**1. Introduction**

Depending on the methodology applied, rural areas comprise about 80% of land in Serbia, and contain between 40% and 50% of the population, which indicates the specific relevance of these areas for the overall Serbian economy. The development of local infrastructure and basic services in rural areas, including leisure and culture services, the renewal of villages and activities aimed at the restoration and upgrading of the cultural and natural heritage of villages is an essential element for the socioeconomic development of rural areas [1]. Rural regions are mainly based on sectors that use natural resources such as agriculture, forestry, fishing, oil, gas, and electricity. Therefore, competitiveness of the primary sector remains the policy focus for developing these areas. In more developed countries, tourism and services, or renewable energy production are associated sectors that also rely on natural resources and make significant contributions to the socioeconomic development of rural areas. The strength of the link between agriculture and other sectors is influenced by various factors such as the natural characteristics of the terrain (land quality, climate, and local tourist attractions), infrastructure, the overall strength of the national economy, the educational level and entrepreneurial potential among the local population, and access to public finances [2]. Namely, in rural areas with demographic problems, rural tourism could be an additional activity that could make the traditional agricultural function of those places secondary, so it could change those areas into multifunctional spaces [3]. Rural areas are usually poorer and less populated than urban areas. Berc et al. [4] indicate that in rural areas poor availability of certain social services is often accompanied by weak coordination of service providers within the social welfare system, which speaks in favour of the present difficulties in the implementation of deinstitutionalization and decentralization of social services.

**Citation:** Jurjevi´c, Ž.; Zeki´c, S.; Ðoki´c,D.; Matkovski, B. Regional Spatial Approach to Differences in Rural Economic Development: Insights from Serbia. *Land* **2021**, *10*, 1211. https://doi.org/10.3390/ land10111211

Academic Editors: Krystyna Kurowska and Cezary Kowalczyk

Received: 30 September 2021 Accepted: 5 November 2021 Published: 8 November 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

The income disparities between rural and urban areas in Serbia have deepened with the process of transforming a centrally planned economy into a market-oriented economy. Following a dynamic transition process, a clear strategic goal for Serbia is integration into the European Union (EU), which requires additional economic and institutional changes. The dynamics of European integration differ among the formerly communist and socialist countries. For Serbia, which is currently a candidate country for EU membership, the experience in pre-accession negotiations of New Member States (NMS), i.e., Central and Eastern Europe (CEE) are a valuable benchmark for future integration processes. Böwer and Turrini [5] investigated the effects of EU accession on NMS, and they concluded that this region has widely benefited from economic and institutional integration with the EU. The socioeconomic growth recorded in NMS after the recovery from transition shock in the early 1990s has been impressive.

The study presented in this paper focuses on Serbia's rural areas, as well as future development strategies for these areas, in order to better prepare them for EU integration. The agricultural sector has a significant position in the overall Serbian economy and even more so in the rural areas, where it is often the dominant activity for most of the population [6]. Based on agri-environmental conditions, the rural areas of Serbia can be divided into two regions: northern and southern. The northern region has exceptional agri-environmental conditions for agricultural development and a high concentration of food industry [7]. The southern part of Serbia is characterized by mountainous areas with relatively poorer conditions for agricultural production. In addition, both regions are characterized by an unfavourable demographic structure with a very low level of education of the rural population. According to Petrovi´c et al. [8], modest knowledge and the absence of supplementary skills in the rural population are limitations for the total capacity and competitiveness of the labour force in rural areas, which can be anticipated as one of the burdening factors in the economic development of these areas.

Creating an appropriate rural policy in Serbia within the conditions of current European integration requires harmonization with the EU's Common Agricultural Policy (CAP), which is one of Serbia's future priorities. The CAP represents the benchmark for future policy, so pressures of the EU accession negotiations, as well as EU pre-accession support will be the key elements in the process of adapting policies to the CAP [9]. However, harmonization of legislation, institutional capacity building, and policy reform in agriculture and rural development are complex issues. Therefore, economic, social, political, and environmental conditions must be taken into account when defining political measures and instruments. Namely, as space and land are limited resources, an era of rapid urbanization should be effectively controlled in line with sustainable development principles [10]. The principles of sustainable development integrate political, economic, and social measures in order to meet the needs of communities without compromising the ability of future generations to meet their needs [11].

The European Union offers various opportunities for receiving financial support, which allows the exchange of best practices whereby the benefit is realized by Serbia, and even certain regions that have adequate cooperation with EU countries, such as Vojvodina [12]. The European Commission [13] estimated that good progress was made by adopting the action plan for acquis alignment in agriculture and rural development and implementing the Instrument for Pre-accession Assistance for Rural Development program (IPARD II). Although it is very important to access these funds, primarily because of the relative significance of agriculture for the overall economy as well as Serbia's rural areas, it must also be emphasized that pre-accession assistance does not solve the key development problems of rural areas, which require a comprehensive, long-term, territorial-based national policy that respects local specificities and the needs of the rural population [14]. Sustainable development of agriculture entails sustained economic growth, technological advancement, efficient resource management, and an increase in quality of life in rural areas [15].

In order to monitor and compare the socioeconomic conditions of different heterogeneous territories across EU countries, the Nomenclature of Statistical Territorial Units (NUTS) was adopted by the EU. The NUTS classification is a framework for determining standardized statistics of all EU Member States at three basic levels: NUTS 1 (population of 3 to 7 million), NUTS 2 (population of 800,000 to 3 million) and NUTS 3 (population of 150,000 to 800,000) [16]. The Nomenclature of Statistical Territorial Units has also been defined for EU candidate countries (Serbia, North Macedonia, Albania, Montenegro, and Turkey), which allows for comparison of regions or districts across Europe. In Serbia, this classification is the basis used to draft documents needed to implement projects that should be financed by the European Union's structural funds [17].

The study presented here is designed to help candidates for membership determine the position of their regions vis-à-vis EU regions and as such can be applied to other candidate countries in addition to Serbia: Montenegro, Albania, North Macedonia, and Turkey. Moreover, this study will be a step forward in comparison with the existing literature, given the minimal amount of research conducted at the regional level in Serbia. The results provide an empirical basis for creating future rural development strategies for Serbia by giving a detailed insight into socioeconomic performance at the regional level, and enabling a comparative analysis with EU countries as well as with regions within Serbia itself. The methodology adopted in this way, and when applied to other candidate countries, would provide an overview of socioeconomic performance across Europe. Identifying a candidate country's level of development in relation to the Member States is important for the harmonization of policies, such as Serbia's rural policy and the EU CAP or regional policies. This indicates the practical contribution this study can provide. The purpose of this study is to determine the socioeconomic performance of rural regions of Serbia and the EU in order to indicate the position of Serbia's rural areas in the process of European integration. More specifically, the aim is to detail the socioeconomic performance of rural regions, which will be evaluated with an Index of Socioeconomic Performance evaluated by Factor Analysis (FA).

Based on the purpose of this paper, the main hypothesis of the research is created:

•*The socioeconomic performance of Serbian rural regions corresponds to the socioeconomic performance of rural regions of NMS.*

### **2. Theoretical Background**

Regional development plays a significant role in the EU. The regional aspect has been given more importance in the EU, primarily through the Cohesion policy, i.e., strengthening the economic, social, and territorial cohesion within the EU. One of the EU's key objectives is to reduce development inequalities between developed and economically underdeveloped regions. EU enlargement to the south, and especially to the east, has been followed by growing inequalities within the Union. The more developed, pre-2004 member states (EU-15) channelled financial resources through the Cohesion policy to less developed NMS to support transformation and economic convergence [18]. The EU Cohesion policy does not include only economic convergence, but it is certainly still the most important objective of this policy due to large income discrepancies. In recognizing a need to assess more place-sensitive policies, highlighting heterogeneity generally contributes to the debate on the future of the post-2020 Cohesion Policy, by providing effective comparative tools to support new policy instruments [19]. In addition to this policy, others, namely the EU rural development policy, have had significant impacts on the regional development of the EU as a whole [20].

The EU Cohesion policy plays an important role in supporting the socioeconomic development of rural areas and, together with the European Agricultural Fund for Rural Development (EAFRD), involves directing financial resources towards the reconstructing and revitalizing these areas. EAFRD is part of the EU's CAP, but with a regional focus [21]. Matthews [22] states that one of the general objectives of the CAP for rural development in the upcoming period (2021–2027) is to "strengthen socio-economic performance in

rural areas" through specific objectives such as attracting and retaining young farmers in rural areas; promoting employment, social inclusion and local development through increased bioenergy production and sustainable forestry; an adequate response to the increased demand for health-safe foods; and the use of innovation and digitalization for both agriculture and rural areas. Increased demand for renewable energy can be a good development opportunity for rural areas. Vukadinovi´c and Ješi´c [23] point out that creating "green jobs" through the concept of a circular economy is important for employment growth, given that economic growth is becoming an effective use of resources and renewable energy resources, as well as the use of comparative advantages of the natural environment. However, the heterogeneity of rural areas or regions significantly impedes the convergence process. The objectives of rural policy have become multidimensional and focus on increasing the wellbeing of rural residents. Generally speaking, quality of life has several dimensions: (1) an economic dimension, in which the income of the population depends on being able to find employment in companies that are productive and competitive; (2) the social dimension, which refers to accessibility to services; and (3) a local dimension, which refers to the environment [2]. Although it is necessary examine rural areas according to several aspects, the socioeconomic aspect is always an important link to future development. The European development model is characterized by balancing economic and social performance, and quality of life, as a top European priority [24]. In most of the rural typologies, the structure of employment by sectors was analysed in order to define the role and significance of agriculture and other sectors within the rural area. Moreover, the importance of the sector is determined by their share in the Gross Value Added (GVA) of the region. The employment structure is also important for the region's socioeconomic development. The traditional approach to identifying regional competitiveness is based on an analysis of GDP per capita [25]. Michalek and Zarnekow [26] pointed to the use of GDP per capita (calculated at NUTS-2 or NUTS-3 level) as: (a) a standard measure of a regional level of welfare; (b) a basic criterion of eligibility criteria for EU funding under structural funds, and (c) the main quantitative indicator of the effectiveness of the policies being pursued. Moreover, Prus et al. [27] have used a significantly larger number of variables to determine the socioeconomic characteristics of certain regions.

Domazet et al. [28] indicate the importance of following up macrocompetitiveness of the EU, or countries around the world by the European Commission in its European Competitiveness Report, which examines the basic performance of the competitiveness of the EU as a whole, member states, or certain economic activities, while the WEF (World Economic Forum) affirmed the GCI Index (Global Competitiveness Index) for following up basic indicators of the competitiveness of countries around the world. Regional competitiveness is the ability of a region to offer an environment attractive and sustainable for businesses and in which residents can live and work [29]. As noted, each round of EU enlargement deepened regional differences. Here, the focus will be on the EU enlargement of 2004, since expansion from that period on included countries with the same historical legacy of centrally planned economies as Serbia (with the exception of two island countries, Cyprus and Malta). The countries of Central Europe, and Eastern Europe in particular, are considered to be less economically developed regions in comparison to the original EU member states (for example Benelux, Germany, and France) due to the strategy of socialism, i.e., industrialization which led to economic, social, and environmental decline. Accordingly, to better facilitate accession for the CEE countries, two EU programs were launched at the end of the 1990s, which strongly shaped the regional policies of the CEE countries: the Instrument for Structural Policies for Pre-Accession (ISPA) and the Special Accession Program for Agriculture and Rural Development (SAPARD), in order to prepare future members for Cohesion Policy (first fund) and for the EU CAP (second fund) [30]. Bachtler and Ferry [31] point to the importance of using these funds when CEE countries join the EU, as well as to different strategies in spending structural funds in these countries, whichhavefurtheraffectedregionalinequalities.

Rural development largely depends not only on national policies (rural, regional, social, etc.) but also on factors that influence heterogeneity, with future development based on addressing specific problems affecting a particular territory [32]. The development of regions, or the convergence of less developed, usually rural regions with developed regions, creates a need to territorialize and regionalize development policies, while also seeking competitive advantages for localities [33]. Resolving the issue of rural development necessitates an integrated approach that requires cross-sectoral cooperation at all levels (national, regional, and local). Issues related to rural policy in Serbia are reflected not only in the low level of funds allocated for rural development but also in the defined measures, which are directed more towards investing in agricultural production itself rather than in the development of infrastructure in rural areas, the environment, or improving quality of life in rural areas [34]. Limited human resources, lack of regulatory framework and funding, and insufficient experience in both policy formulation and major project managemen<sup>t</sup> are the main obstacles to effective rural development policies in Serbia, the most important task of which should be strengthening local self-government capacity [35]. In the EU, the LEADER (an acronym for the French Liaison Entre Actions de Développement de l'Economie Rurale) program supports capacity-building of local authorities and the development of local partnerships, and emphasizes the importance of rural development projects launched at the local level to revitalize rural localities [36]. The involvement of Local Action Groups (LAGs) in the decision-making and priority-setting process for local territorial development, i.e., a bottomup approach, is a significant segmen<sup>t</sup> of EU rural development policy. However, the process of regionalization and decentralization in Serbia is insufficient; the distribution of responsibilities is asymmetric at different territorial levels, which will be unsustainable in the future. Although today we are in the Fourth Industrial Revolution with significant social transformation in parallel with technological change [37], rural areas lag significantly behind urban areas, with limited access to technology, information, and new knowledge.

### **3. Materials and Methods**

In this research, the methods of multivariate statistical analysis were used, i.e., Factor Analysis, which aims to reduce large numbers of variables to a more manageable number while discarding a minimum amount of useful information. The advantage of FA is that it enables researchers to take an important step towards deeper understanding of a complex and multidimensional territory such as, in this case, rural areas [38]. Moreover, the advantage of using factorial techniques is that no prejudgment of the results is required, as the technique itself determines the importance of individual factors (dimensions) within any solution derived from it [39]. The conditions required for the FA to be applied were checked by Bartlett's test and by Kaiser–Meyer–Olkin (KMO) sampling adequacy testing [40]. The KMO value is a measure of adequacy of the correlation matrix to perform the FA. The KMO test ranges from 0.0 to 1.0, but values should be greater than 0.5.

Factor loadings represent the correlation between the original variables and the factors and are key to understanding the nature of a particular factor. When using practical significance as the criteria, Hair et al. [40] suggested that factor loadings in range ±0.50 or greater are considered practically significant. Interpretation of factors, based on factor loadings, is an important step. If it is necessary, factor rotation should be performed. The goal of VARIMAX rotation is to maximize the variance of factor loadings by making high loadings higher and low ones lower for each factor [41]. After factor extraction, it is necessary to calculate the factor scores for each unit of observation for each factor. Factor scores are standardized to have a mean of 0 and a standard deviation of 1. Factor score calculations enable creation of an index for each factor so that observation units can be ranked. There are also some limitations and disadvantages of the method used in this analysis. For example, Cloke and Edwards [42] admit that multivariate classification techniques in general are subject to considerable methodological disagreement and that

the validity of individual classifications rest on their usefulness far more than on their methodological basis.

More systematically, this research will follow certain statistical assumptions and procedures when conducting FA:


In this analysis, NUTS 3 were selected as the observation units for both the EU countries (with the exception of Germany, for which NUTS 2 were used) and Serbia. Despite criticism, for example, Hedlund [43] pointed to need for typologies based on high-resolution data, beyond the urban–rural continuum, respectively the administrative boundary, this level was chosen for two reasons. First, it represented the lowest administrative and territorial level at which data could be found for all countries included in the analysis; second, regional typologies applied at this level aimed to analyse and monitor rural and urban development by implementing certain regional and rural policies and programs. NUTS 3 allowed a detailed representation of EU rural space [44]. It is worth noting that the degree of differentiation among European rural regions is in line with the transitional processes described in the literature [45], which is especially significant for former socialist states, both in the EU and in those, such as Serbia, which are candidates for membership. Spatial (i.e., territorial) distribution of regional performance was considered: different components followed different territorial paths across Europe, suggesting the existence of a puzzled core–periphery pattern, where within-region differences also matter [46].

Since this study focuses on rural regions, predominantly urban regions were excluded as defined by Tercet (Regulation (EU) 2017/2391) [47], i.e., the EU's Urban–rural typology, because they have urban centres with over 500,000 inhabitants, and they contain at least 25% of these regions' populations. Instead, the focus was primarily on predominantly rural and intermediate regions. These two groups of areas were defined as "non-urban" areas [48]. Certain limitations to this approach should also be noted. Primarily, intermediate regions were of particular concern, since they have a wide range of different spatial characteristics. However, the inclusion of intermediate regions in the analysis was justified by the need to consider as large a geographical area as possible, as well as by the need to include most of Serbia in the analysis, which, according to the Urban–rural classification of the EU, was designated as a state with one predominantly urban region (Belgrade District), five predominantly rural regions, and 19 intermediate regions. A total of 691 units were included in the analysis, of which 667 were at NUTS 3 and 24 at NUTS 2. Certain areas, although classified as intermediate or predominantly rural, were excluded from the analysis due to lack of data (primarily for the newly created NUTS 3 areas), or due being located geographically outside of the European continent.

The Eurostat database [49–51] was used for this study, and the time period was a seven-year average from 2012 to 2018, with some exceptions for France and Poland (threeyear average from 2014 to 2016). The analysed period also coincided with the period of candidacy for Serbia's EU membership (from 2012 to the most recent data available). The observation units in this paper were all EU countries and Serbia. The Statistical Package for the Social Sciences program-SPSS Statistics 20.0 was used for the purposes of this paper. Variables used to create the regional Index of Socioeconomic Performances, using FA, were: share of employees in the primary sector in the total number of employees (%) (EMPL\_PRIMARY); gross domestic product (GDP) per capita (purchasing power

standard-PPS) (GDP\_PER\_CAPITA); primary sector share in total gross value added (GVA) (%) (GVA\_PRIMARY); total labour productivity (total GVA of all activities per employee) (EUR/person) (LABOUR\_TOTAL); and labour productivity in the primary sector (GVA of the primary sector per employee in the primary sector) (EUR/person) (LABOUR\_PRIMARY). The selection of variables was determined by the availability of data in the database used. Bearing in mind that Serbia is a candidate country for EU membership, the choice of data in the Eurostat database is scarce. Additionally, according to previous research, the selected variables well describe the socioeconomic performance of rural areas, which is the main subject of this analysis.
