Thoughts on economics and liberty

URGENT: Herd immunity for COVID-19 could be as low as 15%


After reviewing comments here, I’ve tentatively concluded:

There are two reasons to explain the resistance: (a) social distancing leading to low viral load, so the virus doesn’t reach transmissible levels, and (b) innate immunity.

The key point, as Vernon Smith has pointed out, is that “the claim is that the 80% “resistance” accounts for discrepancies across different data“.

I read two other examples cited above in the comments which suggest that innate immunity is unlikely to be much greater than the 10-15% I had earlier estimated.

But even if innate immunity (and cross-reactivity) is in that range, it can have a dramatic impact. Stockholm’s case suggests it is already reaching herd immunity.
And if one adds the natural distancing that is part of life in many rural communities (with reduced viral loads in human interactions), it would drop further – which is probably why Sweden’s rural areas have also seen a rapid drop-off in cases.

The cases which oppose the studies that Knott has cited are:

At least 44 of 70 UT-Austin students who chartered a plane to Mexico for spring break tested positive (they flew back separately), as did 52 of 61 who attended a choir practice in Washington. [a comment in the FB post linked above]

This strongly suggests that viral load is a key variable. However, it also suggests that in communities where people don’t get cooped up together, the virus will never spread to the same extent. I’m assuming something on the lines of an inverse square relationship to distance applies: i.e. viral load is proportional to (1/distance^2).

Therefore there are two main sub-groups of the resistant population:

a) those who are innately immune (around 10-15% perhaps),

b) those who rarely, if ever, come into close proximity with others for an extended duration of time. Depending on the nation, the community, and occupation, this could be anywhere from 10-25% of the people. E.g. farmers and tradies are unlikely to come into close contact with virtually anyone outside their own household for an extended duration.

I want to note that Knott’s flu analysis is flawed. At least I couldn’t confirm it from here or here.

I’ve also studied this chart by Michael Levitt (which was linked by Knott here):

While virus transmission for an individual is random, it is a constant for society, a sum of constants for each segment (occupational, rural/urban, cultural).

The sum of this constant + innate immunity will inform herd immunity.

This virus likely won’t infect 60-65% of the world’s population because of insufficient viral load. Another 10-15% will beat it off with their innate immunity.

Only 20-30% will likely get the disease, of which only a very few will die.

This also means that in cities and countries that have seen significant deaths, the pandemic has effectively passed (some precautions can still help).

In other places, people can revert to normal with a few extra extra precautions.


This is the most pathbreaking piece of research I’ve come across so far on COVID-19, by Anne Marie Knott – Washington University’s Olin’s Robert and Barbara Frick Professor of Business.

This research was shared by the best (in my view) economist alive today – Vernon Smith (Nobelist, of course) on his FB page. He doesn’t share such stuff mindlessly.

This research suggests 80% innate immunity. If so, the IFR might well be high but the total effect is rather low. Herd immunity could be as low as 15%.

This analysis is very persuasive and we need to take it very seriously. This is very strongly supportive of Sunetra Gupta’s claims. I wish she had provided me with serious data and evidence – since the whole world needs evidence and logic – not just claims.

I’m going to share this widely – so everyone with some capacity can analyse this to identify any loopholes/ gaps. I will write about this on my TOI blog after I’ve had a chance to think about it, and read any critiques.


I have been exploring innate immunity recently and believe it is an area with a huge gap in our understanding:

Just earlier today I had raised my estimates from 10-15% to much higher (

To me this analysis by Knott is perhaps the most persuasive so far.


Anne Marie Knott, innate immunity, herd immunity


Sanjeev Sabhlok

View more posts from this author
2 thoughts on “URGENT: Herd immunity for COVID-19 could be as low as 15%
  1. Sanjeev Sabhlok

    Most likely! I’ve written to Dr Knott to find out if she knows anything about this.


Leave a Reply

Your email address will not be published. Required fields are marked *