This paper is devoted to an extension of Dibooglu and Kutan’s work [Journal of Comparative Economics, June 2001], in two directions. First, a bivariate VAR, including the real effective exchange rate (REER) and inflation, is tested not only for Hungary and Poland, but also for the Czech Republic, over the 1993:1-2002:12 time-period, which excludes the early transition years. Second, industrial production and the nominal interest rate are incorporated in a multivariate VAR to investigate the effects of policy shocks on the exchange rate. This helps to illustrate the possible gains and costs of the present exchange-rate regime in these CEECs and to draw some conclusions on the likely economic outcomes of adopting a firmly fixed exchange-rate regime like ERM II. Three main conclusions emerge: (1) contrary to Dibooglu and Kutan (2001), results testify to a high degree of flexibility in the CPI; (2) in contrast to the Czech Republic and Hungary, nominal shocks have a strong effect on the REER in Poland and, more specifically, there is evidence that the fluctuations in the nominal exchange rate explain a large part of REER fluctuations in this latter country; (3) in Poland, the policy mix seems more credible than in the other two countries. We infer from these conclusions that moving to the ERM II and the EU will be of less benefit to Poland than to the Czech Republic and Hungary.