It might not entirely be incorrect either. Very few deaths are officially reported but the deaths which weren't going to be reported, were of the same subset o

It might not entirely be incorrect either. Very few deaths are officially reported but the deaths which weren't going to be reported, were of the same subset of people who never tested either, despite symptoms. So it would be safe to assume while both the infected and fatality numbers are way off, they are at least representative. Hence, I believe the IFR isn't incorrect.