Contagion: Bank Runs and COVID-19

“[I]f there is a single dominant lesson from 1918, it’s that governments need to tell the truth in a crisis...Those in authority must retain the public’s trust. The way to do that is to distort nothing, to put the best face on nothing, to try to manipulate no one.” John M. Barry, “The Great Influenza: The Epic Story of the Deadliest Plague in History. (Cited by @michikokakutani, Twitter, February 28)

There are currently more than 85,000 confirmed cases of COVID-19 in at least 60 countries. The contagion rate could be double that of the common flu, with the fatality rate as much as 20 times higher. But, these estimates—the rate of transmission, the frequency with which people exhibit symptoms, and the consequence of becoming ill—are all extremely uncertain. In addition, since it is a new virus, we have neither tested therapies nor vaccines.

So we know very little about this pathogen, except that everyone is worried. And, with the number of cases rising each day, intensifying concerns probably will lead many people to behave in ways that undermine economic activity. They will shy away from places where the virus can be transmitted. That means avoiding mass transit, schools, and workplaces.

Moreover, many people will stay away until they are confident that the disease is manageable. That confidence probably requires an effective treatment, a very low likelihood of infection, or both. Not surprisingly, many observers are reducing their projections for economic growth this year, while financial market participants anticipate easier monetary policy to cushion the shock.

The challenge of re-establishing public confidence that it is safe to venture out bears striking similarity to the one that authorities face in stemming a bank run. Our ability to identify and quarantine people infected with COVID-19 is analogous to our ability to recognize and isolate a bank bordering on insolvency.

Banks are like black boxes: outside observers know little about the value of their assets, especially in the aftermath of a shock, like a broad-based plunge in asset prices. As a result, bad news can lead depositors to question the solvency of their bank.

Furthermore, banks are vulnerable: the sequential process of redeeming deposits at face value creates a first-mover advantage: those who get to the bank first get paid in full, while those who are patient (or just slow) may receive nothing. This leads to a run.

What is more, like viral illnesses, bank runs are contagious. The news about a run on a specific bank alerts everyone to the fact that there may be other “lemons” among the universe of banks, turning a run in to a panic. Put differently, when people have insufficient information, shocks can cause them to behave in ways that amplify rather than dampen disturbances. Even if everyone believes that most banks are solvent, uncertainty about this bank or that bank can be enough to motivate a run.

These similarities suggest that the means we use to control bank runs also may be useful in managing the economic consequences of an emerging pandemic like COVID-19.

By lending against good collateral to solvent banks, a central bank can easily manage a liquidity-driven run. But if banks’ solvency is in question, then the problem shifts to one where authorities need to credibly demonstrate the health of the banks. Amid frozen markets and fire sales, how can they do that?

In our experience, the most effective mechanism to arrest financial contagion driven by solvency concerns is an extraordinary disclosure mechanism. Stress tests that aim to reveal banks’ true condition are the most powerful such tool. In late 2008, doubts about the capital adequacy of the largest U.S. intermediaries made potential investors, creditors, and customers wary of doing business with them, leading to a virtual collapse of unsecured finance. The May 2009 publication of stress test results for the 19 largest U.S. banks constituted a key part of the remedy.

Why did the U.S. stress tests restore confidence? One reason is that they were serious: a bank that passed the test could still lend to healthy borrowers despite a deep recession. But people also had to believe that the disclosure was truthful. Wouldn’t policymakers have an incentive to declare all banks healthy, even if some were not?

The key to U.S. authorities’ credibility in the midst of the financial crisis was that, even after making large capital injections in late 2008, they still had the means to bail out a failing institution. As a result, investors accepted the news that the stress-tested banks “only” needed to add $75 billion in equity funding, allowing newly confident private markets to re-capitalize them for the first time since Lehman’s failure.

To limit the economic fallout from a pandemic, the requirement of thorough and credible disclosure is the same. Even if people believe that almost everyone is healthy, there is an incentive to stay away from places where you may encounter someone carrying the illness. Daily news of transmission in dozens of countries leads to the obvious conclusion that infection can happen anywhere. And, just as it is costly to observe the health of a bank in a crisis, in a pandemic it is difficult or impossible to observe whether someone sitting next to you is carrying (and spreading) the disease. Every cough and sniffle trigger fear.

The lesson from the 2009 stress test is that sound science and public health policy are critical to limiting economically destabilizing behavior. For people to regain the confidence needed to go about their normal daily business, governments will need to demonstrate some combination of: (a) credible testing to demonstrate the population is nearly virus free; (b) the effective quarantine of those stricken; and (c) advances in treatment that limit the pathogen’s impact.

Success requires that people view authorities as extraordinarily trustworthy. This means providing detailed, up-to-date information on the spread of the illness, its severity and the methods available for treatment and control. As John Barry argues (see the opening quote), they have to stick to the facts, and shun politics entirely. Any attempt to color the facts weakens the credibility of the announcements and delays the point at which confidence returns.

Note: We wrote this for the forthcoming CEPR instant eBook, “Economics in the time of COVID-19,” edited by Richard Baldwin and Beatrice Weder de Mauro.