That is , both of the input arguments , as well as the output , are inferred to have type class Num. Had we explicitly declared a type like Int , the function would operate over a narrower range of values ( which is good or bad , depending on our needs )