The survey also smooths over the bargaining problem. Woman wants 2 kids, husband wants 3. They eventually agree to 2. You put them both in a survey and you get desired 2.5, actual 2. Is it a societal problem the woman “won?” Idk. Just a dumb example but I suspect it might be more complicated.
Hard for me to imagine a scenario where desired fertility from surveys <= actual fertility without extreme poverty, rules against birth control, or one side of a relationship consistently being a subject of the other
Happened during the immediate postwar baby boom if you look at the Gallup time series but since I think 1960 it’s been desired < actual
Sorry I screwed up the inequality lol but you know what I mean
(Usually all three)