Covid-19 India prize (post with fixed links)

It goes to the COVIN Working Group for their paper “Adaptive control of COVID: Local, gradual, and trigger-based exit from lockdown in India.”

As India ends its lockdown, the team, led by Anup Malani, has developed a strategy to inform state policy using what is called an adaptive control strategy.  This adaptive control strategy has three parts.  First, introduction of activity should be done gradually.  States are still learning how people respond to policy and how COVID responds to behavior.  Small changes will allow states to avoid big mistakes.  Second, states should set and track epidemiological targets, such as reducing the reproductive rate below 1, and adjust social distancing every week or two to meet those targets.  Third, states should adopt different policies in different districts or city wards depending on the local conditions.

This project provides a path that allows states to contain epidemics in local areas and open up more of the economy.  Going forward the team plans to help address shocks such as recent flows of laborers out of cities and estimate how effective different social distancing policies are at reducing mobility and contact rates.

This project has 14 authors (Anup MalaniSatej SomanSam AsherClement ImbertVaidehi TandelAnish AgarwalAbdullah AlomarArnab Sarker, Devavrat ShahDennis Shen, Jonathan GruberStuti SachdevaDavid Kaiser, and Luis Bettencourt) across five institutions (University of Chicago Law School and Mansueto Institute, MIT Economics Department and Institute for Data Systems and SocietyIDFC InstituteJohn Hopkins University SAIS, and University of Warwick Economics Department). 

Draft of the full paper is here. And for the visualizations see their website https://www.adaptivecontrol.org

Congrats to all the authors of the paper and their institutions.  And here are links to the previous Emergent Ventures anti-Covid prize winners.

And I thank Shruti for her help with this.

Comments

With a cast of thousands, you will be amazed at the zoom spectacle of global action among economists being presented on the screen in living color.

It does not matter what you do it will not work because it is a highly infectious disease. 70% of the population WILL get it. Everything else is wishful thinking.

Respond

Add Comment

Respond

Add Comment

Sounds a lot like what they are (in theory) doing in the United States. What could possibly go wrong?

Bollywood, Hollywood - the main thing is that the audience will be entertained.

Respond

Add Comment

Respond

Add Comment

Tyler, would you please add the link to the draft paper. This is an excellent application for control theory.

Respond

Add Comment

They have the dimocrap impeachment hoax virus in India? Fake news.

Respond

Add Comment

We’ve all lost respect for the academics and healthcare “experts” due to this Covid lockdown. They never knew what they were doing from the start and still don’t.

They made hundreds of millions of people unnecessarily miserable and did great economic damage. At the same time they failed to protect those who should’ve been protected and actually made it worse for them. Thanks, Best University People.

'They never knew what they were doing from the start and still don’t.'

You are absolutely right - time to expose the germ theory of disease for the absolute catastrophe it is.

Respond

Add Comment

Respond

Add Comment

As India ends it's lockdown, what did it get out of this "premature imitation"? Will we know what they lost from this "premature imitation"? (Hattip Alex Tabbarok for the term!)

Respond

Add Comment

What about the Stalin prize? Who won? Maybe we should not support totalitarian regimes.

I, too, am shocked by those references ro Modi's regime.

Respond

Add Comment

Respond

Add Comment

But how good is the data?

I don't just mean the lack of testing, which varies considerably from region to region in the country. But fine, that's something we have to live with.

https://www.adaptivecontrol.org/

If you zoom into Maharashtra, you see that Pune has an Rt of 0. Wtf?

And Mumbai has an Rt of 1.13. Again, completely meaningless. According to the official statistics, cases are doubling every ~40 days, but deaths are doubling every ~13 days. This is just the official data from the local administrative body (BMC).

In Mumbai, the local body has continuously changed its 'rules' in order to make the data look "good". Another example: what counts as a patient 'recovered': a negative test is not required. It used to be 5 days without a fever and the patient is considered recovered and discharged. Now it's 3 days without a fever. I can go on and on.

When the entities responsible for generating the data have every incentive to use every trick in the book to generate exceptionally biased and misleading data, what's the point of patting ourselves in the back for analyzing such data in "clever ways"?

I am an economist by training, and I do get the desire to applaud such efforts. I don't mean to belittle them at all, but the situation is extremely frustrating. The reality on the ground is so different from what any such effort suggests. It's hard not to think its just academic intellectual masturbation. What's the point of such a "prize"? I bet Robin Hanson (and I'd have thought Tyler Cowen too) would have a field day theorizing the incentives that create such a perversity.

I'd think a *much* more interesting exercise would be to chronicle the methods employed by various public (and perhaps private) bodies around the country (and perhaps the world) in order to distort reality as effectively as possible. But that would require considerably more effort than this paper/project. And it would mostly require the researchers to reliably identify workers on the ground who are willing to speak up about what's happening in reality.

So the incentives to cook the books are great enough that the data is suspect at best, garbage at worst. GIGO.

What might we conclude more generally from this example, about other centrally planned endeavors, particularly managing the economy?

Respond

Add Comment

Respond

Add Comment

Respond

Add Comment