We consider a discrete-time two-dimensional process {(X 1,n , X 2,n )} on Z 2 with a background process {J n } on a finite set S 0 , where individual processes {X 1,n } and {X 2,n } are both skip free. We assume that the joint process {Y n } = {(X 1,n , X 2,n , J n )} is Markovian and that the transition probabilities of the two-dimensional process {(X 1,n , X 2,n )} vary according to the state of the background process {J n }. This modulation is assumed to be space homogeneous. We refer to this process as a two-dimensional skip-free Markov modulate random walk. For y, y ∈ Z 2 + × S 0 , consider the process {Y n } n≥0 starting from the state y and let qy,y be the expected number of visits to the state y before the process leaves the nonnegative area Z 2 + × S 0 for the first time. For y = (x 1 , x 2 , j) ∈ Z 2 + × S 0 , the measure (q y,y ; y = (x 1 , x 2 , j ) ∈ Z 2 + × S 0 ) is called an occupation measure. Our main aim is to obtain asymptotic decay rate of the occupation measure as the values of x 1 and x 2 go to infinity in a given direction. We also obtain the convergence domain of the matrix moment generating function of the occupation measures.