Counterbalancing the Development Territorial Disparities in the Implementation of the Community-Led Local Development EU Approach in Romania

Abstract

LEADER is a rural development method based on a participative approach, which was tailored in 1991 as a complement to the traditional common agricultural policy (CAP) measures. One of its most important objectives is to reduce the differences between rural and urban areas by building on local knowledge and potential. The aim of the present paper is to identify what are the most important characteristics of the LAGs that can counterbalance the existing economic disparities in the rural regions. The research was conducted in the northwest development region of Romania (2014–2020 programming period), using the principal component analysis and the hierarchical cluster analysis. Two types of data were collected: indicators of performance, such as the number of projects contracted and jobs created, were used to assess the success of the method, while the territorial and LAG characteristics were used to explain these results. The findings confirm the presence of an unequal distribution of LEADER support in favor of the most urbanized and developed areas. However, the results also show that the experience and economic and administrative capacity of LAGs could help counterbalance the influence of the territorial features previously mentioned, and therefore to reduce the gap between them and the more developed groups.

1. Introduction

Rural areas cover 90% of the European Union’s (EU) territory and host more than half of its population. For decades, those areas have been facing some important socio-economic and demographic challenges, such as high unemployment, population decline, migration, low levels of income, and high dependency ratios on the agricultural sector [1,2,3]. For those reasons, rural development became one of the most important subjects of the EU’s common agricultural policy (CAP), with a strong focus on the multi-functionality of agricultural activities, economic diversification, innovation and knowledge transfer, and the resilience of cultural and environmental heritage [1]. However, as Alonso and Masot [1] argue, rural areas are struggling to achieve sustainable development using only the traditional rural development CAP’s measures developed in national or regional top-down rural development programs (RDPs). They point toward a territorial imbalance in the distribution of economic resources, as the more developed municipalities, with an already existing business infrastructure can attract more investment funds compared to the more economically disadvantaged areas. Thus, in the early 1990s, the EU started to look for alternative solutions in the form of participative approaches and in the development of social capital as driving forces for the revitalization of rural communities [4]. One such solution was LEADER (“Liaison Entre Actions de Développement de l’Économie Rurale”), which became one of the most important alternative bottom-up rural development methods used in the EU in the past three decades. The approach was firstly introduced in 1991 to improve the potential of rural areas by drawing on local knowledge, initiative, and skills [5]. In its first stage (LEADER I, 1991–1994), the method was implemented in 217 pilot rural regions, while in the second edition (LEADER II, 1994–1999), the number already grew to over 900 areas. Due to its success, the method was further expanded under LEADER+ (2000–2006) to cover all types of rural areas [5]. Starting with its fourth edition, the method was financed through the European Agricultural Fund for Rural Development (EAFRD) and became a mandatory component of all rural development programs (RDP) [6]. In the most recent programming period (2014–2020), LEADER was extended under the broader concept of community-led local development (CLLD) [7]. CLLD is designed to build on the communities’ strengths by offering them not only long-term funding but also the power to decide how to use it, employing a participative bottom-up approach. This transforms the local population from passive beneficiaries into genuine drivers of local development [8]. The method achieves this by enabling the local stakeholders to come together and create area-based local development strategies (LDSs) funded through the LEADER program. Thus, these multi-actor partnerships, called local action groups (LAGs), are responsible for implementing the LDSs, and therefore the method, at the local level [9,10].

From an economic point of view, LEADER receives less funding compared to the more traditional programs; thus, LAGs must concentrate on more innovative projects that help to improve the living conditions in rural areas [11]. However, those projects are more complex and harder to implement, being dependent on the existence of strong and experienced stakeholder networks in their territories [12]. The literature shows that although LAGs have contributed to the diversification of the rural economy, especially in tourism [6,13], they struggled to create non-agricultural jobs [13,14,15,16]. Contrary to the principles and objectives of the program, the results also show similar tendencies regarding the territorial distribution of LEADER support, as in the case of the more traditional rural development measures developed under the rural development plans. The more economically developed areas managed to attract more funds and create more jobs than the poorer rural and highly remote municipalities [9,11,17,18,19,20]. Concerning this issue, Canete et al. [18] argue that even though LAGs were trying to reduce these territorial imbalances, they might have produced the opposite effect, making them more pronounced.

However, despite the above-mentioned shortcomings, the program has indeed raised the quality of life in rural areas by satisfying social needs and facilitating investments in non-productive but essential projects, such as training programs, innovation support, and knowledge transfer [12,15,16]. The method made it possible to activate and to build on local knowledge [6], to engage the local actors [21], and to create sustainable stakeholder networks [22,23]. Those results, however, are strongly dependent on the experience of LAGs. Konečný [24] shows that the lack of experience (in the case of the newer EU member states) reduced LAGs capacity to successfully implement the method.

Concerns related to the urban–rural dichotomy are also raised in Romania, one of the most rural (43% of the population in 2017) EU member states (MS) [25]. The Romanian rural areas are characterized by demographic decline, a lower life expectancy when compared to the urban areas, and a low level of professional training [26]. Here the local economic development is strongly linked to the proximity to the big urban centers and the connection to the European-roads network [25,27]. Moreover, only the most developed municipalities have the institutional and economic capacity to implement projects using EU programs [25]. Pavel et al. [28] also show that the more isolated rural communities experienced the slowest recovery from the financial crises 2007–2008. Here, LEADER is at its second implementation period. The program is implemented by 239 LAGs that together manage a budget of EUR 563 million, financed through the European Agricultural Fund for Rural Development (EAFRD) [29]. Responding to socio-economic problems such as demographic decline and low levels of income, Romanian LAGs mostly focused on measures from EUs’ Priority P6—“Promoting social inclusion, poverty reduction and economic development in rural areas” and Priority P2—“Enhancing farm viability and competitiveness of all types of agriculture in all regions and promoting innovative farm technologies and the sustainable management of forests” [30]. This was a similar approach to other new EU member states, which have still attributed a major role in rural development to agriculture [24]. An important consequence of this approach was a lack of focus on other important priorities, especially for supporting the shift toward a low-carbon climate-resilient economy (Priority P5) and knowledge transfer and innovation (Priority P1) [30]. Moreover, Olar and Jitea [31] have noted that many of the Romanian LAGs had failed to fully engage in core LEADER activities such as cooperation and networking and were not able to develop their own projects using alternative non-LEADER financial instruments. In this case, their success was mostly linked to their previously existing experience and the size of their team.

Thus, as stated before, LEADER should employ a territorial-based bottom-up approach and act as an alternative to the traditional national RDP, especially in the most economically disadvantaged rural areas. The literature shows that in Western Europe, the success of the program is strongly influenced by the pre-existing level of development of rural areas, remoteness, and experience of LAGs and that the method does not always produce the desired results. Therefore, on the eve of a new programming period, it is important to verify if those aspects are also true in the new EU member states, which have similar territorial characteristics as the regions previously mentioned. Thus, the paper aims to contribute to the literature by identifying the most important characteristics of the LAGs that can counterbalance the existing economic disparities in rural regions. Such results are useful for the successful future implementation of the program. Therefore, three hypotheses were formulated: H1 the implementation of CLLD is strongly dependent on territorial characteristics of LAGs such as the existence of a developed business framework in or near their territories, in line with the mainstream top-down approach critics; H2 the economic and administrative capacities of LAGs (budget, experience, and stakeholders) are important characteristics that are explaining their success. Therefore, they can be used as a method to help counterbalance the influence of the territorial features mentioned at the previous point; H3 LAGs are not fully able to follow and implement the principles of CLLD and to act as a complement to the measures of the national RDP, and thus are not able to respond to the existing economic disparities in the rural regions and demographic problems. The results should offer important lessons and insights regarding community-led local development, indicating if and how the present model needs to be adjusted in the future editions in order to create adequate responses for the demographic and socio-economic challenges in rural areas.

2. Materials and Methods

2.1. Study Area

The research was conducted in the northwest development region of Romania (corresponding to the second level of the European Nomenclature of Territorial Units for Statistics—NUTS 2). The region is characterized by a relatively large cultural and ethnic diversity, with almost a quarter of the population coming from minority ethnic groups. After Romanians, the largest ethnic groups are Hungarians, Romani, and Ukrainians. A total of 31 LAGs are active in the region (with a total budget of EUR 72 million), representing 13% of the national total. Compared to the previous programming period (2007–2013), their territory presents more homogeneous features in terms of relief, climate, demography, socio-economic and cultural environment. The area was chosen due to its similarities to other regions used in previous research [9,11]. Moreover, the region has both deep-rural territories, especially in the Apuseni Mountains, but also relatively big urban centers, serving as a suitable laboratory for the proposed objectives. However, due to the inherited limitations of a NUTS 2 development region, the results cannot be generalized to the entire EU area. Nevertheless, they are still representative for member states that joined the EU after 2004 and especially for Romania.

2.2. Data

Using the methodology established in previous similar studies [1,11], quantitative data related to the 2014–2020 LEADER edition in Romania were collected from official sources such as Romanian Statistical Yearbook [32] and LAGs own websites. Two main types of data were collected. The first set contains key socio-economic, demographic, and territorial indicators, which reflect the characteristics of the region and LAGs, while the second one contains the performance indicators related to the implementation of LDSs. Due to the lack of data availability at the municipality level, all variables were considered and calculated at the LAG level and standardized. This approach comes with some limitations. The distance to the nearest urban centers bigger than 50.000 inhabitants was calculated using the LAG administrative center as a reference, and therefore might not always be representative for all municipalities of the group. All the variables refer to the 2014–2020 programming period.

In total, more than 50 variables were considered for the analysis. A few exploratory principal component analyses were performed in order to reduce their number. The variables with the lowest explanatory factors were subsequently removed. At the end of this process, 16 variables were selected for the final statistical analysis (

Table 1. Descriptive statistics of the selected variables.

Variable Name	Minimum	Maximum	Mean	Std. Deviation
Number. of municipalities (no)	5.00	24.00	12.12	5.04
Population (no)	11,891.00	92,558.00	39,247.13	18,851.29
Operating budget (EUR)	187,230.00	797,652.12	443,957.55	140,094.94
Number of projects (no)	12.00	53.00	30.06	11.51
Rural population/total population (%)	75.88	100.00	91.33	9.72
Number of jobs created (no)	0.00	49.00	17.22	10.84
Jobs created/jobs planned (%)	0.00	433.33	131.45	98.18
Firms/1000 pop	2.30	36.30	6.71	6.13
Young/old population ratio (%) (pop. 0–14 y.o./pop 65+ y.o.) Per 100 inhab.)	28.57	187.23	81.36	35.88
Old-age dependency ratio (%) (pop. 65+ y.o./pop 15–64 y.o.) Per 100 inhab.)	16.69	63.64	36.45	12.54
Distance to urban center (>50 k pop) (km)	9.00	86.00	35.70	19.96
Number of jobs initially planned by LAGs (no)	3.00	26.00	14.12	4.57
Paid funds (%)	14.02	70.20	46.38	14.74
Contracted funds (%)	56.13	99.81	88.75	11.44
Share of public partners (%)	15.38	35.71	24.94	5.72
Age of LAG (years)	4.00	10.00	7.80	2.70

Source: Own analysis using SPSS 20.0.

).

2.3. Principal Component Analysis (PCA) and Hierarchical Cluster Analysis

Principal component analysis and hierarchical cluster analysis were performed using the Statistical Package for the Social Sciences (SPSS 20.0, Armonk, NY, US, IBM Corp) in order to test the hypotheses presented at the beginning of this work.

PCA is a factorial analysis that is used to identify a small number of factors that together explain an important part of the total variance from a large number of variables. The method uses large correlations between the items to compute those factors, known as principal components (PC), that best represent the data set [33]. In this case, the PCs were selected using the computed eigenvalues (greater than 1), and the interpretation was performed using a varimax matrix that obtains factors as different as possible [33]. Subsequently, the Kaiser-Meyer-Olkin (KMO) test was used to check the validity of data in the PCs analysis. The result was 0.627; therefore, the decision to perform the analysis with those variables is acceptable [1]. The method is often used in LEADER-related studies. Masot et al. [11] used it to analyze the investments and projects carried out by Spanish and Portuguese LAGs, while in Romania, Olar and Jitea used it to analyze the quality of LDSs [30] and the multiplier effects that LAGs had generated in their territories [31].

Hierarchical cluster analysis is used to classify similar objects into homogenous groups, allowing, in this case, to empirically identify the most important characteristics that determine LAGs’ success in the implementation of the LDSs [33]. The analysis was performed using the squared Euclidean distance and Ward’s method in order to produce similar-sized clusters with a similar degree of tightness [33]. Hierarchical clustering requires no a priori decision regarding the number of clusters [34]. While a possible limitation of the method is the fact that there is no accepted rule regarding the number of cases, it will still be able to provide a solution even with a small sample. However, in this case, a trade-off in accuracy could be expected [33]. As in the case of PCA, hierarchical cluster analysis proved its usefulness in other similar studies [10,31,35,36].

3. Results

3.1. Principal Component Analysis Results

Five principal components were selected using the computed eigenvalues (greater than 1) (

Table 2. Results of the principal component analysis.

PC	Eigen Values	% Variation Explained	% Variation Accumulated	Indicators and Correlation with the PCs (The Most Discriminant Variables, Above ± 0.3)
PC1	4.308	26.926	26.926	Number. of municipalities (no) 0.944 Population (no) 0.910 Operating budget (EUR) 0.908 Number of projects (no) 0.803 Rural population/total population (%) −0.714 Jobs created/jobs planned (%) 0.316 Age of LAG (years) 0.576
PC2	2.585	16.155	43.081	Number of projects (no) 0.401 Number of jobs created (no) 0.938 Jobs created/jobs planned (%) 0.682 Firms/1000 pop 0.651 Distance to urban center (>50 k pop) (km) −0.366 Contracted funds (%) 0.492
PC3	2.279	14.245	57.326	Firms/1000 pop −0.565 Young/old population ratio (%) −0.908 Old-age dependency ratio (%) 0.839 Distance to urban center (>50 k pop) (km) 0527
PC4	2.254	14.086	71.412	Jobs created/jobs planned (%) 0.439 Paid funds (%) 0.897 Number of jobs initially planned (no) −0.871 Contracted funds (%) 0.512
PC5	1.270	7.938	79.350	Distance to urban center (>50 k pop) (km) −0.332 Share of public partners (%) 0.814 Age of LAG (years) 0.586

Source: own analysis. Extraction method: principal component analysis. Rotation method: varimax with Kaiser normalization.

; Figure 1). Together, they explain 79.35% of the variance, a satisfactory level as shown in previous studies [1].

Component 1—Relations between LAGs and their territorial characteristics, and the implementation results (PC1).

The first component explains 26.9% of the total variance and deals with the relations between the LAGs characteristics and the number of projects implemented. The results show a strong positive correlation between the number of projects implemented by LAGs and their operation budgets and experience. They are also correlated with territorial characteristics such as population and number of municipalities. Finally, there is a negative correlation with the share of the rural population, suggesting that those areas have implemented, on average, fewer projects than the more urbanized LAGs.

Component 2—Factors that explain the success of LDS implementation (PC2).

The second component (16.15% of variance) explains the factors that influence the success of LDS implementation. Most of the variables that show LAG performance (such as the number of projects financed, share of contracted funds, and number of jobs created in the territory) are positively correlated with the number of firms present in the territory (per 1000 inhabitants) and negatively correlated with the distance to the nearest urban center. Therefore, this PC shows that, in general, the best results are obtained by the LAGs that already have or are located near a developed business framework.

Component 3—Relations between the social and economic variables (PC3).

The third PC shows how the variables discussed at the previous point (namely the distance to the nearest urban center and the business framework) are influencing the social characteristics of the rural areas. Thus, in general, LAGs that are more remote from big urban centers are experiencing more acute demographic problems and have a weaker business framework.

Component 4—Expectations versus reality (PC4).

The fourth component shows a negative correlation between the number of jobs initially planned by LAGs and their success rate in this regard, suggesting that LAGs have failed to correctly anticipate the needs or potential of the territory, or, contrary, they failed to meet their proposed objectives.

Component 5—Links between remoteness and partnership (PC5).

The last PC shows that, in general, the LAGs that are closer to big urban centers have a larger territory (therefore more public partners) and more experience. This can be explained by the fact that LAGs initially appeared in those areas, and only in the second LEADER edition started to be active in the more rural territories.

3.2. Hierarchical Cluster Analysis Results

Cluster analysis reveals three cluster types in the region (

Table 3. Hierarchical cluster analysis results.

Variables	Cluster 1	Cluster 2	Cluster 3	Mean
Number of municipalities (no)	17.75	13.00	7.58	12.12
Population (no)	61,939.88	40,367.27	23,091.83	39,247.12
Operating budget (EUR)	618,171.04	470,906.25	303,112.27	443,957.55
Number of projects (no)	39.88	33.09	20.75	30.06
Rural population/total population (%)	81.15	91.86	97.65	91.33
Number of jobs created (no)	18.25	20.00	14.00	17.22
Jobs created/jobs planned (%)	178.68	140.37	91.79	131.45
Firms/1000 pop	7.35	5.85	7.08	6.71
Young/old population ratio (%)	96.79	65.51	85.62	81.36
Old-age dependency ratio (%)	32.09	43.07	33.29	36.45
Distance to urban center (>50 k pop) (km)	26.38	40.55	37.50	35.70
Number of jobs initially planned (no)	14.25	14.91	13.33	14.12
Paid funds (%)	45.03	47.37	46.38	46.38
Contracted funds (%)	90.01	92.36	84.61	88.75
Share of public partners (%)	26.41	25.49	23.46	24.94
Age of LAG (years)	9.00	9.18	5.75	7.80

Source: own analysis. Extraction method: cluster analysis. Squared Euclidean distance. Ward linkage.

, Figure 2).

Cluster 1. LAGs with best results (CL1, 8 LAGs).

The first cluster groups the LAGs with the best performance regarding the implementation of local development strategies. On average, they have implemented 40 projects, almost double compared to the ones from Cluster 3, and have created more jobs than initially planned. They are characterized by a large territory and population (more than double the amount of municipalities and almost three times the population of Cluster 3), high operating budgets, and a somewhat developed business framework. However, probably the most important characteristics are their location (closest to urban centers) and urbanization level (highest of the three clusters).

Cluster 2. LAGs with moderate results (CL2, 11 LAGs).

The LAGs from the second cluster have obtained similar results regarding the contracted and paid aids and the number of created jobs, such as those from CL1, but, overall, have implemented fewer projects. They have an average size and operating budget, and although they are located farthest away from important urban centers.

Cluster 3. LAGs with the lowest performance (CL3, 12 LAGs).

The LAGs from CL3 are the smallest (both in terms of size and budget), the most ruralized, and the least experienced. They have the lowest performance regarding the number of projects and job creation, being the only ones who have created fewer jobs than originally planned. Most of them are new organizations formed during the most recent CAP programming period.

4. Discussion

As stated at the beginning of this work, the revitalization of rural areas through means of economic diversification, improvement in the quality of life, inter-sectoral cooperation, and demographic stabilization are some of the core objectives of CLLD [4]. Thus LEADER aims to use its specific characteristics to complement the traditional CAP aids and even to serve as an alternative to them in the most rural and remote areas. However, as some scholars have shown, the program was not always successful in this regard. Some regions from Spain and Portugal experienced a similar situation as in the case of CAP aids, meaning that the most urbanized LAGs and municipalities had managed to attract significantly more funds at the expense of the more remote and underdeveloped areas [11]. As a new CAP programming period is getting closer, and with it a new edition of LEADER, it is important to check if this situation is also happening in new EU member states, such as Romania. Moreover, the present work aimed to identify some important features of LAGs that help counterbalance such territorial imbalances. Thus, three hypotheses were formulated and tested. The first hypothesis (H1) stated that the implementation of CLLD is strongly dependent on LAGs’ territorial characteristics, such as the existence of a developed business framework in or near their territories; thus, in Romania, it is not able to solve the mainstream top-down approach critics. The assumption is verified by the results of the first two PC (PC1 and PC2). This territorial imbalance is further emphasized by PC3 and the results of the cluster analysis, showing that the success of the method is highly dependent on the available private investment [18]. The results are in line with those signaled in the older EU member states [9,11,17,18,19,20]. From this perspective, similar to the RDPs, LEADER continues to increase the disparities between the more developed LAGs and those considered less attractive to investment, questioning the capacity of the method to create real, sustainable development.

However, there are other important features of LAGs, highlighted in the first PC and cluster analysis, such as the experience and economic and administrative capacities of LAGs. In this sense, LAGs from the CL1 are enjoying a greater capacity to successfully implement their LDSs, given by their much higher operating budget and experience, especially when compared to those from CL3. Moreover, the LAGs from the second cluster (CL2) have reported similar results regarding the contracted funds, paid aids, and number of jobs created, although they are the farthest away from important urban centers. This suggests that these aspects can be seen as a counterbalance to the influence of the territorial features mentioned at the previous point, thus confirming H2. The results are also in line with those presented in previous studies [20,24,31,37]. Konečný [24] argues that LAGs from the newer member states had both fewer resources and lack of experience, which represented important barriers in the implementation of LEADER, while Patkós [37] highlights the importance of continuity in the implementation of LEADER. Moreover, Olar and Jitea [31], in a study conducted in the same region, pointed toward the LAGs’ age as one of the main characteristics that enable LAGs’ success in creating lasting multiplier effects in their territories. This is more evident when considering that the same groups (those created in the 2014–2020 programming period) had the poorest results in both studies. As stated before, LAGs need to mobilize and build on local knowledge; thus, the lack of human capacity and experience could represent a serious challenge when dealing with a complex methodology [38]. Contrary, the presence of such assets could be a decisive factor in enabling the success of LEADER. In the next CAP programming period, LAGs should look for alternative financial instruments to complement the existing ones, and therefore, to strengthen their economic capacity and to be able to not only maintain their team but also to hire additional experts. The increase in the administrative capacity and the experience already accumulated should allow the more remote LAGs to close the distance between them and the peri-urban groups.

The results are also implying that LAGs are not able to fully implement the principles of CLLD, thus verifying the H3. LEADER is disproportionately favoring those areas with an already dynamic economic environment, while, as Olar and Jitea [30] are pointing out in a study on the same region, they are still employing similar measures as those from the national RDP. However, Dargan and Shucksmith [23] showed that the national and regional authorities have, in general, a different vision regarding the needs of rural areas. In the view of the top-down policies, the innovation is often approached from a technological and product perspective, without the social dimension promoted by LEADER. However, in order to achieve the best results, LAGs need to interact with and complement the RDPs, not copy them [39]. Therefore, the situation raises questions regarding the method’s capacity to address the territorial needs and to create resilient rural areas and calls for a serious rethinking of the LEADER delivery approach. Cañete et al. [18] are arguing that this situation could be addressed in the very design of the program by favoring in the selection criteria LAGs with smaller and more remote municipalities. However, this solution can only lead to limited results if the core principles of LEADER will not be successfully employed. In this sense, the participative approach is crucial. Previous studies in the region underlined the importance of private and NGO partners for the successful implementation of the program. In general, the non-public stakeholders are engaging in more innovative projects, while the public ones are contributing mainly to the economic and administrative capacity of the groups [31]. However, as the same authors are pointing out, only the experienced LAGs are able to fully engage their local actors. Moreover, the public partners still retain an important degree of influence in the settlement of LDSs, a fact illustrated by the high number of measures from Priority 6B (fostering local development in rural areas). On average, 35% of the measures were allocated to this priority (measures associated with public partners), compared to the 20% in Priority 6A (that mostly deals with job creation). This situation calls for improvements regarding LAGs’ accountability and transparency [22]. They should build on the existing local knowledge in order to find the right opportunities for their territories, as shown in previous studies [12,15,16]. The local actors could be stimulated by being involved in networking and cooperation activities with other LAGs, which represent an opportunity to introduce them to new good practices models [40]. However, this will prove to be difficult in countries with a low participative culture [41,42], such as Romania.

Another important shortcoming in the implementation of the CLLD in Romania is represented by the absence of measures tailored to address problems related to demographic decline and rapid aging of the rural population. Multiple factors are at the root of this reality. Probably the most important one is the traditional model of evaluation employed by the management authorities. This approach allows an easy way of quantifying the results of LAGs based on indicators such as the number of jobs created and budgetary execution. However, taken in isolation, it might paint a false image. The failure of meeting the assumed objectives could lead to penalties, not only in the next selection of the LDSs but also during the transition period. Therefore, this might force some LAGs to lower their targets and to abandon the territorial-based approach in favor of a “success-based” strategy by selecting the measures that are the easiest to implement. This situation requires a rapid rethinking of the present evaluation model by also considering LAGs’ results regarding social capital and innovation. This should allow them more liberty in designing LDSs, and better responses to the above-mentioned challenges [16].

However, in some cases, LAGs themselves do not understand the real needs of the territories and are failing to address them in LDSs [30]. In this case, LAGs are limiting their responses to the more traditional RDP measures such as “the setting up of young farmers” while disregarding measures related to education and knowledge transfer. Navarro-Valverde et al. [43] are pointing out that the inability to implement the principles of the area-based approach is proof of a top-down vision in both the cases of LAGs and the management authorities. In recent years, the management authorities attempted to address the lack of diversity of LDSs by allowing and encouraging LAGs to create and employ “atypical measures” [44], which are measures that are not similar to those from the national RDP. A total of 12 out of the 31 LAGs from the region opted to implement such instruments. In total, 14 measures were created, dealing with the promotion and creation of local identity and brands, quality schemes, knowledge transfer, and the development of social and environmental capital. Although not directly aimed at solving the problems of sparsely populated areas, this approach could represent an important milestone for future policies. Konecny et al. [45] emphasize the necessity of such measures in order to truly fulfill the needs of the territory, arguing that without them, the essence of CLLD ceases to exist. Thus, LAGs should remain truthful and embrace the core principles of the participative bottom-up approach promoted by LEADER in order to respond to the more specific local threats [46]. Moreover, in the next programming period, LAGs should also focus on finding additional non-LEADER instruments in order to create multi-funded strategies. This approach should allow them even more liberty in designing and implementing initiatives that are better suited to the territorial needs [47].

5. Conclusions

The objective of the present paper was to verify whether or not LAGs that can counterbalance the existing economic disparities in the rural regions and, under the principles of community-led local development, are managing to revitalize Romanian rural areas by implementing tailored sustainable strategies in their territories. The results show that in many cases, the method falls short of this objective. LAGs success is not only dependent on internal factors such as their experience and resources but is also highly influenced by external variables such as the presence of highly urbanized areas near their territory. The dependence on previously established business infrastructure reduces the capacity of the more rural areas, which were the primary target of the program from its very beginnings, to benefit from LEADER support and remain competitive, worsening the territorial imbalance. The top-down approach tendencies, lack of vision, and experience in both the cases of LAGs and the management authority are further limiting the success of CLLD in the region. This shows the incapacity of the method, in its present form of implementation, to respond to the problems of the sparsely populated areas, raising questions regarding its potential to create resilient rural territories. This calls for a serious rethinking of the LEADER delivery approach and evaluation models. However, the results also suggest a potential solution for reducing the aforementioned disparities. In this regard, the experience and the economic and administrative capacity of LAGs are proving to be crucial and should allow the more remote LAGs to close the gap between them and the more developed groups. This is especially important now, in the eve of a new programming period and in the case of the former socialist countries, where the rural space was highly affected. Moreover, this should help strengthen the relations between small municipalities and other local stakeholders, allowing them to apply for and implement projects that otherwise would exceed their individual economic capacities to support such initiatives [25]. In the next CAP programming period, better results can be obtained by favoring the disadvantaged areas both in the resource allocation and in the selection process and by putting even more emphasis on core LEADER principles such as the involvement of the local stakeholders in decision-making solution delivery.

The study has some important limitations. Due to its focus on a NUTS 2 region, the results cannot be generalized to the entire EU area, being representative only for Romanian and the new EU member states. However, the study complements previous research conducted in Western Europe, so together help building a more complete image. In addition, the results highlight some important solutions to the issues identified in the mentioned literature. The study is also limited by the small number of cases (31 LAGs), which might reduce the accuracy of the methods. Future research is advisable to be conducted at the municipality level, and with a broader scope, by also considering the distribution of support from each EU rural development priority. This will offer better insights into the distribution of LEADER aids and ultimately lead to more tailored solutions.