Abstract: COVID-19 has severely disrupted the conduct of business around the globe. In jurisdictions that impose one or more ‘lockdowns’, multiple sectors of the real economy must endure prolonged periods of reduced trading or even total shutdowns. The associated revenue losses will push many businesses into bankruptcy. No public policy response can recover these losses. States can, however, act to reduce the amplification of the shock by the way in which they treat the cohort of newly bankrupt businesses. In jurisdictions where a well-functioning reorganisation procedure is capable of producing value-maximising outcomes in normal conditions, the temptation may be to subject this cohort to treatment by such procedures. This temptation should be resisted, not only because of the (significant) costs of these procedures, or because of concerns about institutional capacity to treat a high volume of cases, but also because such procedures are likely to be a poor ‘fit’ for the treatment of COVID-19 distress. In our view, the more attractive routes to relief are bail-ins (one-time orders to creditors or counterparties, or some class thereof, to forgive), bail-outs (offers to assume the debtor’s liabilities, or a class thereof), or some combination of the two. In this paper, we explain why a public policy response is necessary to mitigate the amplification of the shock caused by trading shutdowns, and compare treatment by the prevailing bankruptcy law with treatment by bail-ins or bail-outs along a range of dimensions. We conclude by tentatively suggesting some principles to help guide the choice between bail-ins and bail-outs, and the design of either form of intervention.