Your argument as it stands would explain why people with college degrees smoke a lot less than people without them. But that's not what the "place effects" is in the article. "Place effects" is the fact that, if we just look at non-college-graduates, the ones who live in rural areas smoke more and have lower life expectancy than the ones who live in urban areas.
The latter effect, I think, can be explained by an argument that's similar to yours: even for non-college graduates, it's a lot more inconvenient to be a smoker in urban areas than in rural areas. You're much more likely to find smoking banned inside the places you go, and to face social disapproval if you try to smoke outdoors in public spaces.
The latter effect, I think, can be explained by an argument that's similar to yours: even for non-college graduates, it's a lot more inconvenient to be a smoker in urban areas than in rural areas. You're much more likely to find smoking banned inside the places you go, and to face social disapproval if you try to smoke outdoors in public spaces.