A Comment on Andersen and Risager (1991) [The Role of Credibility for the Effects of a Change in the Exchange-Rate Policy]