I'm not really debating that ARR for hospitalisation (or worse) may be only a few percent on average, but the article you linked to doesn't help. If I understan

I'm not really debating that ARR for hospitalisation (or worse) may be only a few percent on average, but the article you linked to doesn't help. If I understand it correctly they say: more people got infected (!) in the control group, but that's still a very low percentage of the entire group. Calculating the ARR in that case is only 'absolute' for the duration of the experiment right? That didn't age well at all, especially with newer variants, it's assumed most will eventually get infected, increasing the ARR. Then again de RRR for infection seems to drop by itself, which works the other way. So the conclusion of this paper is debatable at the least, and regardless it doesn't even cover what we were discussing, ARR for hospitalisation or worse. I really was trying to answer your question regarding pushing mandates. Again I'm not saying I'm for them (here in the Netherlands we're staying away from mandates, for now at least). Individual efficacy (vs risk) is not the ONLY argument for mandates (actually I'd find that a less strong reason by itself even with a higher ARR), my argument would be the benefit for society as a whole. Sure, we should be able to make it through also without 100% vaccination rate, that's probably why mandates are less clearly on the table in places where it's high without mandates already.