There is also a great deal of randomness. I'd hestitate to draw conclusions by comparing countries, there is not enough data. If we could compare different lo

There is also a great deal of randomness. I'd hestitate to draw conclusions by comparing countries, there is not enough data. If we could compare different local areas that had different policies, we might be able to get somewhere. Even then, it's not easy. The locales who are happy to lockdown are probably also composed of people happy to wear a mask and happy to vaccinate. There is also an ugly reverse causality thing going on where lockdowns are a response to growing community spread, which makes it look like they don't work when maybe they actually do. It's not an easy question to answer empirically. I'm happy to do the finger in the air approach of assuming they work based on common sense. That doesn't mean that they should be implemented, which is a related but different question.