I have offered an earlier version of my article ‘The excess mortality is strongly underestimated’ to ESB, a renowned economic statistical magazine in the Netherlands. The CBS was asked for a reaction.
According to the CBS I used the wrong number for comparing mortality. The CBS had a point.
I used the expected mortality for 2021 from the CBS estimate from 2020. I compared this estimate with my adjustment of 8K. See image below.
I realized that should have used a pre-pandemic estimate from 2019 as a base. The estimate for 2020 was adjusted from the estimate from 2019. See table below.
This is the correct way to make the necessary calculations. I changed my article accordingly. Note that the net effect in the (rounded) difference is minimal.
I presume one disturbing factor, covid. Hence I adjust the base estimate from 2019 with 8K.
The CBS claims that my correction was included in their estimate. The CBS calculates their estimate from the population size per 1-1-2021. According to the CBS this includes the fact that in 2020 more people died from covid. And that means that the covid-effect is included. With covid as only disturbing factor the estimate from the CBS should also have come more or less 8.000 lower. But the CBS estimate is 8.000 – 345 = 7.655 higher. Hence there must be a second disturbing factor that explains the difference of 7.655. The CBS doesn’t explain this.
Or otherwise: The CBS claims that without covid their estimate would be 8.000 higher, because the 8.000 is included in their estimate. Then the estimate from the CBS would be 154.887 + 8.000 = 163.887. This is 7.655 higher than the estimate of 155.232 from 2019. Without major disturbances this is no realistic increase.
The following might be of importance.
The CBS uses a mortality probability per age group that is based on historical numbers. However, covid hits the weak the hardest. Hence covid might have struck hard on the weakest per age group. This can affect the distribution in the age group and thus the percentage that is needed to calculate the expected mortality in the group. With this the expected mortality may be overestimated and hence the excess mortality is then underestimated.
See for this my article Expected mortality adjusted for distribution in age groups . Note: I wrote this article later and it was not included in the comment from the CBS.
I think now my article on excess mortality is correct. So please read my article and decide for yourself.