Nominal and real convergence in Estonia: the Balassa-Samuelson (Dis)connection

The objective of the paper is to analyse the nominal and real convergence process in Estonia drawing on the Balassa-Samuelson (B-S) framework. A 15-sectoral breakdown for GDP and a 5-digit level CPI data disaggregation with over 260 items is used for the period 1993:Q1 to 2002:Q1 to show that the productivity differential is related to the GDP-deflator relative price of non-tradable goods in the long run. Furthermore, the role of regulated prices in the CPI basket is also investigated - we show that excluding regulated prices makes it possible to detect a robust relationship between productivity and the relative price of market services in CPI. The B-S effect could have possibly contributed to CPI by a yearly average of 2-3% over the sample period, and more specifically 1-4% at the beginning of the period and 0.5-1% in 2000 and 2001. The potential long-run impact of the B-S effect in Estonia is estimated to amount to 1-2%. Analysis of the influence of the B-S effect on the inflation differential and the real appreciation of the exchange rate against Finland, Sweden, Germany and the UK, shows that, whereas the inflation differential attributable to the B-S effect seems to have been higher in the early 1990s, it better explains the real appreciation occurring in recent years.