2021 of vaccines that protect half of Americans by June 1 reduces the death toll to 672,000. Effectively, disease-control measures, although periodically relaxed, buy time for the development of vaccines. Another simulation, which assumes that disease-control measures were imposed May 1, 2020 and not relaxed until vaccines are distributed, produces a death toll of only 292,000. (Actual U.S. deaths from COVID-19 through late-March total more than 540,000.) This simulation shows what could have been achieved by implementing more aggressive testing, tracing, and isolation while waiting for deployment of the vaccine.

Actual and predicted COVID deaths under carious models.

Atkeson concludes that, to better manage future pandemics, authorities should improve infectious disease surveillance worldwide, invest in new models to accelerate the development and distribution of vaccines and cures, and stand ready to quickly deploy measures that slow disease spread with the least cost to the economy.

“Public efforts at disease control can save a lot of lives over the long run by controlling disease while we wait for a vaccine or a cure,” Atkeson tells The Brookings Institution. “We have a tremendous opportunity to learn from international experience with COVID on how to do that without tanking the economy.”


David Skidmore authored the summary language for this paper. Becca Portman assisted with data visualization.


Atkeson, Andrew. 2021. “Behavior and the Dynamic of Epidemics.Brookings Papers on Economic Activity, Spring, 67-88.

Conflict of Interest Disclosure

The author did not receive financial support from any firm or person for this article or from any firm or person with a financial or political interest in this paper. They are currently not an officer, director, or board member of any organization with an interest in this paper.


The views expressed by the authors, discussants and conference participants in BPEA are strictly those of the authors, discussants and conference participants, and not of the Brookings Institution. As an independent think tank, the Brookings Institution does not take institutional positions on any issue. Please contact Haowen Chen ([email protected]) if you have any questions.