I mean, if we are talking about teaching elementary school, then you don't need to be world class researcher to get them ready for next level. You need to like math and, imo, you need to like problem solving, you need to understand math instead of having it memorize etc.
However, you definitely don't need to be actual researcher. I would even argue that it will be more beneficial to study psychology, child development, child behavior etc then spending time doing serious math research.
I'm not talking about elementary school, I'm talking about masters or bachelor's level advanced mathematics that has nonetheless been established for decades or centuries.
All I'm saying is that teaching and researching are different skill sets. One does not prepare you for the other except indirectly, nor is one a prerequisite for the other. One could just as easily say some pedagogy is required to communicate original research. It's true to an extent, but you don't need to be a excellent teacher to be an excellent researcher, or vice versa.
However, you definitely don't need to be actual researcher. I would even argue that it will be more beneficial to study psychology, child development, child behavior etc then spending time doing serious math research.