* An aggregate re-calculated for earlier periods using the current composition of the euro area.
Convergence among the peripheral member nations of the euro area developed similarly. The initial asymmetric shocks were primarily the consequences of German reunification and the EMS crisis. Except in the case of Portugal, correlation coefficients of between 0.4 and 0.8 were typical both in this group of countries and the accession countries towards the second half of the decade, which suggests the adequate synchrony of the cycles. It is worthy noting that in the past 5 years the cyclical asymmetry among countries has moderated across Europe, except in Portugal and the non-EU member Norway.
Structural Vector Autoregression (SVAR) models may help to decompose business cycles into shocks originating from the demand and supply sides. Distinguishing between demand and supply effects may be useful for two reasons. First, supply shocks may be interpreted as autonomous factors which affect the economy anyway, even in a currency union. Explanation for this is that supply shocks do not reflect the effect of monetary policy influencing typically the demand side. At the same time, monetary policy may behave differently from country to country, which affects the symmetry of demand shocks. Consequently, they give limited information on future co-movements within the euro area.
The second advantage of segregating demand and supply shocks is that in this way we can distinguish between long-term (supply) and short-term (demand) effects. Supply shocks tend to be more persistent, as they are partly attributable to changes in and development of production technologies. According to modern-day theory, the primary task of monetary policy may be to smooth out temporary fluctuations rather than influence trends. Accordingly, it can achieve price stability and contribute to more balanced output growth mainly by neutralising demand shocks. So, somewhat in contrast with our previous argument, the degree to which a currency area is optimal will depend on the similarities of demand shocks.
An analysis of the past suggests that the second approach can be seen as more relevant, as the post-communist countries experienced economic restructuring in the 1990s, which was accompanied by a series of country-specific supply shocks. In contrast with technological innovation in the developed regions of Europe, technology imports appear to have played a dominant role in supply shocks in the accession countries. That is why it is not surprising that we have been able to find only modest correlation for supply shocks. However, the catching-up process will likely entail a
reduction in supply-side asymmetries, and thus supply correlation is expected to increase in the coming years.
Our method, pioneered by Blanchard and Quah (1989), was introduced to the literature on OCAs by Bayoumi and Eichengreen(1993). Segregation of demand and supply fundamentally rests on the simple assumption that demand-pull price increases induce a temporaryrise in output, whereas supply-push inflation triggers a lastingdrop.48
After calculating demand and supply shocks for each country, they can be compared to demand and supply series relating to Europe or the euro area as a whole. Three methods are used to determine the latter. These methods are similar insofar as the European or euro-area demand and supply shocks are derived as some kind of weighted average of the individual country series. In the first method, the basis of comparison is the component that principally accounts for variance in the series of the West European countries. In the second method, individual countries’ demand and supply shocks are weighted together on the basis of their effect on inflation in the current euro area. Finally, in view of the fact that external demand is crucially important for the state of the Hungarian business cycles and external balance, comparison is made with the weighted average of the roles of Hungary’s trading partners in Hungarian exports.
The advantage of using the principal component analysis to derive European demand and supply serie49is that larger countries will only carry a higher weight to the extent that they exhibit actual correlation with the cycles of several other countries.
Thus, this approach provides a numerical representation of genuine shocks common to several countries simultaneously. During the period between 1992 Q1 and 2000 Q4, France, Germany and Belgium showed much higher-than-average correlation with the European component in respect of both demand and supply. Similarly, Switzerland, Spain and Finland also exhibited significant correlation in terms of both constituents. Luxembourg and the Netherlands preceded all the other European states with regard to demand and supply correlation, respectively. Of the
48For a detailed description of the method applied, refer to Appendix F.2.
49The inclusion of other principal components in addition to the first principal component, which accounts for 10-50% of the demand shocks and 0-45% of the supply shocks to non-peripheral EU member states, will not significant increase explanatory power. This justifies the proposition to view the first principal components as common European components, with their values interpreted as pan-European demand and supply shocks.
accession countries under review, the Czech Republic can be viewed as the country least exposed to asymmetric shocks in the 1990s. Demand shocks affecting Hungary showed broadly the same degree of correlation (0.4) with Europe as those
in Italy, the Czech Republic and Great Britain. Its correlation coefficient of 0.2 for supply fluctuations, corresponding to the level for Norway and Luxembourg, represents a slightly greater degree of symmetry than that of Austria and Greece, and slightly smaller degree than that exhibited by Spain and Portugal.
It is primarily Hungary’s demand correlation that exhibits symmetry of a magnitude that corresponds to that of many current euro-area member nations. At the same time, the inference suggested by the country’s supply correlation is that the exchange rate of the home currency is needed as a shock absorber. Nevertheless, the situation appears to be more promising from the perspective of accession,