The IFR is far below 2% for young people--a recent meta-analysis[1] got 0.7% over all ages. That study didn't break it down by age, but an older one that did[2]

The IFR is far below 2% for young people--a recent meta-analysis[1] got 0.7% over all ages. That study didn't break it down by age, but an older one that did[2] (and got a similar IFR over all ages) got 0.14% for <60 years old, mostly from the older end of that. There's moderate evidence for seasonality, from studies comparing countries with different climates and trying to control for other factors, trying to regress against actual weather over weeks, etc. That's empirically not enough to bring R < 1 in the summer in the USA in combination with whatever we've been doing, but a big effect might still be there--we might just have relaxed the social distancing by enough to more than offset. If that's the case, then we're potentially in for a terrible fall and winter (and places that have built up some degree of herd immunity--whether quasi-intentionally like Sweden or accidentally like NYC--could end up better off in the end). As to excess deaths: As far as I understand, most of those actually observed are probably unconfirmed coronavirus deaths, rather than unrelated deaths due to an overrun medical system or people afraid to seek care. I've seen speculation that the latter may occur, but no strong evidence (since even NYC barely occupied the Navy hospital ship or other temporary facilities, and since no excess deaths have been seen in areas with the lockdowns that would discourage people from seeking care but without big coronavirus prevalence). To be clear, I'm strongly pro-mask, and pro-social-distancing. (I do think stuff like factory and school closures comes at an unacceptable social cost for the benefit to public health, though that's been reversed most places.) I just want to make the argument with the right facts. 1. https://www.medrxiv.org/content/10.1101/2020.05.03.20089854v4.full.pdf https://www.medrxiv.org/content/10.1101/2020.05.03.20089854v... 2. https://www.thelancet.com/journals/laninf/article/PIIS1473-3099(20)30243-7/fulltext https://www.thelancet.com/journals/laninf/article/PIIS1473-3...