1. What % of people have natural immunity to Covid? We get some information from closed environments where a large % of people were exposed. Here are some numbers from prisons, a meat plant, and call center. From this, seems that >80% are susceptible under the right conditions. 
2. Note that only the Korean call-center number is based on a scientific study. However, the call center was shut down early and so it's likely that >55% would have been infected if it had stayed open. I think >65% is plausible for prisons, but not sure about Marion result.
3. Some hospital wards and care homes also had large %es infected. But older people are more susceptible. Prisons, meat plants and call centers cover a wide range of ages (18-70). In the table I adjust for the false-negative rate of PCR testing, which is ~20-30% for mass testing.
4. It might be that people have some immunity but not enough to stop infection in a prison with no distancing and poor hygiene. The call center seems closer to normal life for most people.
The table was generated using my spreadsheet:
docs.google.com/spreadsheets/d…
docs.google.com/spreadsheets/d…
NB: household secondary attack rate is only 20-30%. What's different about these environments where >50% got infected? Each individual would be exposed to multiple distinct infectious people vs just one. And there's a high chance of close contact with presymptomatic spreader.
I'd excited for people to add other examples. I found one worker dormitory in Singapore with what seemed to be >20% infected (i.e. 2.6k/14k). There are HCW antibody studies with 26-30% in UK and Spain.
docs.google.com/spreadsheets/d…
docs.google.com/spreadsheets/d…
