It's poor coding, as it can confuse new programmers.
The function, assuming it is passing by value, like in the example above (as opposed to passing by reference) takes a copy of y, decrements it, and assigns it to b. It does not alter the argument passed to the function when it was called.