1: \begin{abstract}

2: We introduce a general algorithm for the computation of the

3: scale functions of a spectrally negative L\'evy process $X$,

4: based on a natural weak approximation of $X$ via upwards

5: skip-free continuous-time Markov chains with stationary

6: independent increments. The algorithm consists of evaluating a finite linear 

7: recursion with its (nonnegative) coefficients given explicitly in terms 

8: of the L\'evy triplet of $X$.

9: Thus it is easy to implement and numerically stable.

10: Our main result establishes sharp rates of convergence 

11: of this algorithm providing an explicit link between the semimartingale

12: characteristics of $X$ and its scale functions, not unlike the

13: one-dimensional It\^o diffusion setting, where scale functions are expressed

14: in terms of certain integrals of the coefficients of the governing

15: SDE.

16: \end{abstract}

17: