1) How do you test scientifically that they actually prevent the spread of COVID? Like, how do you make an actual scientific study about it, with a meaningful p-value? Mechanicistic evidence is not meaningful here, because whether or not lockdowns work is a matter of how people actually respond to it (For example, many people responded to lockdowns by rushing into grocery stores and buying unreasonable quantities of everyday products, arguably contributing to the spread of COVID.). To me it seems like you can make essentially two types of study about it:
a) Do non-controlled experiments, like the famous Delaware study. And any conclusion following from them is, of course, a post hoc ergo propter hoc fallacy. I think it should not be controversial that the Delaware study is seriously flawed. First, they imply that the 88% reduction in deaths between April and July in Delaware is due to the lockdown. However, during the same time period, the deaths with COVID also drastically declined in Sweden.
But Sweden had no lockdown at that time. That strongly suggests the drastic reduction in mortality during that time period in Delaware is mostly not due to lockdown, but due to some other factor. Maybe it is Vitamin D, since Vitamin D deficiency (which appears to drastically increase the incidence and severity of COVID) is lower in summer. Second reason why that Delaware study is flawed is that, as they say, the mortality in Delaware peaked one week after the lockdown began. Though they cite that as evidence the mortality indeed started decreasing because of the lockdown, I'd argue it is evidence of the opposite. In order to die from COVID, you need to have it for around three weeks. So, the reduction in the number of infections had to start around two weeks before the lockdown.
b) Do cross-coutry comparisons, to see if there is a correlation between the severety of the lockdown and COVID-related deaths. And, of course, the only way to do that is to lump good and policies together, like the Economic Freedom Index is doing. Obviously, any such study is next to meaningless.
It seems that the claim that lockdowns work against COVID is one of those claims that, while they seem testable at first, are not actually meaningfully testable. Like the claim that circumcision decreases sexual pleasure: how exactly would you do a scientifically valid study about that, one that controls for the placebo effect? If we should have a government in the first place, it should be a government that bases its policies on science.
2) Lockdowns probably have side-effects in the form of damaging mental health (increased suicide, especially among the young) and, caused by that, economic damages. The constant fear-mongering about COVID is, to young people, certainly more harmful than COVID itself is. And it is far more wrong if a policy results in a death of a young and healthy person, who would otherwise live for another 100 years (or by whatever amount of time the human lifetime increases this century), than if it causes a death of somebody who would otherwise live just a few more months. And if people are depressed, the economy cannot work, causing even more deaths. To be fair, it is hard to tell how much effect lockdowns themselves have on mental health and the economy. Obviously, economic damages happen even in countries without a lockdown, as has happened in Sweden. On the other hand, economies of countries are interconnected, and, if the world's economy suffers, Swedish economy will suffer because of that. I think it is hard to deny lockdowns have played a significant negative effect on the economy, even if we do not know exactly how much (compared to just the fear-mongering). The predictions that the economy will return back to normal a few weeks or months after the lockdowns have been implemented world-wide have, as far as I understand it, proven spectacularly wrong.
3) If the studies showing a link between Vitamin D deficiency and COVID mortality are correct, lockdowns are probably counter-productive. Now, this is, as far as I understand it, a very complicated topic. Somewhat similar to the question of whether low-carbohydrate low-protein diets somehow help against epilepsy: many studies show they do, but there is a complete lack of scientific explanation for how they might. I think the most honest position to take here is not to bet on the either side: do not assume that Vitamin D protects against COVID, but do not assume it does not help either. However, implementing lockdowns is basically betting on the association between COVID and Vitamin D deficiency not being true, which does not seem very reasonable.
4) In just about every country, lockdowns are unconstitutional. If we allow the governments to break the law now, they will have more justification for doing so in the future. Government overreach is a serious problem, which can significantly affect the quality of life of those who are young today.
5) Even if we take for granted that properly implemented lockdowns work, there is little relation between what a proper lockdown would be, and what the governments are actually doing. The US government, led by Andrew Cuomo, was putting COVID patients into nursing homes not to overwhelm the hospitals, going wildly against science and undoubtedly leading to even more deaths. The Croatian government organized massive commemorations of the events from the Yugoslav Wars (the commemoration of Vukovar Massacre was attended by around 30000 people) and World War 2, in the middle of the pandemic, which led to a spike of COVID cases.
