It all depends on how big that 10% is contributing to the GDP. Let's assume the least profitable employees are effected the most and thus have to absorb the largest unemployment hit. Than the percentage of the economy actually effected has to be a fraction of of the increase in unemployment.
> Let's assume the least profitable employees are effected the most and thus have to absorb the largest unemployment hit.
That assumption is wrong in this context. The employees most affected aren't the ones providing the least value under normal circumstances, they're the ones providing the least value during a quarantine (presumably because they're not working at all), which isn't strongly correlated with providing low value in general. For example, a hotel may have to do layoffs, but that doesn't mean hotels are unproductive/unprofitable in the general case when there isn't a quarantine.
And the likes of hotels may not be the quickest to rebound either. A lot of people are using their vacation time right now, and will have a ton of work to catch up on when they go back and not be in a position to take more time off to travel. Even after people go back to work, they're going to be wary about traveling to hang out in a huge venue with a bunch of strangers for a while.
So what you want is to create a lot of general demand so that former hotel workers can find work doing something else until demand for travel comes back, on top of the workers who had been doing those things already.
