I suspect that a lot of the "productivity drop" in general is a result of trying to measure productivity for remote workers but not office workers. If someone is in the office, managers commonly assume they're working, even if they're spending most of the day faffing around on the internet. If they're at home, you can't see what they're doing so you try to measure it by results. Which shows that they're only doing a couple hours of real work in a day, which gets called a productivity drop even though it was just the baseline.
I mean think about it. If people were less productive then to do the same work you'd need them to work more hours, not fewer, right?
And it's even worse if your metrics do something silly like measure by lines of code or time spent typing. If someone spends an hour walking in a park figuring out how to implement something and then five minutes when they get home to type it into a computer, it's obviously bad accounting to call that an hour of leisure and five minutes of working.