The top five nations testing for COVID by 9 April 2020 were these:
With total tests per million at left, all 5 of them had more than 20,000 tests per million population (more than 2%). Because of testing such a large share of the population before mid-April, the total test positivity rates for them should be reasonably accurate regarding nationwide infection prevalence:
With total tests and total positivity of those tests, you can impute the implied total infection prevalence from population size, and then compare that total count of all COVID infections to the total count of all COVID deaths 17 days later (the usual time it takes for COVID to kill you):
As can be seen in cell J18, that leads to a median COVID IFR of 0.1% — which is similar to seasonal flu.
And imagine how much lower it would have been if no pandemic declared and usual treatments for all the respiratory illnesses had been carried out?