This year, family or other gatherings have fewer possible negative externalities—because vaccines are broadly available. Let’s first assume a gathering composed exclusively of fully vaccinated people. Your risk of hospitalization is down a huge amount relative to last year. Your risk of contracting and passing along the coronavirus is down, too, though not as much, particularly because Omicron seems quite good at breaking through vaccine defenses against infection. In the event of a breakthrough, however, the chances are lower that the next person in the chain will end up with severe symptoms.

Now consider a mixed gathering that includes unvaccinated people. If you’re vaccinated, your personal risk remains low. The problem for society arises when an unvaccinated person contracts the virus and takes up hospital space—or passes the disease along to another person who then takes up hospital space. That’s the issue, the externality. But thinking about the puzzle this way can be clarifying. If the unvaccinated people are low-risk (because of age, or perhaps because of prior infection), the external cost is less significant. Being very explicit about these externalities will let us appropriately weigh them in our decisions.

Even in the first scenario, the risk of a bad outcome is not zero; gatherings will always carry some risk. But as a society, we should weigh negative externalities for the community against the benefit to individuals. Restricting—or encouraging the restriction of—family gatherings of fully vaccinated individuals is not rational, because that would impose large costs on individuals without a commensurate benefit to others. And, indeed, the CDC has not categorically advised against holiday reunions this year.