How does this code has 2 power N complexity as opposed to N*2powerN?

def allFib(n):
 for i in range(n):
   print(str(i) + ":, "+ str(fib(i))

def fib(n):
  if n<=0:
    return 0 
  elif n == 1:
    return 1 
  return fib(n-1) + fib(n-2)

Shouldn't fib being called N time be accounted here?