As given above: in a dogmatic systems department, recursion is to be avoided because its naive implementations use more machine resources than their equivalent loops; in a dogmatic theory department recursion is to be avoided because it is more difficult to prove properties of potentially unstructured recursions than of structured folds.
The dogmatic systems department wishes to replace recursion with something more concrete; the dogmatic theory department with something more abstract.
we can form a committee from both departments, and after enough iteration, they may agree to a compromise that annoys everyone equally: replacing recursion with recursion.