The starting point for using reduction theorems to prove fixed point theorems is the following result which seems to have become part of the folklore in this area:
