Not just (relative) declines but an absolute decline: they're down to a small handful of cases; COVID-19 is basically gone from Sweden. Moreover, if the cause of that is widespread immunity, it means they don't need to concern themselves with second waves or additional lockdowns as other European nations do right now.

It doesn't prove that lockdowns were a mistake, but it means we should start considering that possibility. Sweden's still doing comparably or better than other nations that locked down heavily (e.g. the UK); even if you take the position that there's absolutely no doing cost/benefit when it comes to potential deaths (which is a radical position for seemingly everything besides COVID, but no matter), if at the end of the day the number of infections/deaths per capita turns out to be roughly the same anyway, the lockdowns were an enormous waste.

My personal favourite insight from all this, though: if you'd applied the Imperial College model (~500,000 dead in the UK without lockdowns, 2.2 million in the US) that initially inspired the lockdowns to Sweden, you would've gotten ~100,000 dead [1]. Instead we're left with ~5000. The virus turned out to be 20x less deadly than we thought. Maybe that calls for a change in tactics?

[1] https://www.medrxiv.org/content/10.1101/2020.04.11.20062133v...

There's a lot of variables in play. It may be 20x less deadly in a specific scenario, but change that scenario slightly and some other variable may explode in influence.

You have to look at the entirety of the response over time, which we can’t judge until this is over.

If everyone ends up about the same as Sweden, then lockdowns were a mistake.