Read on Twitter
Interesting discussion these days about various versions of this figure (this one is taken from the NYTimes). 1/n
The typical reaction has been to point out that U.S. policies have been less successful in controlling the epidemic - noticing the resilience of the epidemic - possibly caused by early re-opening in various states. 2/n
The NYTimes again reports a comment by Jonathan Bernstein (a political scientist and Bloomberg Opinion columnist): “Government efforts to inform the public about the pandemic have been a colossal failure, ..." 3/n
https://www.nytimes.com/2020/06/17/briefing/dexamethasone-india-china-john-bolton-your-wednesday-briefing.html?login=email&auth=login-email
https://www.nytimes.com/2020/06/17/briefing/dexamethasone-india-china-john-bolton-your-wednesday-briefing.html?login=email&auth=login-email
While I have no desire to praise the US government policies - especially the Federal Government policies - and while I am extremely worried about the numbers in FL, AZ, TX, and other states, I think this reading of the figure is superficial and biased. Two reasons: 4/n
1. Comparisons across geographical units are very difficult (a bit of self-promotion: @andreamoro and I have written a paper about this - see below). 5/n
https://s18798.pcdn.co/albertobisin/wp-content/uploads/sites/16384/2020/06/spatialSIR-1.pdf
https://s18798.pcdn.co/albertobisin/wp-content/uploads/sites/16384/2020/06/spatialSIR-1.pdf
The figure above normalizes by population size But various other characteristics matter greatly - density, openness to outbreaks, and especially heterogeneity of these characteristics inside each geographical unit (EU v. US). These comparisons need much greater care. 6/n
2. Lockdowns and other Non Pharmaceutical Interventions generally (but not always -this is an important point in our paper) do not have an effect on the long-run - that is, the best you can do with this policies is to flatten the curve ... to avoid lines in front of ICU's. 7/n
Well, if this is the objective, the US did it better than the EU, according with the figure (see the lower peak in US?)
I am NOT saying that this is the case, necessarily, nor that - if it is the case - it is due to good policies in the US. 8/n
I am NOT saying that this is the case, necessarily, nor that - if it is the case - it is due to good policies in the US. 8/n
The fact that the epidemic started later in the US is obviously important. Agents' behavioral responses are fundamental too (and it is very hard to distinguish statistically the effects of behavioral responses from those of NPI's - again we discuss this in the paper). 9/n
What I am saying is that this reading (of the NYTimes and several other commentators) is superficial and biased. And, as an economist, I don't like that. 10/n
(And I got to self-promote my paper with @andreamoro)
