One way to find out is to look at the overall death rates (of all causes) in an area and see how it evolved. NY started doing that and found there's a discrepancy. One explanation among others is that some people who've caught it and died from it never got a positive test (either no test, or false negative).
That seems to point to the opposite happening: a higher mortality that is still not fully understood for lack of accurate data.
I do overall agree with you, but Norway has lower than usual number of deaths this year, so not sure what to do with that. I guess if you just compare the deaths to average you are still under reporting because of the stay at home orders it is sort of expected that less people die.
Don't forget the secondary effects of shutting down entire countries: depression, supply chain failures, sudden poverty, degradation of infrastructure, and increasingly lack of exercise and unhealthy eating as this drags on.
I suspect that will account for most of the differential, and possibly even be the more important number. (because many countries diagnose deaths with covid as deaths by covid)
Later on it may be a major factor, but right now it is probably outweighed in the counts by less car accident deaths, lower transmission of other communicable respiratory diseases, and less work and leisure accidents.
The secondary effects are much worse. The virus is being used as an excuse to get more control of the people and planet by the psychopaths. They don't really care about us.
I don't know why something so easily disproved is being repeated so often. The worst-case of an economic shutdown is not worse than 2 million dead people.
Added bonus: the economic outcome of 2 million dead people is bad too, perhaps worse than a shutdown of (largely) healthy people. It's not an either/or, economic hardship right now is unavoidable. The choice is between getting a handle on the virus or not, and thousands dead instead of millions.
That seems to point to the opposite happening: a higher mortality that is still not fully understood for lack of accurate data.