MONETARY UNION ACCESSION AND THE SEVERITY OF POTENTIAL ASYMMETRIC SHOCKS - THE CASE OF LITHUANIA AND POLAND

The invention relates to a 1.4-1.52 mum-band optical amplifier. This optical amplifier includes (a) a first pumping source for a first pumping light in a wavelength range defined as being not less than 0.65 mum and being less than 0.77 mum; (b) a second pumping source for a second pumping light in a wavelength range of from 1.0 to 1.2 mum; (c) an optical multiplexer for combining said first and second pumping lights with a signal light, to produce a combined light; and (d) a light amplifying optical waveguide for amplifying said signal light by receiving said combined light, said optical waveguide comprising a core doped with Tm. The optical amplifier is highly efficient.


Introduction
The decision on joining a monetary union is rather fundamental in terms of economic policy. In the case of Lithuania and Poland, it can be only compared with the decision of economic system transformation from the beginning of the nineties. By accessing the European Union, both of the countries have committed themselves to join also the euro monetary union, but the question about the specific date of replacing the local currencies by the euro stilI remains open. Lithuania has already started the preparatory phase entering the ERM2, however, a lot of doubts have emerged about the accurate timing for this decision since the moment the whole process has started four years ago. Polish economists are also not unanimous in their opinions about the calendar of joining the euro area, while some of them even undermine the sense of forgoing the Polish Zloty itself CLan, 2007).
During the discussion about the potential economic costs of common currency, the question of asymmetric shocks is perhaps the most popular one. We define an asymmetric shock as a disturbance that causes different in size negative! output deviations from the potential level among the member countries of a monetary union. Under the common currency regime there is no longer possibility to neutralize these shocks using countryspecific tools of monetary policy, hence a special attention is paid to this problem. According to the seminal paper by Mundell (1961), there are two main conditions that predestinate a candidate country to join the monetary union, establishing a situation where the severity of potential asymmetric shocks will be minimized. These conditions are high labour mobility and the flexibility of wages, especially in downward direction. If the labour market is flexible enough, the economy easier restores an equilibrium with the initial level of prices and income per capita.
We cannot treat the above suggestions as a finite list of conditions which, being fulfilled, will guarantee the success of the accession process. There are, of course, many other detailed factors, e.g. the degree of openness of an economy (McKinnon, 1963) or the degree of product diversification among the member countries (Kenen, 1969); however, the Mundell's conclusions can be viewed as some general prerequisites to the accession process. Therefore, the first goal aim of this paper is to verify, using the available labour market statistics, whether the Lithuanian and Polish job markets are flexible enough to easily neutralize the potential economic disturbance. I The Common Currency Area theory pays attention to negative deviations from the potential level of output.

8
The latter target focuses on estimating the chance of emerging a special type of an asymmetric shock.
Disturbances that cause asymmetric reaction can be classified in a number of ways. European Parliament (1998) divides shocks into real and financial as well as into exogenous and policy-induced shocks. Borowski (2001) distinguishes two main types of asymmetric shocks: "classical", i. e. countryspecific shocks that do not affect the other countries of monetary union, and another group of shocks that influence all the common currency area, however, to a different degree.
Balancing the benefits and costs of joining the monetary union and discussing the optimal timing of accession, we need a more pragmatic way of describing the asymmetric shocks. We suggest, therefore, to categorize shocks from the point of view of our ability to predict these disturbances. Having established this classification, we may then better identify the potential threats of monetary union accession.
Perhaps there are no totally unpredictable as well as definitely certain shocks. The natural disasters are those that are always hard or even impossible to predict. The productivity shocks are also generally unpredictable, but very often they are also difficult to identify. Using the portfolio theory language, these types of shocks may be treated as "non-diversifable risks", hence we should pay attention mainly to these sources of shocks that can be "diversified" by the appropriate timing of the monetary union accession. We would like, therefore to focus on the common monetary policy that may be a source of the asymmetric shock itself.
Under the common central banking regime, the monetary authorities' target is the average inflation rate for the whole union area, so it is not difficult to imagine a situation where the distribution of the inflation rates among the member countries is dispersed and at the same time the Central Bank reaches its target. In such a case, the common monetary policy can destabilize some of the local economies leading them to an unnecessary recession or overheating the economy. Looking at the country-inflation rates of the euro area and keeping in mind the 2% target inflation rate, we may observe this kind of dispersion.
There is a group of countries (Ireland, Greece, Spain, Luxembourg, Portugal) which have country inflation rates permanently above the target, and we may distinguish countries with a permanently lower inflation (Germany and in the recent years the economic perspective, however, there are some doubts whether fulfilling the Maastricht criteria really guarantees economic convergence; therefore, we will run a separate analysis of business cycles synchronization that might provide some valuable conclusions. Finally, our second research target is to elucidate whether for Lithuania and Poland the common monetary policy may be a source of an asymmetric shock.

Data and methodology
The two broad definitions of the business cycle are recognized in the literature (Reichlin, 2005): the so-called classical cycle and the growth or deviation cycle. The classical business cycle is a sequence of expansions and contractions in the absolute level of economic activity. The growth cycle is defined in terms of the deviation from trend or potential output, and thus within an additive or multiplicative trend-cycle decomposition. The rationale for investigating the deviation cycle is that absolute prolonged declines in the level of economic activity tend to be rare events, so, in practice, many economies have not exhibited recessions in classical terms. The notion of the cycle as a deviation of current output from its potential has become increasingly relevant for the conduct of monetary policy, and it is this concept of the cycle that is the focus of attention here. In this paper, deviation cycles are identified for Poland, Lithuania and for the EU-13 as an aggregate. The four key economic indicators are reviewed in terms of their cyclical movements across countries, and both activity and price developments are covered. The indicators considered are: real GDp, unemployment rate, industrial production index and the harmonised index of consumer prices (HICP). These play an important role in an assessment of the economic situation. Our sample covers the period between 1995-2007 for quarterly data on GDP and the period between 1998-2007 for the other monthly data. As a data source, the official Eurostat webpage was used. Most of the series are given by the Eurostat in an already seasonally and working-day-adjusted form.
Although it can prove difficult to distinguish empirically between tren·d and cycle, such a distinction is helpful, as these two phenomena frequently tend to be discussed in connection with different economic issues.
For exampl~, the extent to which trend developments in output growth diverge or narrow across the countries is discussed in the context of countries "catching up" with one another, while the degree of similarity of cyclical movements across the countries is discussed in connection with the issue of "synchronization". Ideally, what is required is an estimate of the asymmetry in shocks that a country will experience in future versus those that will impact its prospective partner. One would like to have estimates of the "no policy" shocks, since it is one of the tasks of policy precisely to offset them. In practice, economists have had recourse to two alternative approaches to get an estimate of the asymmetry of shocks. One, following Bayoumi and Eichengreen (1993), isolates the shocks as the error terms in an estimate of a twovariable (output and prices) structural vector auto regression model (SVAR) on which some identification restrictions have been imposed. Once supply and demand shocks are identified separately for individual countries, synchronization is assessed by the correlation between the shocks. However, the use of SVARs is debated even for countries having long sample periods (Darvas, Szapary, 2004). Imposing long-run identifying restriction for data available for Lithuania and Poland would not make much sense in the framework of the SVAR model (see Faust, Leeper, 1997;Cooley, Dwyer, 1998). Moreover, some ofthe inflation data series are not stationary and seem to be even an 1(2) process which raises a problem that is quite difficult to handle.
Due to these theoretical and practical deficiencies of the SVAR technique, in this paper we employ the alternative and more "atheoretical" approach based on identifying asymmetries in the business cycle phase as a signal of asymmetric shocks in the relevant sense. The adopted approach consists of using detrended time series as cyclical measures and calculating various synchronization measures based on them. The same method was used by Artis and Zhang (1995), Borowski (2001), Darvas and Szapary (2004). In the following, we describe the methodological issues related to detrending and the measurement of synchronization.
The first issue is the decomposition of the observed series into a trend movement and cyclical component. There are a number of different ways of extracting the trend from the data. However, while different detrending methods give rise to somewhat different results, it appears that the principal results and the validity of broad conclusions are not sensitive to the method chosen (ECB, 1999). For the purposes of this paper, the trend series are derived on the basis of purely statistical considerations rather than from a specific economic theory. Due to its widespread use in empirical economics, the Hodrick-Prescott (HP) filter has been applied here in order to mechanically decompose the individual indicators into a trend movement and a cyclical component. The HP filter can be specified as ~~n{~(y, _g,)2 +A~[(gl+l -g,)-_(g, _g,_I)Y}' (1) where Yt denotes the raw series, g, the growth component and (y, -g) the cyclical component.
The first part measures the fitness and the second is a measure of smoothness. The parameter A. is the signal-to-noise ratio and weights the relative importance of the two conflicting goals in the loss function: when A. = 0 the filter gives the original series; as A. goes to infinity, the HP filter collapses to a linear trend. Hodrick and Prescott (1997) suggest a value of A. = 14400 for monthly data and A. = 1600 for quarterly data, and these values have been adopted in the estimates here. The HP filter is easy to implement, but suffers from the end-point bias owing to the fact that both lagged and lead values of the series are taken into consideration when calculating the trend, thus making it less precise at the beginning and at the end of the sample. Nevertheless, we decided to use the HP filter, because all alternative detrending methods also suffer from specific shortcomings.
The techniques used to investigate the question of synchronization exploit correlation analysis between the cyclical series. A high coefficient of correlation will indicate that countries tend to be in similar states of cyclical movement. We used three measures. Firstly, we calculated simple contemporaneous unconditional correlations between the cycle of the euro area and those of individual countries for each indicator, which provide the information about overall linkage between cyclical movements. Next, we examined developments in synchronization over time on the basis of the contemporaneous correlation coefficients for rolling 3-year periods, which allows for a fairly comprehensive analysis. While evidence of increasing or decreasing synchronization may emerge, there is uncertainty as to whether this is due to generally higher or lower linkages in cyclical developments or simply to a phase shift of the cycles. With this end in view, we provided an explicit measure of phase shift by finding the lead or lag at which the maximum correlation is obtained. Thus, for a given pair of cycles, X; and y~ Pt; (X" Y,t;) denotes the correlation between X; and Y, at displacement ± i . In this paper, in order not to decrease the degrees of freedom too much, the maximum value of I P±; I is chosen for i ~ 12 for monthly data and for i ~ 4 for quarterly data. From the perspective of the optimum currency area, zero or smallleadnag would be optimal.

Labour market flexibility
Flexible labour market conditions stimulate people's professional activity, hen.ce the employment rate may be treated as the most straightforward indicator of labour market flexibility. Table 2 displays the data.
The assessment of labour market flexibility of the two candidate countries differs significantly. The Lithuanian employment rate is nearly equal to the euro area average, while Polish employment is the lowest in the whole sample. According to the conclusions of the Optimum Currency Area theory, Lithuania is more predestined to join the euro area. Comparing, however, the euro area employment rate with the data form Japan and USA where the employment rate is about 70%, there is still a need to reform the labour market in euro area to enhance its flexibility and competiti veness. The reasons for the low market flexibility and at the same moment the targets of desired labour market reforms have been repeated several times by the economists (Balcerowicz, 2008): too high a tax rate on labour, too strong a position of the trade unions. In these circumstances, the crucial -0.59 -0.05 0.24 0.58 point seems to be the lack of political will rather than the lack of good ideas. Table 3 shows the overall degree of synchronization between EU-13 as an aggregate and the individual countries' cycles for the whole sample period. Figure 1 shows three-year rolling sample correlations for quarterly data on GDP between EU-13 and Lithuania, Poland and Germany, the latter being graphed here for comparison purposes only.

Synchronization of business cycles
The coefficients in Figure 1 refer to the end of the respective period, i.e. the latest available data for 1998Ql reflect the average ........ points to an increasing rather than decreasing synchronization since 2006, especially in industrial production and unemployment rate. Synchronization of the cyclical element of price increase with that of the euro area as a whole was characterized by relatively low correlation coefficients ranging from 0 to 0.25 during the last two years. However, in periods of relatively low inflation rates, more temporary country-specific factors may dominate, and it becomes increasingly difficult to disentangle cyclical movements from a purely statistical noise caused by all kinds of relatively small shocks to the price level (e.g. differing seasonal patterns, liberalization measures, administrative price changes and asymmetric effects caused by differences in the composition of the basket of consumer goods and services). Synchronization patterns for Lithuania are different. While the correlation of the cyclical element of price developments remained relatively stable at around 0.5 for the examined period (see also Table 3), there is a noticeably lower degree of synchronization in the unemployment rate and industrial production compared with Poland. This surprising observation, especially in case of unemployment rate, has its origins in the high labour force migration from Lithuania to the countries (mainly UK and Ireland) that decided to open their labour markets for the citizens of the new member states from the very beginning, i. e. from May 2004. The low correlation coefficient, therefore, is a statistical fact rather than a real economic phenomenon: in case of Poland, the unemployment rate has been diminishing more slowly, which has been more in line with the average rate for the euro area countries, while the Lithuanian unemployment rate has been decreasing faster, hence the correlation is lower. Labour force migration in case of both countries has been a kind of macroeconomic shock; therefore, more credible results of cyclical components synchronization will be possible to obtain for a few years.
Additionally, we have estimated the values of the leads/lags in the cycle components, which are denoted by i, and the corresponding highest correlation value I P±i I between the euro area and the individual countries examined. Table 4 displays these results.
The interpretation of the results for this measure is the following: the zero value indicates that the contemporaneous correlation is the highest, negative values indicate that the euro area leads the country studied, while a positive number indicates the reverse. The price indexes for both countries perform the

Conclusions
Our analysis has led to two main conclusions: labour markets in both countries are not highly flexible (especially the Polish labour market) and the business cycles are not strongly correlated with the euro area average, indicating that at the present moment common monetary policy could be a source of an asymmetric shock, so for Lithuania and Poland the current macroeconomic situation does not seem to be the optimal moment for entering the euro area. For practical reasons, however, we think that the best strategy of joining the monetary union is the ASAP strategy. Recommending early terms for entering the euro area may work like a self-fulfilling forecast. It may stimulate the government to conduct the sound economic refonns which helps to fulfil the Maastricht criteria. Emphasizing the potential benefits of the common currency may help also to find the public support ofthis idea. In our opinion, discussing the scenarios of abandoning the process of entering the euro area may be perceived only in economic-fic-tion tenns because it would inevitability be followed by the costs of losing the country's investment credibility, which are even hard to assess. Poland and Lithuania are still emerging market economies, their currencies are not the major world currencies, so there is no chance for successfully repeating the UK case.
It is also necessary to note the questions that this paper does not investigate. It does not examine the sources of shocks] i. e. whether the business fluctuations are caused by supply or demand shocks. Many authors (for instance, B1anchard and Quah, 1989) have found that both demand and supply shocks contribute to fluctuations, the fonner dominating in the shorter frequencies and the latter becoming important in the longer run. Identifying the sources of shocks is important because monetary policy cannot deal with all types of shocks similarly. Another question our paper does not investigate is the channels of transmission of business cycles from one country to another. The empirical evidence discussed in the literature shows that openness, trade integration and a similarity of economic structures have a strong effect on international co-movements. Investigating the sources of shocks and the transmission mechanism of business cycles, however, exceeds the scope of this paper, but, in our opinion, both ofthese challenging issues should become the dominant areas of future research on monetary integration.