Opening Statement Paul Cockshott
In defence of the COVID lockdown
There is no doubt that the best option for a country is to use border controls, tracking tracing an isolation to prevent COVID-19 even starting widespread community transmission. But if that fails, Lockdowns are needed. I have been challenged to debate by an opponent of lockdowns for the Covid-19 crisis. I will be arguing for a policy in which lockdowns are a key element. It will cover the following points:
First I will show that COVID-19 is a serious danger.
Next I will show that the herd immunity policy that the Tories initially advocated would lead to catastrophic numbers of dead in any country that followed it.
I will then review the evidence for the success of lockdowns.
Finally I will outline what should be the long term policy of any government trying to deal with the crisis.
How fatal is COVID-19
As of the end of March the best estimate of fatality rates was around 1%. An article in The Lancet reported :
A unique situation has arisen for quite an accurate estimate of the CFR of COVID-19. Among individuals onboard the Diamond Princess cruise ship, data on the denominator are fairly robust. The outbreak of COVID-19 led passengers to be quarantined between Jan 20, and Feb 29, 2020. This scenario provided a population living in a defined territory without most other confounders, such as imported cases, defaulters of screening, or lack of testing capability. 3711 passengers and crew were on board, of whom 705 became sick and tested positive for COVID-19 and seven died, giving a CFR of 0·99%.
This figure of 1% was based on 7 deaths, so was subject to considerable random noise. It could have been too low or too high. It is necessary to have larger samples to be more confident.
More data comes from the epidemics in Italy and New York. For instance the province of Bergamo has had a population death rate of 0.58%, [Modi et al May 2020] . The caveat to this lower bound is that we lack data on what percentage of the Bergamo population were infected.
The infection death rate has to be higher than the population death rate, only part of the population will have been infected. Modi et al estimate that the lower bound on the infection death rate from the Lombardy data is 0.61%, the actual rate is likely to be higher.
More accurate data has become available from a serology survey in New York. A summary is given on the Worldometer site.
12.3% of the population in the state had COVID-19 antibodies as of May 1, 2020. This percentage would indicate that 1,671,351 people infected with SARS-CoV-2 as of May 1. Total of actual deaths: 13,156 confirmed + 5,126 probable + 5,148 additional excess deaths calculated by CDC = 23,430 actual COVID-19 deaths as of May 1, 2020 in New York City. The CDC added in deaths of those who died outside of hospital settings.
Allow for the fact that the disease takes some 3 weeks to kill people, and allow 1 week for antibodies to appear, this implies we need to look for total deaths in New York on or around the 14 May, to estimate fatality from those who had been infected by May 1. Below I present a calculation assuming the same ratios between hospital and non hospital deaths.
So this is somewhat higher and is still conservative, since not all people infected on the 1 May who will not recover will have died by the 13th May.
We have a range of estimates of the death rate going from an absolute lower limit of 0.61% for Italy to 1.68 for New York, with the cruise line figure of 1% sitting in the middle.
The New York data is arguably more reliable than the Bergamo data, based as it is on larger samples and being backed by random sampling of the population for antibodies.