As the European Union grows eastwards, the criteria for joining the inner sanctum of the Union, the EMU, draws renewed attention. In 2004, ten Central and Eastern European Countries were admitted to the EU and as of 1 January 2007, Romania and Bulgaria followed. Among these new members, only Slovenia has so far managed to adopt the euro and by that taken the final step towards full European integration. De Grauwe & Schnabl (2004) predicted that EMU entry for the countries joining the EU in 2004 would be within 2006. Why has only Slovenia succeeded?
To become a member of the EMU a candidate state must meet the Maastricht criteria. Two of these criteria are of special interest to this thesis. The first is the exchange rate criterion, stating that the candidate must have pegged its currency to the euro and kept it within a range of ±15 % (The ERM II band) with respect to a central rate against the euro - for at least 2 years. The second is the inflation criterion, stating that the candidate’s inflation rate must be no higher than or equal to 1,5 % above the average of the three best performers of the EU. All candidates have signed an obligation to adopt the euro in the EU Treaty of Accession and by that committed themselves to work towards fulfilling the Maastricht criteria. However, the countries do struggle in order to accomplish this task and one of the main challenges is inflation.
This thesis seeks to explore the reason for the high inflation rates in the three Baltic countries, and in the Slovak Republic, since inflation seems to be the main obstacle to their membership in the EMU. I look at these four candidate economies since they seem to be the next ones in line to qualify for EMU membership. Slovenia is also included in the analysis, because it became the first candidate from the East to join the EMU and its performance therefore is of interest as a benchmark for the ones next in line. I have chosen to look for the existence of the Balassa-Samuelson effect, as this effect explains a natural economic mechanism, stating that transition countries with high productivity growth in the tradable sector, and with equal nominal wage growth between the sectors, experience a temporary higher inflation rate than what you see in Western Europe. This excessive inflation rate is supposed to vanish when the convergence process is completed (productivity growth flattens out). I look for the effect relative to the euro area, individually for each country. Is it present? If that is the case could not an inflation rate slightly higher than the required rate be taken as proof of a healthy economy, showing that inflation is not out of control due to poor economic management, but rather a consequence of a successful transition? Based on the results I wish to explore whether the inflation criteria can be satisfied by a country in convergence or alternatively whether the Maastricht bottleneck is too tight. I find that the Balassa-Samuelson effect is present in Latvia and Lithuania and that the effect has increased rather than diminished in recent years. In the Slovak Republic the effect is significant, but has the wrong sign, which is puzzling. I also find that changes in the nominal exchange rate vis-à-vis the euro have an effect on the inflation differential between the accession country and the Euro area, but that this effect is smaller than the effect through the relative labor productivity differential - for all countries but Slovenia. A possible relative wage growth differential, incorporated as an additional explanatory variable, turned out to have no significant effect on the inflation differential in any country, including the ones experiencing non-uniform wage growth. For Estonia significant results were only obtained for changes in the exchange rate which is odd and suggest errors in the Estonian data.
The results lead me to conclude that the transition process is not yet over in the countries which were selected for this study. The Balassa-Samuelson effect is still present, but small in Latvia and Lithuania over the whole sample period, meaning that it is not able to explain a lot of the inflation differential, but that at least some of the difference in inflation rates between the candidates and the euro area is due to a “natural” development. The effect has been rather large in Lithuania in recent years and definitely present when the country applied for EMU membership in 2006. At that time an inflation rate 0,1 % higher than the required inflation rate was the only impediment to EMU membership. If only a small part of the Balassa-Samuelson effect had been allowed and adjusted for then, as a sign of naturally higher inflation, Lithuania might have been an EMU-member today. The significant effect of changes in the nominal exchange rate on the inflation differential leads me to conclude that meeting both the exchange rate and the inflation rate criteria at the same time may be easier if the exchange rate is not entirely fixed, but allowed to vary within the ERM II band.