07 Jan 2004
CASE Network Reports

Harrod-Balassa-Samuelson Effect in Selected Countries of Central and Eastern Europe


This study investigates the HBS effect in a panel of nine CEECs during 1993:Q1-2003:Q4 (unbalanced panel). Prior to estimating the model, we analyze several key assumptions of the model (e.g. wage equalisation, PPP and sectoral division) and elaborate on possible consequences of their failure to hold. In the empirical part of the paper, we check the level of integration of the variables in our panel using the Pedroni panel-stationarity tests. We then investigate the internal and external version of the HBS effect with the Pedroni panel-cointegration tests as well as by means of group-mean FMOLS and PMGE estimations to conclude that there is a strong evidence in support of the internal HBS and ambiguous evidence regarding the external HBS. Our estimates of the size of inflation and real appreciation consistent with the HBS effect turned out generally within the range of previous estimates in the literature (0-3 % per annum).  However, we warn against drawing automatic policy conclusions based on these figures due to very strong assumptions on which they rest (which may not be met in near future). Finally, following the hypotheses put forward in the literature, we elaborate and attempt to evaluate empirically the potential impact of exchange rate regimes on the magnitude of the HBS effect.