Monday, April 13, 2020

V, U or L? Three views

Will the recovery be V shaped, U shaped or L shaped?

The 1918 influenza stood in most economists' memories as the paradigm -- a short, small V shaped slowdown, despite massive mortality. But 1918 was different. People and public policy went about their business ignoring the death toll to a large extent.

François Velde writes about the 1918 Influenza.
Burns and Mitchell (1946, 109) found a recession of “exceptional brevity and moderate amplitude.” I confirm their judgment by examining a variety of high-frequency data. Industrial output fell sharply but rebounded within months. Retail seemed little affected and there is no evidence of increased business failures or stressed financial system... Interventions to hinder the contagion were brief (typically a month) 
Of course the fact that interventions were brief has a lot to do with the mildness of the recession. Still,  it was on this historical evidence that most economists thought in February that this might be the great vacation.
I then use high-frequency cross-sectional data to confirm the visible but brief impact of the epidemic and of the intervention measures (closings of certain businesses) that were adopted at the time. Banking data shows a financial sector functioning as it should (increasing loans). Conspicuously, there is no evidence of stressed balance sheets in the nonfinancial sector: business failures were on an uninterrupted downward trend, and cross-sectional data fails to find any effect of mortality....
Here is the picture of a short sharp V shaped 1918 -- followed by the deeper but equally V shaped 1921, not influenza related

And here is the absence of business failure, all before the CARES act, stimulus, massive Fed lending, and so forth. 

François constructs
an index of local business conditions from weekly qualitative reports and use it, along with measures of payments volumes, to examine if the speed with which economically costly interventions were put in place made a difference in economic outcomes. I find clear evidence that interventions changed the dynamics of the epidemic and affected economic activity by reducing the number of infected, though broader effects (through a reduction in demand or activity) proves elusive.
Things are not always the same
[Interventions] imposed mostly by cities but sometimes at the state level, took a wide variety of forms ranging from shutting down public gatherings and crowded places to staggering business hours, closing schools, imposing quarantines for infected people, requiring masks, etc. No intervention went as far as closing non-essential businesses, as have the lockdowns of the 2020 pandemic.
Velds goes on to make an important point. There are lots of pandemic macroeconomic models being built at breakneck speed right now. They should fit not only this one -- a massive recession -- but also fit the milder earlier ones. 


Looking across the US, 
more exposed areas experience a sharp and persistent decline in economic activity. The estimates imply that the pandemic reduced manufacturing output by 18%. ...
This differs a bit with Velde. In part they emphasize the longer lasting cross sectional effects, but I'm not sure if that totally accounts for the difference. 
We find that cities that intervened earlier and more aggressively do not perform worse and, if anything, grow faster after the pandemic is over. Our findings thus indicate that NPIs not only lower mortality; they may also mitigate the adverse economic consequences of a pandemic.
This is surely true. The way to nip exponential growth in the bud is to stop it early.


Scott Baker, Nick Bloom, Steven J. Davis and Stephen J. Terry COVID-Induced Economic Uncertainty are much more pessimistic. They use their measures of economic uncertainty, derived from newspaper reports and (panicked) financial markets, fed through a model that relates uncertainty to economic outcomes, to forecast a long, deep recession.

Here too, the past may or may not be a guide to the future. The hope is for a swift all-clear, people return to the jobs they had and businesses reopen, without the agonizingly slow dynamics of a usual recovery. Well, that's the hope.


Robert J. Barro, José F. Ursua, Joanna Weng "The Coronavirus and the Great Influenza Epidemic" look across countries,
Regressions with annual information on flu deaths 1918-1920 and war deaths during WWI imply flu-generated economic declines for GDP and consumption in the typical country of 6 and 8 percent, respectively
The death rates were enormous relative to now
150 million deaths worldwide when applied to the world’s population of around 7.5 billion in 2020. 
and they note that we are choosing much more economic dislocation to save lives.
However, extreme mitigation efforts—such as widespread cancellations of travel, meetings, and major events—will themselves contribute to the depressed economic activity.
I'm not sure what implications to take for our question.



  1. I think people aren't taking into account the "flattening of the curve". The idea is to prevent the healthcare system from being overwhelmed but does it change the AUC? Same number of cases spread over a longer period of time? The more you flatten the curve the longer the pandemic lasts.

    Frankly I think modeling is futile right now since there are insufficient data. Fauci pretty much said as much, and so far all the models have been wrong. Might as well buy a Magic 8 Ball.

    Every day new information comes out. South Korea has seen "reinfections " but are these true reinfections, meaning having C19 doesn't confer immunity, or is is reactivation, the way varicella-zoster virus starts as a childhood chickenpox infection but the virus lays dormant for decades and re-emerges as shingles?

    And the info keeps changing. Initially it was estimated that 18% of cases were asymptomatic. Now it's 80%.

    I keep hearing about a vaccine in 12-18 months. It's been almost 40 years without an HIV vaccine.

    The plural of "anecdote" is not "data".

  2. Kashkari says Fed's coronavirus response is "as aggressive as possible" By Melissa Quinn CBS News April 12, 2020

    "It goes back to the progression of the virus," he said. "If we're going to have economic distress until we have a vaccine, then it's going to be up to Congress to keep coming back to provide support to the American people."

    Kashkari said the nation should be looking at an 18-month strategy for the economy and health care system based on other countries further along in their recovery from the coronavirus outbreak, and acknowledged that while the U.S. economy cannot be shut down for that long, officials need to identify ways for those who are not infected with the coronavirus to return safely to work.

    "We need to find ways of getting the people who are healthy, who are at lower risk back to work and then providing the assistance to those who are most at risk, who are going to need to be quarantined or isolated for the foreseeable future," he said. "That's a real challenge for all of us."

    President Trump has signaled an eagerness to reopen the U.S. economy, the strength of which was central to his reelection campaign, and discussions have started shifting toward what a return to normalcy may look like. Public health officials, however, have urged Americans not to let up on the aggressive social distancing measures taken to curb the spread of the coronavirus, which they acknowledge are having positive effects.

    Kashkari batted down expectations that the economy will sharply and quickly rebound, saying that's unlikely unless a new treatment is developed in the coming months.

    "This could be a long, hard road that we have ahead of us until we get to either an effective therapy or a vaccine," he said. "It's hard for me to see a V-shaped recovery under that scenario."

  3. The importance of data as it relates to this whole mess:

  4. Correia, Luck and Verner don't control for quadratic effects, which is a huge oversight. Should we have a 5 year quarentine for two fold gdp growth? The data says yes

  5. I still think the proper macro analog, not for the virus per se but the policy responses to it, is to invasion and occupation, as of France 1914-18 -->


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.