> If data does show a correlation, don't hosts have good reason to discriminate, just as insurers do? Well, they have an incentive to discriminate, but not nec

> If data does show a correlation, don't hosts have good reason to discriminate, just as insurers do? Well, they have an incentive to discriminate, but not necessarily a morally good reason. Regarding insurance: Being middle-aged is something we all get a chance at. Being white or female is not. If a restaurant owner can show that blacks do not tip as well as whites, is it acceptable for him to always seat blacks at the back of the restaurant?