However it seemed last month that the correct thing to do was to close down. At least so that hospitals were not overwhelmed, causing extra deaths beyond the "natural" amount for this disease and others that people use hospitals to treat. In other words, a non-linear effect on death rates caused by the sudden spike.
Now we're between a rock and a hard place, because while taking a week or two off would be tough but OK for most businesses, taking 4 weeks or 8 weeks off is not simply 4x or 8x as bad. Thus, another, competing, non-linear effect.
We also have the problem that the pandemic is not synchronized, so any area that does manage to keep it down is going to have to keep the border closed towards people from places where it's still going on.
The only solution that seems like a good compromise is to have lots of testing so that people who know they are infected, even ones who don't feel ill, can be quarantined. Also of course generally keep old people away from too much contact with everyone else. But somehow even months after the virus was discovered, the number of people tested anywhere is not really high. I would have thought we'd ramp up production of test kits massively so that everyone who's been near someone positive is tested, plus everyone else is randomly tested.