ECONOMIC PROSPECTS IN THE CONTEXT OF GROWING REGIONAL INTERDEPENDENCIES : THE EUROPEAN UNION AND THE EASTERN PARTNERSHIP

Abstract. The paper deals with the European Union programme devoted to the eastern neighboring states. Through its European Neighbourhood Policy (ENP), the EU works with its southern and eastern neighbours to achieve the closest possible political association and the greatest possible degree of economic integration. This goal builds on common interests and values — democracy, the rule of law, respect for human rights, and social cohesion. The EU is concerned that, despite sufficient funding and support from the EU, the targeted states did not raise to the EU targets for the programme or at least to a relevant one. We assume that such fact happened mostly because, although having very diverse economic and reform pasts emerged from the post-soviet period, they were considered and approached as a single group. The main hypothesis: has the umbrella of the EU funds in terms of the EaP provided for the six targeted states to intensify the growth of regional interdependencies as well as political cooperation and progressive economic integration? The main goal of the paper is to assess, by means of the statistical and comparison approach, the development and the economic sustainability of six targeted states (Belarus, Moldova, Ukraine, Armenia, Azerbaijan, and Georgia) in the period before and after the programme launching – the degree of regional interdependence and economic integration. The research was conducted using the methods of empirical (regression) analysis, theoretical explanations, descriptive analysis, and the Granger causality test.


Introduction
One goal of strategic importance for the EU has been to reinforce relationships with the neighbor border states since the 1990s.The EU's Eastern and North-Eastern neighbors include six post-soviet countries -Belarus, Moldova, Ukraine and the three countries of the South Caucasus (Armenia, Azerbaijan, and Georgia).The Eastern Partnership (EaP) is the European Union's leading policy initiative to forge closer ties with six countries in Eastern Europe and the South Caucasus.Established in 2009, the partnership seeks to promote regional stability through trade agreements and democratic institution-building.The financing relations that the EU maintains with these member countries has been

Literature review
The browsing of the Google Scholar (a broad and famous depository of scientific papers in Open Access) provides about 103 thousand research papers linked to "the EU and the Eastern partnership" word combinations, and there are only 45 papers that contain these word combinations exactly.This is a rather miserable amount given the significance of this policy.But most interesting is the "temperature / attitude" of researches on the topic -which mostly look concerned, doubtful and uncertain as to the ability of the EU to make the policy effective and legitimate in the region (Korosteleva, 2012).The timeline of scientific thoughts about the topic is also rather demonstrative.Thus, up to the year 2008 when the idea of EaP had been presented by the foreign ministers of Poland and Sweden in Brussels, only about 40% of the current scientific collection on the topic had been published in the Google Scholar depository.It should be particularly emphasized that all papers in the indicated period considered mostly the EU partnership with some particular states or the aspects of integration / expansion of the EU to the East, as well as the possibilities of enlarging strategies and looking for buffer zones between the EU and Russia.Only after the political decision (2008) this combination of words -"EU & EaP" -appeared in the titles and texts of scientific papers, as well as further ENP / ENPI.
Taking into account that the Google Scholar contains and offers only 45 published papers, particularly after 2009, which have the "EU & EaP (ENP)" combination in the title or in the body text, we considered those most cited for this analysis (Table 1).
As it is clear from Table 1, most scientists came to express deep concern about the successful realization of such policy.However, a detailed analysis of fragmentation motives that appeal and provide a full understanding of "groundwater flows" in the six target countries on their way to the absorption and realization of reforms.The establishment of the EaP in 2009 can be considered as a central element of the new European diplomacy in the eastern borderlands where the means of the neighborhood policy become a target (Lapenko, Arshinov, 2010).The funds for supporting this effort are notably massive, i.e. overall 2.5 billion euro available for the European Neighborhood Instrument in the following quotas for 2011-2013 (eeas.europe.eu):Armenia -€ 182 million; Azerbaijan -€ 75.5 million; Belarus -€ 41.5 million; Georgia -€ 208 million; Moldova -€ 308 million; Ukraine -€ 389 million, as well as some funds on flagship initiatives.
Even a slight look at these numbers can catch quite obviously the lack of statistical analysis to explain the choice of sums and proportions; at least it is not open for the majority.Also, the literature review revealed the absence of statistical researches on the topic.However, there are some statistical researches of the EU & ENP that are concentrated on migration tendencies (Barbone et al., 2013).
The aim of our research is not to glance at the source of the programme steps for the six target countries in 2008-2014, and not even to consider the exact funds and their effectiveness in particular metrics, but to consider the realization of the main aim of the EaP (and following ENP / ENPI) at promoting an enhanced political cooperation and progressive economic integration between the EU and the six countries from the position of grounded statistical results and the tracing of main tendencies.
The main hypothesis: does the umbrella of the EU funds in terms of the EaP provide for the six targeted states to intensify the regional interdependencies as well as political cooperation and progressive economic integration?

No statistical analysis
Reviews the experience of implementing the EU assistance in the region of the Eastern Partnership in the current financial perspective (2007)(2008)(2009)(2010)(2011)(2012)(2013), suggesting ways in which it can be made into a more effective instrument for realizing the political priorities of cooperation Excessive thematic fragmentation, inconsistent application of the "more for more" principle, and the insufficient volume of aid to civil society Source: author's compilation.
We claim that the EU implemented funds and policy tools unified all the six targeted states without considering the unequal levels of civil society and economic development at the time of the Programme establishment.This was not quite appropriate.So, the process of integration and the EU standards' diffusion remained diverse and slow in the target states as could be expected without any funding.Also, the reflection of old, postsoviet tendencies is so strong in these states that an "individual" approach should be used first by the EU based not only on the political point of view and assumptions, but mostly on the results of a complex survey of the six target economies on their way to the EU (preferably economic-mathematical one).
Main methods: to achieve our goal, we use the knowledge of general scientific methods (analysis and synthesis, comparative, historical, and logical) and statistical approach (empirical (regression) analysis, descriptive analysis, and the Granger causality test), which seems to be the best possible approach to analyze trends for the development of six targeted states during the EU programme funding and before it.The choice of these methods is due to the logic of the study as today the application of mathematical methods is a prerequisite for a complex analysis of economic processes, ensuring high requirements to the validity, effectiveness and feasibility of the model forecasts for economic processes.This, in turn, makes it possible to avoid random one-sided conclusions and increases the reliability and validity of the final results of the statistical analysis.
Practical outcome: understanding the macro-economic trends in the six targeted states will help to develop the policies whose target is not to make these states part of the EU, but first of all to make them predictable, synergetic and a reliable buffer for the EU.

Statistics and empirical assessment
The EU has allocated 175 million Euros in 2011-2013 to the programs related to the institutional development and reforms in the countries of the EaP.What do these funds represent for the recipient states?For the post-soviet "Eastern Partnership" republics, this is a way to get funding from the EU, which can cause some concern for Brussels: a country applying for the EU membership in advance behaves as a subsidized member but does not reflect the EU standards and interests.The EU is challenging this approach by asking -"Eastern Partnership" versus European integration: with or instead of?
When we consider these six countries from the position of their reforms and their internal macroeconomic development (Table 2), it seems that the situation is entirely stable in the worsening direction: 24 positions have shown a deterioration in the period of the EaP funding fulfilment, and only 14 positions show an improvement.Also, a negative signal for the targeted group of countries is that the scores evaluated by the Freedom House, despite some changes, remained in the negative limit level during the whole period of 2005-2014 without any sign of improvement.

Corruption
Source: author's compilation based on the scores of the Freedom House 'Nations in Transit' surveys 2006-2014.https://www.freedomhouse.org/report-types/nations-transit#.VMz70y6NvVo Table 2 demonstrates that the targeted countries, being unequal in the democratic performance until engaging in the EaP (further ENP / ENPI), still kept the same unequal performance even after receiving the first funds and launching the projects of action.Thus, the most improved scores are presented by Georgia and Moldova; only their judiciary system is still getting worse, but the other main positions are better or at least the same.This fact is an evidence that the correspondingly higher support of the EU to these two states has stimulated civil society and democratic reforms in them.As to the other states, there is an evidence that the funding of the projects of action and other initiatives have been productive at a rather low level and mostly kept the situation at the same milestone or were a stimulus to worsening (possibly caused by corruption and the non-transparent use of funds).As for Ukraine, despite the rather high sum of funding in comparison with the other targeted states (actually the largest level), the country suffers the internal and border conflicts.The situation and scale of Ukrainian society is not reflected correspondingly in the sum and structure of the EU projects.Even during 2013-2014 the situation was still deteriorating (Table 2).The most likely reason is that, first of all, the internal climate and features of business climate in such a large country were not analyzed enough and taken into account when the sum and the drivers for its delivery were considered by the EU.
As the next step of our research we use a descriptive statistical analysis to describe quantitatively the main features of collecting information on the main macroeconomic indicators of dynamics for the six targeted states and the EU before, during, and after the Programme launching.Figures 1-5 are demonstrative for the following conclusions: 1) the economic growth of the states was quite diverse (Fig. 1).There is no synergy / convergence in the stripes of the correspondent indicators.The crisis years in the 1990s and in 2009 were highly dramatic for the states, what shows that the EU integration direction did not support the economic consistency and robustness to shocks of the target states.The joining of Azerbaijan to this programme seems a bit unclear, as the macroeconomic development as well as the democratic progress (Figs. 1 and 2, Table 2) are absolutely different in this state in comparison to the EU and the rest of targeted states.However, some macroeconomic stability was established in the analyzed region during the first half of the 1990s and has been maintained since then; 2) the GNI (current US$) values for the EU and the targeted states are incomparable as the EU level is 100 times higher than for the lead indicator value in the analyzed group for the period 1990-2013 and after 2008 (the year of the programme launch) the situation just depreciated.But as to GDP and GNI per capita growth (annual %), it is possible to see the same average level and appearance of a convergence in the dynamics after 2008 (Figs.2-3); 3) the trends of trade of the EU (in its part of GDP) are synergetic in its dynamics with the targeted states' trends for the period until and after the programme launch.However, trade volumes (as % of GDP) are higher in the six states than in the EU itself (Fig. 4).This fact puts under consideration the necessity of establishing the project in action of the ENP initiative "Deep and Comprehensive Free Trade Area (DCFTA)"; 4) Most positive tendencies that accompanied the programme are noted in the aspect of taming the inflation in the region (Fig. 5).Despite the existing internal challenges, all the six states (beside Belarus) managed to harmonize their inflation rates and bring them close to the European standard; at least the volatility of rates kept quasi-equal.
Summing up the results of the main macroeconomic indicators in dynamics, we can conclude that the comparison analysis proved the general possibility of the targeted states to integrate into the EU in unison to the EU dynamics.Most politicians and economists find that the unison dynamics of the main indicators of a country's health is the first positive sign -litmus -that reforms are effective (the best known case of Poland and the Baltic states in the years of their integration to the EU).But there is quite a high diversity inside the target group itself.The evidence is quite obvious that these six states have been chosen not from the economic point of view but from the political one.As from the economic point of view, it would be better to separate the same programme projects in action into two different blocks or better to consider and fund states separately, according to their particular needs and unstableness.However, there is still a tendency of keeping the same idea and the same aims.Source: author's calculations on the base of the World Bank data.http://data.worldbank.org/country4) Most positive tendencies that accompanied the programme are in the aspect of taming the inflation in the region (Fig. 5).Despite the existing internal challenges, all six states (beside Belarus) managed to harmonize their inflation rates and bring it close to the European standard, at least the volatility of rates kept quasi-equal.The ENP Multilateral Platforms (European Commission Memorandum, 2012) are considered in terms of the four main directions: • Platform 1 -"Democracy, Good Governance and Stability" • Platform 2 -"Economic Integration and Convergence with EU Policies" • Platform 3 -"Energy Security" • Platform 4 -"Contacts among people".Our next step is to trace the dynamics of the main world-known representative indexes for these six targeted states, which we believe can reflect each of the platform ideas: •  Thus, in the mirror of the representative indices (according to the four highlighted platforms) for the period 2008-2014, we can see a clear tendency to improve for Armenia and Georgia after the programme platforms have come into force.
Thus, the statistical and analytical analysis of the performance indexes and data of the main targeted objectives of the EU project in the direction to integrate / close six targeted eastern neighbor states to the EU standards and values provides the evidence that the programme is unbalanced.
The next step is implemented by us to depict the understanding of internal nets and levers for the development and integration of the six states to the EU.The usage of powerful statistical tools for the evaluation of statistical relationships, involving dependence, is supposed as most appropriate for this aim.We try to indicate by means of the correlation analysis the predictive relationships that can be exploited in practice -to trace economic prospects in the context of growing regional interdependencies for the EU and the eastern partnership of the six states.Note that a correlation coefficient is a measure of the strength and direction of the linear relationship between the two variables, which is defined as the (sample) covariance of the variables divided by the product of their (sample) standard deviations (Green, 1993).The correlation analysis cannot be interpreted as establishing cause-and-effect relationships.The correlation coefficient measures only the degree of linear association between the two variables.It can indicate only how, or to what extent, the variables are associated with each other; this is appropriate to reach the declared goal of the paper -does the macroeconomic level of the six states associate in its dynamics with the EU?The programme efficiency is based to some extent on reaching this result.
The large data base was used for the research: 2157 statistical data on 1960-2013 for the EU and the six targeted states in the cutaway of such indicators that represent 14 main tendencies of the states' development as it is considered below: Our article has intended to show that there are a lot of relationship analyses but quite nothing is said (as the literature analyses show) on the mathematical point of view on the topic.Thus, we used a correlation that refers to any of a broad class of statistical relationships involving dependence.We would like to supplement the known thoughts and opinions by the objective results of the mathematical approach as a sober look at the problem.A correlation analysis proves at the significance level of 0.05 that there are some associated relationships (Table 4) for the analyzed period (on the average).The most unexpected result of Table 4 is the fact that the EU variables are in a minor correlation with the six states' indicators.Thus, we interpret this as the evidence that in general during the analyzed period the EU and the targeted group were not in convergence.The country that has appeared to be most synergetic to the EU is Ukraine.It is quite an unexpected result, as for the programme to be successful we should consider rather high values of correlation at least for the main indicators of the EU and targeted states.
Despite compiling Table 4 on the correlation matrix analogue, we do not consider the direction of the impact as it is unclear in the capacity of the correlation analysis.The correlation coefficient just shows us the density and effective communication among the factor variables for their linear dependence.By means of correlation we can detect only a strong interdependence in the time series of representative indexes, but we cannot be deep as to the nature of such dependency.The direction of the dependency remains unclear.The causes preceding the correlation, if any, may be indirect and unknown.High correlations also overlap with identity relations (tautologies) where no causal process exists.For depicting the main causes and sequences in tendencies in the analysis, we propose to use the Granger causality test.We have pushed off the assumptions that the correlation does not necessarily imply causation in any meaningful sense of the word.The econometric graveyard is full of magnificent correlations which are simply spurious or meaningless.The Granger approach (1969) to the question of whether X (independent variable) causes Y (depended variable) is to see how much of the current Y can be explained by the past values of Y and then to see whether adding the lagged values of X can improve the explanation (Green, 1993).This approach helps us to understand which main development indicator and of which state can cause the integration / development tendencies and can be the best indicator of its happening.Before the application of the Granger test we had clarified each of the time-series to determine their order of integration -involved a test (such as the ADF test) for which the null hypothesis is non-stationarity.The implementation of the Granger causality test in EViews provided us with the following resulting claims (at the appropriate level of F-stat) about link directions for considered data and states (Annex 2): 1) as to GDP growth: there is the mutual Granger causality for the six targeted states and the EU; besides, there are one-way directions for: AZE_GDPGR → EUU_GDPGR, ARM_GDPGR→ AZE_GDPGR, ARM_GDPGR→BLR_GDP-GR, AZE_GDPGR → UKR_GDPGR, MDA_GDPGR → UKR_GDPGR, GEO_ GDPGR → UKR_GDPGR, GEO_GDPGR → BLR_GDPGR, GEO_GDPGR → MDA_GDPGR; 2) as to GDP growth per capita, the case is a bit similar, but there are mutual causality pairs for Armenia and Azerbaijan, and Moldova vs Ukraine; however, there are one-way directionsfor AZE_GDPGRPC → GEO_GDPGRPC, AZE_GDP-GRPC → BLR_GDPGRPC; 3) as to the GNI growth, we can considerate the mutual Granger causality only for Ukraine and the EU, as to the other five states there is only one-way Granger causality from the EU to a state; 4) as to GNI per capita growth, we consider the mutual Granger causality for the six states and the EU in the analyzed period; 5) in the aspect of the most representative indicator of security -military expenditure (% of GDP), we have detected a mutual causality for the EU with Azerbaijan and Belarus and the one-way run from the EU to Armenia, Moldova, Georgia.As to Ukraine, we received a statistically insignificant result; 6) as to the gross national expenditure (% of GDP), we can consider the existence of the mutual Granger causality between the EU and the targeted states; 7) as to exports of goods and services (% of GDP), the result is quite diverse, thus, there is mutual causality on Granger for the EU and Armenia, Azerbaijan and Ukraine.But there is one way from the EU to Moldova, Georgia and Belarus, and not the other way; 8) as to trade (% of GDP), there is the mutual Granger causality for the EU and targeted states -that is a very exciting result and litmus that the Free Trade Action works properly; however, there is one-way run from Belarus to the EU, by which some period in the international status of Belarus can be explained; 9) as to the unemployment indicator, it demonstrates quite predictable results: there is the mutual Granger causality for the EU and Moldova, Armenia, but the one-way: AZE → EU, BLR → EU, GEO → EU, and for Ukraine one way from the EU.Note that we considered the following indicators only for the interregional level, because these indicators demonstrate exclusively the internal process and the way of emerging the targeted state: 1) as to the current account balance as the % to GDP, the mutual Granger causality was detected among all the targeted states; 2) as to a short-term debt (% of total reserves), for the analyzed states we saw the one-way Granger causality for pairs: GEO → ARM, ARM → AZE, ARM → UKR, MDA → GEO, UKR → GEO, BLR → UKR; 3) as to the real interest rate, it is the mutually Granger causal besides one-way for MDA → ARM, MDA → GEO, UKR → MDA, BLR → GEO, AZE → GEO; 4) in the level of poverty, the states demonstrate the full mutual Granger causality; 5) as to the causality in minimizing the inflation rates, we detect mutual causality for most of the combinations of states in the target group but a one-way run for UKR → ARM, AZE → ARM, ARM → MDA, UKR → MDA, AZE → MDA; (i.e. the result "Null hypothesis Probability GEO_EXP does not Granger Cause EUU_EXP 0.6389 EUU_EXP does not Granger Cause GEO_EXP 0.0022 says us that having such probability values we cannot reject the hypothesis that the GEO_EXP does not Granger cause EUU_EXP, but we do reject the hypothesis that EUU_EXP does not Granger cause GEO_EXP.Therefore, it appears that the Granger causality runs one-way from EUU_EXP to GEO_EXP and not the other way.) Not venturing in the causes and sources of found results that are quite clear and repeat the known agenda, we can conclude that mostly in trade and social aspects the programme works rather optimistically.However, the direction of macroeconomic growth and security requires enhancing the actions.Also, the group is not yet homogeneous.The economic position of the states has been quite diverse and unequal at the starting point.This gives no hope for the further smooth and efficient parallel integration of the six states to the EU.Quite a definite proposal is to separate the states in this policy and provide for a unique action, specific for each state.Such actions can bring a more expected positive result for the EU.

Conclusions and discussion
The Eastern Neighborhood European Partnership seems to be an up-to-date and necessary objective aimed at: promoting democracy and good governance; strengthening energy security; promoting sector reform and environment protection; encouraging people-to-people contacts; supporting the economic and social development; providing an additional funding for projects to reduce social inequality and to increase stability; implementation of the Integrated Border Management Programme, as well as the SME Flagship Initiative; defending the regional energy markets and energy efficiency besides the diversification of energy supply (like the Southern Energy Corridor); common prevention of, preparedness for, and response to natural and man-made disasters.However, this idea could be considered as fully political and standing on a very fragile economic basis and no socio-economic reasons for choosing the six targeted states.The targeted six Eastern neighborhood states appeared to be quite diverse in the statistical sense of their economic development.A mathematically based research proved a rather high diversity of inter-state tendencies and development trends before and after the EU programme launching, which can minimize or neglect / eliminate all the EU attempts involved to spread the EU policy and standards through its borders to the centers of interest.
One can argue that the research methods used in the article are not sufficient as it is not enough to draw pictures and, based on the difference in trends of the GDP and similar macroeconomic indicators for the EU and the analyzed countries to conclude that the EU funds are not efficiently distributed.One can say that such difference could be explained by many reasons (internal and external); besides, the period (2009-2013) was very heterogeneous and cannot be treated not mostly by the EU funds.This is the attitude we try to overcome.First of all, for the analysis we have used only the classic methods that are always instruments for any economic analysis.Our research sets a goal to highlight the problem (it is a novel idea as the literature review proves), to make the first steps in indicating the possible reasons why the EU is still concerned with slow reforms despite the large funding of the targeted states.We avoided repeating the political reasons that are widely known.These reasons are explained, funds are delivered, but their effectiveness is not as high as expected.The demonstration, the most obvious proof of it is exactly the dynamics and convergence of macroeconomic trends, the fact that the targeted states have experienced this heterogeneous period, that they are still heterogeneous in most of the main indicators.However, having an efficient absorption of the incoming funds should maintain the resistant and robus economies in the states that have much less reacted to the external and internal challenges and, according to the synergetic law, mirror the dynamics of the EU economy.
One can argue as well that we have not consider one exact factor -the EU funding -in our research.Our argumentation is that we considered it as an umbrella, as a climate in which countries developed during the period.Also, the analyzed indicators are considered according to the main platforms the funding was targeted, thus they can be a reflection of the funding effectiveness in the region.
The findings that contribute to the literature are that the paper highlights the fact that, despite emerging from the same system, the six states of the Eastern partnership are different in their way of transformation and development, which is the main reason for the EU in its policies to consider the states as unique subjects of the programmes but not to apply a universal approach that can have a great probability to fail in its effectiveness.However, in the aspects of trade we have received the evidence of positive results of the EU actions.

Fig. 1 .FIG. 2 .
Fig. 1.Dynamics of GDP growth (annual %) in the EU and the targeted states (1961-2014) Source: author's calculations on the base of the World Bank data http://data.worldbank.org/country FIG. 5. Dynamics of the inflation rate, consumer prices (annual %) in the EU and the targeted states (2005-2013) Source: author's compilation on the base of the World Bank data.http://data.worldbank.org/country

TABLE 4 . Significant high correlations among the states' indicators * (we avoided intentionally the cor- relations among indicators of the same state)
ukr (-), InflSource: author's calculations and compilation.* (-) -means the opposite direction of indicators, as increasing one indicator can be accompanied with decreasing another factor (in linear dependence).

Granger analysis result: test on causality for 6ENP and the EU time series, 1960-2013
Source: calculated and compiled by authors.