Regional Potential of Armenia for SMEs Clustering: an Empirical Evidence

. The process of clustering of small and medium-sized enterprises (SMEs) is complex and multidimensional, as well as risky and costly. The purpose of this study is to empirically substantiate the presence of regional potential for clustering of SMEs in the Republic of Armenia (RA) by evaluating the most appropriate set of indicators from the available variety. The quantitative analysis encompasses the following measures: Kim’s Location Quotient, Krugman’s Concentration Quotient, shift-share indicator with quotient of national shift, industrial mix and regional shift, structural shift along with the Gini’s Localization Quotient. The data were obtained from the database of the Statistical Committee of the Republic of Armenia (SCRA), which, however, did not fully encompass a detailed knowledge base concerning all economic sectors of the country’s regions. This served as a limitation for the study. In addition, there exist no similar quantitative studies on the problems of regional diagnostics of Armenia in the scientific literature. The study made it possible to identify promising regions for SMEs clustering in RA. The obtained results can be applied in future for a more specific study of opportunities of forming a supercluster that will unite the potential of several regions of RA at once.


Identification of Regional Potential for SMEs Clustering
Starting from the period of formation of early clusters and until present, theorists and practitioners of economics and management face the problem of identifying the most promising regions for the formation of cluster agglomerations [1,2,3]. Today there is no universal approach, the application of which can foster the optimal identification of regional potential for SMEs clustering. The process of measurement is complicated by the fact that each country has its own unique business structure. In this sense, RA is no exception: in fact, despite the forecasts of analysts and draft projects of state programs for regional development, not a single SMEs cluster operates in the country.

Approaches to Regional Potential Identification for SMEs Clustering and Their Relevance in RA
There exists a variety of quantitative and qualitative tools that can be used to identify regional potential for cluster formation that can boost a country's economic performance [4]. Some of the tools are only applicable in countries with successful experience in creating SMEs clusters: this is due to the fact that calculations require historical data that are used in complex extrapolation forecasts [5,6]. The purpose of this study is to reflect several methods of identification of regional potential for SMEs clustering on the realities of RA and to reveal the most promising regions for cluster formation. To achieve this goal, the following indicators were calculated: Kim's Location Quotient, Krugman's Concentration Quotient, shift-share indicator with quotients of national shift, industrial mix and regional shift, structural shift and Gini's Quotient.
The article includes: Section 2 with research methodology; Section 3 with study results; Section 4 results discussion and Section 5 with final conclusions.

Materials and methods
The lack of studies on identification of RA's regional potential on the one hand complicates the task of choosing necessary indicators, but on the other hand allows to select a unique number of methods from a variety of alternatives and to conduct the regional analysis. Thus, a number of indicators were taken as the basis for calculations, namely: • Kim's Location Quotient (LQ) [7], which measures the level of entrepreneurial specialization of a region in a particular industry; • Krugman's Concentration Quotient (CQ) [8], which allows to determine the level of concentration of enterprises in economic sectors in specific territories; • Shift-share analysis (ZP) [9], which examines regional economic growth from the standpoint of the sum of the combined effect of three components: growth at the national level (ZN), growth in the industry structure -industrial mix (ZO) and growth due to other (local) factors -regional shift (ZR) [10]; • Gini's Localization Quotient [11].
The above-listed indicators were considered in detail and reflected in complex researches [12,13,14]. This analysis encompasses data from the Statistical Committee of the Republic of Armenia [15]. It should be noted that official sources do not show a sufficiently detailed sectoral presentation of statistical data. Thus the categories are roughly divided into two groupsagricultural (AG) and non-agricultural (NAG).

Results
It is estimated that the growth of employment in AG sector for the period of 2015-2020 decreased by about 40%, while the growth in NAG sector increased by almost 19% (Table 1). The total level of specialization of regions in AG prevails over the LQ of NAG. This is evidenced by the LQ values, which must exceed 1.25 to consider the specialization of the region as high. These are the results for AG for the following regions: Aragatsotn, Ararat, Armavir, Gegharkunik, Shirak, three of which (Aragatsotn, Syunik, Shirak) in the period of 2019-2020 showed positive LQ growth. In these three regions the share of employment in agricultural sector is higher than the national average. In a number of regions (Vayots Dzor, Kotayk, Lori and Tavush) LQ values were higher than 1, while Vayots Dzor and Kotayk regions showed positive growth. Calculation of CQ and Cumulative CQ enabled to find trends in industry diversity ( Table 2). It is known that a high level of CQ (which is in the range from 0 to 2) symbolizes a high indicator of concentration in regions and in the country as a whole. The cumulative CQ in the AG sector exceeds those in the NAG sector, but in both cases there is a negative increase in 2020 compared to 2019. However, it can be seen that the regions of Armavir (in AG sector) and Kotayk (in NAG sector) stand out with relatively high CQ values.
Based on the data presented in Table 3, it can be stated that, on average, in RA in the period of 2018-2020, there has been an increase in the level of employment by 1.16%, which is evidenced by the value of the national shift (ZN). The component of the industrial mix (ZO) valued constantly negative for the AG sector (-0.14%) and a constantly positive (but not high enough) for NAG sector (0.05%), which indicates a disproportionate level of employment in country. The regional shift (ZR) shows regional growth that is different from the national one, which in turn makes it possible to identify leading and lagging industries. For the same regions, the structural shift (ZP) exceeds 1, but the regions of Ararat and Gegharkunik did not show high results.
As can be seen from Table 4, the largest spatial concentration of firms is in the AG sector of the following regions: Armavir, Ararat and Shirak, as well as in the NAG sector of Kotayk. Despite this, it can be stated that the Gini's Quotient has not shown high results in any region. Only in the AG sector of Armavir the value of the coefficient equal to 0.255 showed a relatively good result considering the highest reference value equal to 0.5. No region demonstrated a complete absence of firm concentration (value equal to 0). Table 5 reflects the final ranking of regions by estimated indicators.

Discussion
The study revealed that there is no region in RA that demonstrates positive results in all estimated indicators. Although there exist unfinished projects and studies on SMEs clustering in the regions of the RA ("USAID's Armenian Tourism Cluster Strategic Action Plan of 2007"; "National Competitiveness Foundation of Armenia's Strategic Project of the Southern Corridor of Tourism of Armenia of 2011") [16] and was also proposed to create a cluster of recreational and sanatorium services in the Vayots Dzor region [17,18], the assumptions have not been confirmed based on quantitative analyses.

Conclusion
This study reflected the attempt of application of well-known methods for determining the regional potential for SMEs clustering on the realities of RA, identifying the most promising regions, for which estimation of a variety of indicators was carried out. The accuracy of the results was hampered by the lack of detailed regional statistics for each of the country's key industries. An additional difficulty arose due to the fact that the research literature does not cover quantitative research in the field of SMEs clustering in RA.
Despite the presence of the identified potential of the regions of Aragatsotn, Armavir, Syunik and Shirak in the AG sector, as well as the regions of Kotayk and Vayots Dzor in NAG sector, there is a need to set relationships between SMEs located in those territories, as well as to determine the role and level of influence of state bodies and scientific communities. It would be advisable to consider the possibility of forming a cluster that combines the potential of several regions, taking into account the cross-sectoral cooperation of regions.