Protecting people now, helping the economy rebound later

Jason Furman 31 May 2020



The coronavirus is a shock to the economy the likes of which none of us has ever seen. A hurricane hitting every place in the United States, Europe, and much of the rest of the world simultaneously. And hitting every day for weeks, months, or perhaps even a year. It is partly a supply shock as workers can no longer work and supply chains get severed. It is also partly a demand shock as people will cut back their demand, not just for restaurants and travel but likely much else throughout the economy given the extreme nervousness about their economic situation. The public health measures to flatten the curve, delaying and spreading out the extent of the virus, will necessarily and appropriately impose large economic costs. The job of economic policy is, to the greatest extent possible, to protect people from those costs now and help ensure the economy is in a position to rebound quickly when the health threat is contained. Doing this will require a multifaceted and ambitious policy response.

This column is taken from the VoxEU eBook Mitigating the COVID Economic Crisis: Act Fast and Do Whatever It Takes. Download the eBook here.

Human suffering in the immediate future

The human toll will be enormous. Most profound and consequential will be the toll on health – the illnesses and deaths from COVID-19 and also the collateral damage to others from an overloaded hospital system. Less profound, but even more widespread, will be the economic toll on workers who lose their jobs or see dramatic reductions in their incomes, either directly because they work in an industry that is shut down as a result of the pandemic or indirectly because they work in anything in the economy which is seeing less demand as a result of all of the income loss and dramatically worsened outlook for the future.

Prolonged reductions in economic activity

The economic problems could persist well after the pandemic is contained. It would be foolish just to think of what is happening as intertemporal substitution – people not going to restaurants today but then making up for it by going to even more restaurants next year. There are four interrelated reasons why the economic problems could be very persistent:

  • Labour market matching means that unemployment rates can rise sharply but not fall sharply. In the US financial crisis, for example, the unemployment rate took less than two and a half years to rise from 4.4% to 10% but, seven years to recover. The pattern is similar in previous recoveries and in other countries, and is the result of the labour market matching process in which it is hard to connect people with jobs.
  • Companies will go bankrupt throughout the economy, although the extent and magnitude will depend on the policy response. In some cases, these could be disorderly bankruptcies that separate firm-specific management capital, worker capital, and a nested set of arrangements and sever all of them simultaneously. Putting that back together will not be easy.
  • Financial institutions could come under tremendous strain and, absent an ambitious policy response, the economic crisis could turn into a financial crisis. Banks had substantial amounts of capital going into the crisis, including both the required levels and additional countercyclical capital buffers in many countries, but not the United States. Business credit lines have already been drawn, loans will be extended or not repaid, and funding could be increasingly difficult – all threatening to freeze up the financial system.
  • The global aftershocks could also be large. Some countries may succeed in containing the virus more quickly than others. Some countries may succeed in containing the economic and financial damage associated with the economy more quickly than others. A rolling set of epidemics in individual countries and economic crises in individual countries would create global limits on travel and continue to strain global supply chains.

Crafting a policy response during a pandemic

Policy faces three constraints during a pandemic. 

The first is uncertainty. Macroeconomic policy is operating under substantial uncertainty at the best of times, given the impossibility of definitive evidence from randomised control trials and the fact that the key structural parameters of the economy are likely fluctuating over time. Now it is operating under massive uncertainty given that we do not know the duration of the pandemic, the duration of the steps that are being taken to flatten the curve, the effect all of these will have on the economy, and how economic policies work in this situation.

The second is time. The change in economic activity has been larger and more abrupt than anything anyone has ever experienced on a global basis. The US housing bubble peaked in 2006, European financial institutions started to have problems in the summer of 2007, economic activity slowed over the course of 2007 in the United States, Bear Stearns needed to be rescued in March 2008, and Lehman Brothers collapsed in September 2008. At times events went quickly, but for the most part the economic situation unfolded slowly. In contrast, each day brings more news about the pandemic and more news about economic closures. Policies that are operational as quickly as possible are necessary.

The third constraint is capacity. During the financial crisis, government employees showed up at their jobs; now many are teleworking and likely all will be soon. Many are scared and distracted by the spread of the virus. Some will get sick and die, or if they do not will care for and grieve for others who do. All of this applies to the people developing policies in places like legislatures, finance ministries and central banks. It also applies to the people implementing the policies. At the best of times it is hard to implement administratively complex new policies. And these are not the best of times.

These constraints have six implications:

  • Better to do too much rather than too little. The current situation is one of radical uncertainty. And given this radical uncertainty, policy has to be based on a risk analysis of the cost of doing too little and the cost of doing too much. In this case the risk analysis is increasingly clear: the cost of doing too much is the time value of money, which right now is negative given the negative real interest rates. The cost of doing too little could potentially be enormous both in terms of immediate human suffering and a prolonged economic crisis that exceeds the one in the wake of the global financial crisis.
  • Use existing mechanisms as much as possible. Franklin D. Roosevelt engaged in “bold and persistent experimentation” in combatting the Great Depression. This process unfolded over a decade. We do not want to be combatting the economic fallout of the pandemic over the course of a decade. Increase funds under existing channels of assistance rather than creating new ones; repeat policies that have been tried and worked (at least administratively) in the past.
  • Invent new programmes where necessary. It will not be possible to use existing mechanisms for everything. The United States, for example, does not have paid leave or mandatory sick days so needs to invent and implement them in the middle of the pandemic. No country has sufficient mechanisms to deal with abrupt revenue stops in a large number of sectors across the economy.
  • Diversify and do not fear duplication or unintended ‘winners’ in the response. Given the uncertainty about the economic situation, the impact of policies, and the new policies that are being invented, it is worth diversifying the response. Many policies will have to be tried. Some will work; others will not. Many will be wasteful, giving money to people or businesses that do not need it – or even giving it to the same ones twice. The risk of duplication is much smaller than the risk of over-targeting that leaves many out.
  • Enlist the private sector as much as possible. The private sector will be operating under many of the same operational constraints as the government. But it has an existing infrastructure, can be nimble, and can form a diversification of the response. Direct government lending is hard, but loan guarantees can enlist the private sector to make the loans. The private sector will make the additional food and hospital equipment but will need financial incentives. Some will end up making a profit out of all of this, now is not the time to be squeamish.
  • Ensure the response is dynamic and persistent. The damage is uncertain. It may vary across places. It may last a long time. Policy needs to be ready to stay in place and even grow in the places and times it is needed. The more that policies can have triggers to automatically continue and expand in places and times they are needed, the better.

What a policy response should look like

The exact policy response will vary from country to country but here are some ways to operationalise the principles above:

  • Spare no expense on health. Testing, hospital systems, antiviral and vaccine research and anything else that is needed should be funded.
  • Targeted assistance using existing programmes. In the United States this means expanding eligibility for unemployment insurance (although that is hard because eligibility rules vary across the 50 states and the District of Columbia), increasing the amount of benefits ($50 per week added), and expanding other programmes for the most vulnerable, like the Supplemental Nutrition Assistance Program (SNAP). In addition, increasing federal funding for states is critical and one of the best mechanisms for this. The United States is increasing the share of Medicaid paid by the federal government.
  • Cash payments to households. Many people will be affected in many ways, with lost jobs, furloughs, lost savings, lost gig work, greatly reduced employment, and the like. Many will fall outside the targeted programmes that already exist or will be set up. Italy and France are delaying bill payments, but those bills will eventually have to be paid. Cash payments to households are a very efficient way to make sure broad-based help is available. This can help cushion the blow in the short run and put people in a better position to spend, driving economic recovery, after the virus passes. In the United States this should be at least $1,000 per adult and $500 per child – and importantly should continue as long as the unemployment rate is over 5.5%.
  • Assistance to businesses. This will require the most creativity. Large-scale lending programmes will be an important part of the response, including partial or full government guarantees of loans made by banks to businesses in order to keep them able to employ people and out of bankruptcy so as to be in a position to resume activity after the pandemic ends – something Germany, among others, has launched. New procedures are needed to enable large-scale workouts and avoid costly bankruptcies and liquidations without relying heavily on government administration. Whether the government should directly cover a large fraction of payroll, as in Denmark, is worth seriously considering. Finally, ensuring banks can extend and pretend on loans while extending new loans will require not just temporary regulatory changes but also backstops for the financial system.


Too often in January, February and the first half of March, policymakers were days or even weeks behind where they needed to be. For public health the consequence of this delay has been enormous; when dealing with what is initially an exponential process, a delay of a few days in implementing social distancing can have a large impact on the trajectory of the virus.

Policymakers are increasingly grasping the gravity of the situation and are implementing ambitious measures to flatten the curve of the virus, help protect people now, and help the economy rebound in the future. Much, much, much more will be needed – and I fully expect that much, much, much more will indeed be done.



Topics:  Covid-19 Macroeconomic policy

Professor of the Practice of Economic Policy, Harvard Kennedy School


CEPR Policy Research