The Impact of LEADER Funding in Romania

Abstract

Established in 1991 as a European initiative to enhance innovation in rural areas, the LEADER programme became an important tool for the spread of sustainable local development actions in almost all regions of the EU. In Romania, although the programme has generated the implementation of many successful local projects since 2007, quantitative evaluations of its socio-economic impact are extremely limited. The aim of this paper is to evaluate whether LEADER funding has had any effect on the eligible territory of Romania, using multiple linear regression analysis with the evolution of demographic and socio-economic indicators for the 2017–2023 period. The results show that, over a certain amount of euros/capita absorbed from LEADER at the administrative level (UAT), the positive dynamics of the statistical indicators is confirmed. This analysis represents quantitative proof of the LEADER programme’s added value, maintaining or increasing its European role in the future as a sustainable policy directed towards communities.

1. Introduction

As it has been over the past thirty years, LEADER is a pillar of the bridge that connects rural communities and businesses to the future of sustainable economic and social development in rural areas. The LEADER programme was developed in response to the growing demand for endogenous resource-based and community-activated sustainable rural development. LEADER was created in 1991 as an experiment aimed at promoting the development of, and testing fresh ideas for, integrated and sustainable rural development within the European Community. It encourages greater utilisation of rural areas’ potential, boosts their competitiveness, and fosters economic expansion.

Permanent (sustainable) development should result from the most efficient use of resources and the progressive reinforcement of endogenous developmental forces. Applying for European funding and then using it wisely might help that process; this is a real-world implementation of the LEADER strategy. Certain characteristics of communities that hope to receive structural financing influence how well they acquire funds. These include soft variables, such as human and social capital, as well as the endogenous capacity of certain communities, their surroundings, and the requirements for obtaining EU capital. For the continuous advancement of the rural community, rural development is a process that aims to bring about social change and sustainable economic development.

1.1. Insight on the Level of Knowledge in the Field

Up until the 1970s, the exogenous approach of rural development was the only model used to better explain and reduce the different gaps (whether they be cultural, economic or even technical) between the rural areas and the cities, which at the time were considered to be the main centres of activity [1].

This type of approach was based on the idea that rural development may only happen through an increase in demand among populations from urban areas [2,3].

These top-down policies started to be largely criticised by different economists, as they were not producing any sustainable development in rural areas [4]. Woods [5] states that this exogenous model, even though it also had some positive aspects, led to:

-

dependency on external investment (which has an effect on the profits of the development not being locally distributed, but instead exported);

-

a democratic deficit obtained by the non-collaborative nature of this top-down approach.

By the beginning of the 1980s, a new approach emerged, and the top-down view of rural development was replaced by a more local-based one, the endogenous approach. In this bottom-up development model, local resources were used as a matter of priority and the local population had a stronger involvement. As opposed to the expected results, the approach led to locally differentiated variations; the overall development of rural areas did not [always] happen as initially intended [6]. Considering the endogenous model, the responsibility for the non-development of some marginal regions was placed on local actors [7]. The new issues that emerged from the bottom-up approach were given a different name—the neo-endogenous development approach. This new type of bottom-up approach aimed to mobilize local actors and local resources to help develop rural areas [8].

This new approach does not only have a pure economic vision, but is based on the idea that local resources and local actions should be integrated in bigger networks, with local control remaining at the heart of rural development [9,10].

There are many papers that state the important internal/external dualism that this new approach has [11,12,13]. “The importance of focusing on social development, as much as economic, has implications for the institutional basis of endogenous rural development” [14]. “Social inclusion should also be explicitly recognised as an objective, and appropriate targets and indicators developed to support this” [13].

These bottom-up rural development policies were translated in Europe through the “Community-Led Local Development” (CLLD) strategy that the LEADER programme is based upon. The LEADER programme was the first attempt at switching policies towards the regionalising of rural development.

Since its creation in 1991, the LEADER programme was rolled out in four large generations: LEADER I (1991–1993); LEADER II (1994–1999); and LEADER + (2000–2006). After 2007, the new LEADER programme was integrated in the National Rural Development Programme of the EU. However, after 2007, one can also identify two key moments in the evolution of the LEADER approach:

-

the first one was in 2007, at the beginning of the 2007–2013 programming, when the experimental nature of the LEADER approach was abandoned, by institutionalizing the programme as the main axis of financing for rural development, with the introduction of a minimum budget; this was called the “integration process”;

-

the second one was in 2016, when the 2007–2013 programming period ended and a new programming period started (2014–2020), which was seen as a ‘unique opportunity to refocus support from the new EAFRD on growth, jobs and sustainability.

This second important moment in the evolution of LEADER reflects the new neo-endogenous approach well, in that it mainly emphasizes the diversity of rural areas and the resulting need to focus not only on agriculture and farmers, but also on other industries and other private and public actors. It also favours both the decentralisation and the regionalisation of political and administrative decision-making structures [15].

For that matter, OECD [16] also highlights the two basic principles of the new rural paradigm:

(1)

“focus on places instead of sectors”;

(2)

“focus on investment instead of subsidies”.

At European level, there is an extensive literature on LEADER:

• analysing the implementation of the programme at the national level (e.g., Spain [17], United Kingdom [18], Czech Republic [19], or Greece [20]), regional level (e.g., Piedmont [21], Extremadura [22,23], Andalusia [24]), and local level (e.g., considering analysis of individual LAGs [25,26]), as well as comparisons between European states implementing LEADER-type programs (e.g., United Kingdom and Italy [27]) and between European and other territories’ endogenous approaches (e.g., European Union—United States of America [28], Ukraine—European Union [29]);
• considering specific fields of interest for LEADER, such as: social innovation [30,31,32], tourism [33,34,35] or entrepreneurship [36].

Although the LEADER programme has been criticised for not yet achieving its full potential [11,12,37,38], so far it is the most similar method used by development policies with neo-endogenous approaches [39]. It has helped to accumulate experience regarding the ‘do`s’ and ‘don’ts’ in rural development policies, representing a “significant progress in the centralist facilitation of endogenous processes” [40].

1.2. Background Information about LEADER in Romania

LEADER-type actions started in Romania in 2006, during the pre-accession period to the European Union. The Ministry of Agriculture and Rural Development identified at that time 121 informal partnerships from 37 counties between NGOs, local administrations, and representatives of the private sector at the local level who wanted to act in the interest of the communities they were part of. These partnerships were known as Local Initiative Groups (GIL). After a series of training sessions, 91 GILs started developing a local development strategy; however, due to the lack of financial resources, they either did not complete it, or completed it but did not implement it. Even if the potential for forming partnerships was a solution for development, in the socio-economic context of that time, people’s reluctance towards forms of cooperation (coming from the communist period) represented the main obstacle to their establishment. Some GILs were legally established into community associations, and others dissolved before or after fulfilling the objective [41].

After the accession to the European Union in 2007, the National Programme for Rural Development 2007–2013 created, for the first time in Romania. the institutional and financial framework for the implementation of LEADER. It included Axis IV LEADER as a priority axis for rural development, which had an increase in the financial allocation from approximately 2.3% to 4.1% by the end of the period, representing 386,164,150 Euros [42].

During this phase of LEADER in Romania, 163 Local Action Groups were authorised by the Ministry of Agriculture to implement Local Development Strategies, the level of the financial execution rate within axis IV being 91.49% [42,43]. According to [42], Axis IV proved ineffective in relation to the indicators regarding the creation of jobs and the indicators related to innovative actions, but it registered a better effectiveness related to indicators aimed at training and strengthening capacities.

The next programming period, 2014–2020 saw an expansion of LEADER, both from the financial point of view, reaching 6.03% of the financial allocation [44], and territorial distribution—239 Local Action Groups were authorised to function, covering 92% of the territory.

In Romania, Local Action Groups are established based on the legislation related to non-governmental organisations, having legal persons from public, private and civil society sectors as members. The territory covered by the partnership, which must be coherent and homogenous from a geographical and strategic point of view, consists of rural administrative units and small urban units with less than 20,000 inhabitants, established as local administrative units (LAU). Geographical homogeneity (in terms of the surface covered) refers to the continuity of the territory, except in cases where a city with more than 20,000 inhabitants is included.

Analysing the specific literature, there is little research related to the effects generated by LEADER funds on the quality of life and on the economic and social situation of the eligible territories for the Romanian territory [45,46,47,48]. Most of the studies carried out contain qualitative analyses using interviews with representatives of the Local Action Groups (LAGs) from various regions in Romania [49,50]. The lack of research on the effects of the programme could be due to data availability.

The main conclusions of the impact studies that also used quantitative data (indicators from the National Statistics Institute, data related to LEADER funding) carried out on the LEADER programme for the 2014–2020 programming period are:

• There is a positive contribution of the funding obtained by the LAGs through measure 19 LEADER to the rural development of the financed regions [47];
• The initial level of development of the LAGs did not prove to be a factor influencing the performance of absorbing LEADER funds [45];
• The implementation of the extended approach of LEADER as Community-Led Local Development (CLLD) is dependent on the territorial characteristics of the LAGs, such as the existence of a developed business framework in or near their territories [46];
• LAGs are not able to fully follow and implement the principles of CLLD and represent a complement to the measures of the National Rural Development Programme and therefore are not able to respond to the existing economic disparities in rural regions and demographic problems [46].

Moreover, as noted by [51] “the range of key performance indicators being used is narrow, their usefulness as a measure of achievement is limited”.

From the point of view of the methods used in analysing the impact of LEADER on certain areas, most researchers have applied standardised questionnaires or interviews and have calculated the significance of Pearson’s coefficients [20,48,52,53]; some used principal component analysis and cluster analysis [23,50,54]; and some used different regression models [45,55].

According to Lillemets [56], “Although there is a rich abundance of academic literature on the impacts of the CAP on rural areas, (…) only 59 publications that estimated the socioeconomic impacts of the CAP were found. The main findings are the following: the reviewed studies have found CAP to have no significant impacts on rural development as an abstract concept and the rural population; positive but negligible effects on economic output, the generational change in farming and gender equality; a positive effect on employment; and limited or inconclusive evidence about the impact on economic diversification, regional cohesion, and civil participation”.

This paper is a sequel of research on the 2014–2020 programming period for [43] the territorial distribution of LEADER funds, and aims to contribute to the existing literature by identifying if LEADER funding had any effect on the eligible territory, by analysing the evolution of demographic and socio-economic indicators (unemployment and tourism). As [57] proved, “in each evaluation process, the territorial distribution of public support should be the preliminary step to understanding the potential impact on the rural socio-economic system”. This analysis could represent quantitative proof of the LEADER programme’s added value, thus covering a research gap in the existing literature.

The paper is structured as follows: Section 2 introduces the design and the methodological approach to quantitative research conducted for the eligible territory of Romania, highlighting the encountered constraints. In Section 3, the results of the research are discussed followed by conclusions, as well as ideas, for future research in Section 4.

2. Materials and Methods

We start this chapter by presenting the evolution of the territorial distribution in the case of the main local development structures (LAGs) that the LEADER programme is based upon, in order to develop as many rural areas as possible. Other than that, we will describe the data used in this study, together with the theoretical deliberations on the empirical model, as consequences of a series of studies in the field.

2.1. Study Area

In Romania, the LEADER concept was first introduced during the EU programming period 2007–2013, i.e., 20 years behind countries such as Ireland, Italy, or France. Following Romania’s accession to the EU, the LEADER programme was implemented, the first 82 LAGs being chosen in 2011, after a rather long administrative establishment period. Another 81 LAGs were chosen at the end of 2012. The 163 LAGs encompassed 72% of the population and 78 percent of the eligible surface, which consists of communes and cities with less than 20,000 residents [58]. For the second programming period (2014–2020), in 2016, the Romanian Ministry of Agriculture and Rural Development authorised 239 LAGs, which accounted for 86% of the population and 92% of the eligible area.

The Local Action Groups (LAGs) are financed under the National Rural Development Programme for the programming period of 2014–2020, including the transition period of 2021–2022, for the implementation of Local Development Strategies (LDS) and for ensuring the running costs for LAGs.

Receptivity to the new bottom-up approach among Romanian rural communities appears to be particularly high, as evidenced by the notable increase in the number of Local Action Groups from one financing period to the next, but more so by their territorial extension. With the rural areas of Romania being underdeveloped and the primary goal of the LEADER programme being to promote development, this can only be fortunate. The evolution of the overall territory is better shown in Figure 1 and Figure 2.

In study [43], the implementation of the LEADER programme in both programming periods was analysed, considering the types of beneficiaries, the uptake value in each administrative unit/no. of inhabitants, and the financial contribution brought by the beneficiaries to the programme.

2.2. Data and Methods

To carry out the research, we used a publicly available database containing all the projects financed by LEADER during the 2014–2020 programming period, i.e., from 2017 (the date of signing the first financing contract) until 1 March 2023. Although we analysed the 2014–2020 programming period, it is important to consider that the implementation of the approved projects will continue until the end of 2025 (due to the transition period 2021–2022 and the N + 3 rule), which means that the first comprehensive data for this period will not be registered until the following year, at the end of 2026 or beginning of 2027.

We have considered a recommendation made by Bosworth [10], that “To enable more detailed evaluation of future programmes, it will be useful to correlate localities with statistical data areas so that the impact of intervention can be more robustly assessed against other criteria”.

Considering the level of detail of the database, for each administrative unit, only the completed contracts were extracted, considering the initial point of the research, the date of signing the first contract and ultimately, the year in which the last project was completed (more precisely, if the project covered at least 6 months of that year). Thus, a database with all the completed contracts at the level of each territorial unit (2853) in the period 2018–2022 resulted. Therefore, the analysis was carried out on the entire population of administrative units (UATs) and LEADER completed contracts, not a sample (none of the variables were measured at the sample level).

The data were mainly used to show the impact of LEADER funding, correlated to the seven statistical indicators considered to be more susceptible to the influence of different amounts allocated in the LEADER program.

From the point of view of the constraints, given that the data on the amounts allocated to LEADER contracts are available only at the total level and not at the level of each year of the contracting period, no panel analysis could be made to consider the annual variation. For annual analyses, it would have been necessary to make slightly unrealistic assumptions regarding the amounts received annually (proportional distribution of LEADER budgets by year).

Also at the constraint level, it should be mentioned that the allocation of projects by localities was not random (we did not have an experimental design), so the inferences in terms of impact may be incorrect.

In carrying out the impact analysis, a series of dependent variables, an independent variable of interest, and control variables were used, as follows:

A.

Dependent variables:

We selected a series of indicators from the National Statistics Institute to ensure that their variations were potentially influenced by variations in the amounts allocated under the LEADER programme for the reference period (2018–2022) [59]. The two criteria that were the basis for choosing these indicators were: (1) the connection with the activities carried out within LEADER projects and (2) the availability and accessibility of the data. Considering the analysis of the two programming periods of LEADER in Romania presented in [43], the main categories of projects receiving LEADER financing were related to: (1) projects aiming at increasing the quality of rural populations by providing basic services, such as: improvement or creation of leisure and sports infrastructures (10% of contracts), creation of social infrastructures (5%), investments in energy efficiency (3%), the development of local emergency services (25%); and (2) projects involving non-agricultural businesses, mainly tourism.

The chosen indicators were the following [60]:

• the population by domicile, which represents the number of persons with Romanian citizenship and domicile on Romanian territory, delimited according to administrative-territorial criteria. The indicator is relevant for measuring the population at the locality level, as it is the only population indicator available at this level of disaggregation from a territorial point of view;
• settling of domicile is registered in case of persons who arrived and were proven to have ensured a dwelling in the given locality;
• departures from the domicile, registered in the case of persons who left the given locality and were proven to have established a dwelling in another one;
• the unemployment rate represents the ratio between the number of unemployed (registered at the employment agencies) and the civilian active population (unemployed and civilian employed population, defined according to the labour force balance methodology);
• The existing (installed) tourist accommodation capacity represents the number of tourist accommodation places recorded in the last reception, homologation, or classification document of the tourist reception establishment with tourist accommodation functions, excluding the additional beds that can be installed if necessary;
• The number of tourists accommodated in tourist reception units includes all persons (Romanians and foreigners) who travel outside the locality where they normally reside, for a period of less than 12 months, and spend at least one night in a tourist accommodation unit in the areas they visit within the country;
• Tourist overnight stay is the 24 h interval, beginning with the hotel hour, in which a person was shown to have registered in a tourist accommodation establishment and benefited from accommodation in exchange for a fee paid for the occupied space on the premises, even if the actual duration of stay was inferior to the mentioned interval. Overnight stays for additional beds (paid for by tourists) were also considered.

For each of these indicators (except the unemployment rate, for which the absolute variation was used), we constructed the dynamics/relative variation at the level of administrative unit, considering the initial and final points previously defined, according to the formula:

$$Relative variation=\frac{value for the end year-value for the start year }{value for the start year}\times 100$$

A positive value represents a relative increase in the indicator, and a negative value indicates a relative decrease, with a value of 0 indicating no variation. Therefore, the dependent variables analysed were precisely these variations of the indicators, not the statistical indicators themselves, i.e., their absolute values.

B.

Independent variable of interest:

For the amounts in euros per inhabitant allocated during the reference period considering all completed LEADER contracts, we used the variable with categories (0, below 20 euros/cap., between 20 and 40 euros/cap., between 40 and 80 euros/cap. and over 80 euros/cap.) and not the continuous variable due to the non-linearity of the distributions and relations.

In the case of analysis by type of beneficiary and type of project, to ensure a sufficient number of cases within each class, the categories are: 0, under 20 euros/capita, between 20 and 40 euros/capita, over 40 euros/capita.

C.

Control variables:

To account for the possible influence of other factors, in this case the well-being of the locality and funding through other projects similar to the LEADER programme, we considered two control variables:

• the average income per inhabitant at the administrative unit level;
• the average income per inhabitant obtained from EU projects/donations at the administrative unit level.

As Martin [61] states, “there is the added difficulty of disentangling the impacts of the policy measures from the effects of a range of other factors that may also change through time and alter the economic performance of the regions concerned”.

Descriptive results of the database and the types of variables chosen for the empirical model are shown in

Table 1. Descriptive statistics (N = 2853).

	%	Mean	SD	Min	Max
Control variables:
average income per capita (mil. RON)		2.78	1.25	0.98	15.07
average income from other EU projects per capita (mil. RON)		0.13	0.21	−0.36	3.86
Dependent variables:
population (relative change)		−1.43	3.77	−15.80	23.91
inhabitants IN (relative change)		5.23	41.36	−96.07	245.45
inhabitants OUT (relative change)		0.10	30.15	−77.42	116.67
unemployment rate change		−0.80	1.98	−9.60	8.40
existing tourist accommodation capacity (relative change)		3.97	35.63	−100.00	275.00
number of tourists accommodated (relative change)		1.80	69.95	−100.00	688.37
tourist overnight stay (relative change)		2.17	71.88	−100.00	688.89
Independent variable:
amount per capita from LEADER (EUR)
0	23.1
<20	21.0
20 < 40	27.5
40 < 80	20.5
80+	7.9

.

To analyse the relationship between the dependent variables and the independent ones, we used multiple linear regression analysis.

However, as Selman Mermi et al. [62] state, multicollinearity is a problem usually encountered when using this statistical technique, which can lead to inaccurate estimates for the coefficients, inaccurate confidence levels, incorrect t-tests, or even to the excessive growth of standard deviations [63]. Taking this into consideration, we chose to use a robust estimation in order to avoid creating high variance in the coefficient estimations.

The results from the linear regression model are depicted in

Table 2. Linear regression results (beta coefficients).

Amount/Cap from LEADER (EUR)	Population	Inhabitants IN	Inhabitants OUT	Unemployment	Tourist Accommodation Capacity	# of Tourists Accommodated	Tourist Overnight Stay
<20	1.617 ***	−1.508	5.676 **	0.279 *	0.155	1.840	0.540
20 < 40	0.347	−0.660	3.561 *	−0.012	−3.731 *	−5.426	−5.544
40 < 80	−0.530 *	3.165	3.995 *	−0.364 **	−0.559	1.560	1.816
80+	−1.267 ***	10.378 **	3.994	−0.314 *	3.533	7.790	8.318
Controls
avg. inc/cap	−0.211 ***	2.601 ***	−0.187	−0.025	−0.413	−1.323	−1.069
avg. inc/cap (EU)	0.757 *	−1.343	−0.745	0.140	17.219 ***	10.910	8.653
Constant	−1.167 ***	−2.790	−2.596	−0.705 ***	3.782	4.267	4.428
R-squared	0.063	0.016	0.004	0.014	0.013	0.004	0.003
N. of cases	2853	2853	2853	2853	2853	2853	2853

* p < 0.05, ** p < 0.01, *** p < 0.001.

and they show, first of all, an absolute change for unemployment and relative change for others.

The marginal effects of the final model are given in

Table 3. Robust regression results.

	Population	Inhabitants IN	Inhabitants OUT	Unemployment	Tourist Accommodation Capacity	# of Tourists Accommodated	Tourist Overnight Stay
Amount/cap from LEADER (EUR) (LN)	−1.249 ***	4.190 ***	−0.680	−0.359 ***	0.102	1.446	2.841
avg. inc/cap	−0.130	2.199 **	0.163	0.043	−0.103	−1.127	−1.017
avg. inc/cap (EU)	0.749	2.713	−1.002	−0.027	17.319 ***	14.252	12.047
Constant	3.172 ***	−15.268 ***	3.099	0.298	1.494	−1.867	−6.358
R-squared	0.082	0.016	0.000	0.018	0.010	0.002	0.002
N. of cases	2194	2194	2194	2194	2194	2194	2194

** p < 0.01, *** p < 0.001.

.

In order to facilitate the interpretation, we have chosen to present the obtained results, i.e., the values of the regression coefficients associated with the classes of the LEADER variable, in a graph form. The resulting graphs present synthetically the relation between the amount per capita (in euros) allocated through LEADER projects at the level of each administrative unit and the variation of the dependent variables considered. Each graph shows, for each financing category (no financing, financing under 20 euros/inhabitant, financing between 20 and 40 euros/inhabitant, etc.) what was the average variation of the reference indicator in relation to municipalities that received no funding, all other things considered being equal.

With the mentioned limits, the observed differences between the funding categories can be interpreted as a potential impact of the LEADER programme depending on the relative size of the absorbed budgets.

The values in the graphs represent the differences between localities (UATs) with LEADER funding and localities without funding, as estimated by multiple linear regression models with controls. So the values in the graphs are not indicating absolute changes (except graphs related to unemployment), but relative changes.

In the graphs, the dots represent the “average impact” at the level of administrative unit, and the associated lines represent the uncertainty of the estimate, i.e., the confidence interval (for 95% probability). If the lines do not include the value zero, we can state that the observed variation in the dependent variable is statistically significantly different from zero (no variation). Similarly, if two lines for some amount categories do not overlap, the difference is statistically significant.

Even though the analysis included all eligible LEADER localities in Romania, so we did not work on a sample, we chose to keep the confidence intervals associated with the estimates to consider the rest of the uncertainties and errors associated with the data and statistical modelling.

The analyses were carried out both in relation to the total value of completed LEADER contracts and by different types of projects:

• depending on the type of beneficiary: public or private;
• depending on the type of financed investment: infrastructure, services, non-agricultural field;
• depending on the age of the beneficiary’s legal representative—young people;
• for the specific field of tourism.

In carrying out this analysis, it would have been useful to include other control variables in the regression models, which would have measured the potential impact of other factors on the variation of the considered indicators (for example, the amounts received following some LEADER projects from the previous programming). However, these data either do not exist or are not available.

3. Results and Discussion

Analysing the impact of LEADER funding as a total value in relation to the seven previously mentioned statistical indicators, the following can be observed:

From the point of view of the impact of financing on the population, it can be observed that as the uptake value/inhabitant increases, the number of people attracted to the locality increases (Figure 3); thus, the localities with more than 80 euros/capita absorbed from LEADER have registered an average increase of 10.4 percentage points in the number of people who settled in the territory during the analysed period. At the same time, a maintenance of the degree of departure with domicile is observed (Figure 4); however, for the localities with over 80 euros/capita absorbed from LEADER, the number of people settled in the territory compensates for the number of people who chose to leave, resulting in an increase of 6.4 pp.

Consequently, the contracts financed by LEADER seem to positively influence this indicator, but only if the financing per capita is consistent (over 80 euros).

In a study carried out on the 2007–2013 programming period using output indicators and financial and physical statistics, [64] observed that “the impact of the LEADER method implementation on individual regions is primarily conditioned by the number of people living in LAG territories”.

The difference between the relative dynamics of people settled in the territory and that of people who left the territory should be reflected in the analysis of the relations between the relative dynamics of the population and the LEADER/capita value. However, from Figure 5 it can be seen that as the value absorbed/inhabitant increases, the population in the locality decreases; thus, the localities with more than 80 euros/capita absorbed from LEADER recorded an average decrease of 1.3 points in the percentage of the total number of inhabitants. This can be justified by:

• Low birth rate in Romania. According to the National Institute of Statistics, “in 2022, the number of live births registered at civil status offices in Romania was 171,132 live-births, the lowest number of live-births since 1930 to date. (…) The birth rate was, in 2022, 7.8 live births per 1000 inhabitants, higher in rural areas (8.6 live births per 1000 inhabitants) than in urban areas (7.1 births in villages per 1000 inhabitants)” [60];
• Increased mortality rate in Romania. “The general mortality rate was 12.4‰ in 2021 compared to 15.2‰ in the previous year, higher in rural areas (14.4‰) compared to urban areas (10.9‰)” [60].

The existence of statistical data related to the birth rate, respectively, the death rate at the level of UAT, or the distribution by age groups of settlements/departures with domicile, could contribute to the justification of the relative decrease in the population despite the increase in the dynamics of settlements with domicile. For example, if, in those localities that have attracted LEADER funds that have led to the improvement of living conditions, the people who settle are from the 65+ age group, and the people who leave home are from the 19–40 age group, then the phenomenon of population aging is accentuated and the decrease in relative population dynamics is justified.

Regarding the dynamics of the unemployment rate, in the case of the class “LEADER funding = 0”, we can observe in Figure 6 that the graph indicates the value 0 (all values associated with this class are 0), in the case of the class “LEADER funding = between 40 and 80 euros/inhabitant” the value −0.36 means that, in the communities where LEADER projects were carried out with higher amounts per inhabitant, the unemployment rate decreased by 0.36 percentage points more (other conditions considered being the same).

Our calculation confirms the conclusion of [55]: “the higher the unemployment rate, the higher the amount requested. This means that the community believes that attracting higher amounts may increase labour demand and thus reduce unemployment”.

Considering the fact that most of the research on LEADER focused on the field of tourism, we performed the analysis of the effect of the total value of the LEADER contracts on three statistical indicators related to this field, the results being highlighted graphically in Figure 7, Figure 8 and Figure 9. The result is that a positive evolution of the indicator can be observed only with an allocation of over 80 euros/inhabitant from LEADER:

• The number of tourists increases by an average of 7.8 percentage points, triggering an increase in accommodation capacity by an average of 3.5 percentage points;
• In the localities that absorbed more than 80 euros/inhabitant from LEADER funds, there is an average increase of 8.3 pp in the number of nights of accommodation; therefore, we can conclude that the lengths of stay of tourists were longer in these localities.

Another analysis that can be carried out concerns the influence of certain types of projects or the influence of the types of beneficiaries—public or private (with the emphasis on young beneficiaries) on the evolution of the considered indicators.

The evolution of the demographic indicator (population by residence) tends to have the same negative slope, both in the analysis according to the type of beneficiary (Figure 10) and according to the type of project (Figure 11).

This decrease in population seems to be compensated by settlements, noting, in Figure 12 and Figure 13, the fact that financing of more than 20 euros/capita of public and private beneficiaries influences the attraction of new residents to those localities; in the case of young beneficiaries, the impact on the indicator is observed only in the case of localities with more than 40 euros/capita. The growth trend of residential establishments is positively influenced by investments in services, but especially in non-agricultural activities.

In the case of projects related to non-agricultural activities, as the absorbed value/capita increases, a continuous increase in residential establishments is recorded, with an increase of 15.2 pp among the localities where the value of these projects was between 20 and 40 euros/capita and the localities where the value of these projects was over 40 euros/capita. Instead, there seems to be a negative effect of infrastructure investments on the dynamics of residential settlements, with a decrease of 9.8 pp recorded between the two categories of localities.

Although the evolution of the population is decreasing (Figure 14 and Figure 15), emigration seems not to be influenced by the public investments made (this indicator remaining relatively constant in the case of all classes of localities), while private investments lead to a positive effect. It is worth noting that in the localities where the contracts of young beneficiaries had values above 40 euros/capita, departures with domicile show a negative trend.

According to the type of beneficiary, Figure 16 shows that the unemployment rate tends to decrease as the amount of LEADER financing/capita increases, thus in the communities where the public beneficiary contracts had values above 40 euros/capita the unemployment rate decreased on average by 0.41 percentage points. At the same time, it is observed that the investments of private beneficiaries contribute to a decrease in the unemployment rate in localities that absorbs between 20 and 40 euros/capita, with the higher values of the absorbed financing maintaining the unemployment rate at relatively similar levels. From the analysis of the existing data, we observe that the biggest decrease in the dynamics of the unemployment rate (0.47 percentage points) is recorded in the case of localities where young people benefited from LEADER contracts that brought between 20 and 40 euros/capita.

Depending on the type of project (infrastructure, services or non-agricultural), the biggest decrease in the dynamics of the unemployment rate (0.56 percentage points) is registered in the case of localities where the infrastructure projects brought more than 40 euros/capita (Figure 17).

In the case of service projects, as the value absorbed/capita increases, there is a continuous decrease in the unemployment rate. On the other hand, according to the applied regression models, the impact of non-agricultural investments on the dynamics of the unemployment rate is very low, being recorded as a decrease of 0.03 pp only in the case of localities where the value of these projects was over 40 euros/capita.

Finally, the analysis of the impact of LEADER funding was done by referring to a series of indicators belonging to the field of tourism, statistically representative for highlighting supply (accommodation capacity) and demand (number of tourists and number of overnight stays).

In localities where the value accessed by public beneficiaries was over 40 euros/inhabitant, an increase in the relative dynamics of the number of tourists is observed of 3.1 pp. Depending on the value absorbed by private beneficiaries, a continuous decrease in the relative dynamics is observed for the number of tourists; the most remarkable being the decrease of 5.1 pp in the case of localities where the value accessed by young people was over 40 euros/inhabitant (Figure 18).

Investments in all types of projects analysed (services, infrastructure, non-agricultural) seem to have had a positive effect on the number of tourists, with the highest increase recorded in the case of localities where the value accessed through projects that targeted non-agricultural activities was over 40 euros/inhabitant, on average over 14.2 percentage points, as can be seen in Figure 19.

The descendent trend in the number of tourists demonstrated in the case of localities where the value accessed by young people was over 40 euros/inhabitant is also maintained in the case of the dynamics of the accommodation capacity (Figure 20) and the number of overnight stays (Figure 21).

From the analysis of existing data, we observe that the greatest increase in the dynamics of accommodation capacity (9.9 percentage points) and the dynamics of the number of overnight stays (19.4 pp) is recorded in the case of localities where the value of investments in non-agricultural activities was over 40 euros/capita (Figure 22 and Figure 23), although all types of projects that brought in at least 20 euros/capita contribute in a positive way to the evolution of the two indicators, creating an upward slope.

4. Conclusions

The purpose of the paper was to find out if the LEADER programme in Romania had a quantifiable effect on the territory in order to obtain obvious proof and influence the political decision of whether to allocate a higher amount of funds to support endogenous rural development. Although LEADER is a bottom-up approach funding programme, it still depends on political decisions by the Ministry of Agriculture to allocate more than the minimum percentage established by European regulations [65]. Therefore, it is important to obtain quantitative results of the programme demonstrating that a higher absorption of LEADER funds leads to socio-economic development. Considering that LEADER funds are implemented by Local Action Groups, this result could also lead to acknowledging the role of public–private partnerships in the rural economy. As [28] stated “strengthening Public-Private Partnerships and non-profit organisations presence in the European context should be acknowledged as a necessity and not an option”.

The following series of key findings resulted from this research:

• Localities with over 80 euros/capita absorbed from the LEADER programme managed to attract new populations, maintaining a constant level of departures from the territory;
• Through the funding attracted, these localities managed to contribute positively to the attractiveness of the territory, obtaining increases in the number of visitors, thus boosting both the development of accommodation capacities and the development of related services that would prolong visitors’ stays in the targeted territories;
• Beyond the differences, most often small, it is worth noting that positive developments in the unemployment rate occur regardless of the type of beneficiary or the type of project;
• For the field of tourism, it is worth noting that the investments of public beneficiaries with more than 40 euros/capita seem to influence an increase in the number of tourists in the respective localities, accommodation capacities, and lengths of stay;
• The major effect on tourism indicators that non-agricultural investments had (over 40 euros/capita) demonstrates that financing in this field brings added value to the territories, while at the same time boosting a decrease in the unemployment rate;
• The funds attracted by young people did not generally target the field of tourism, a fact highlighted by the negative tendency of specific indicators; however, the funds make a significant contribution to attracting new residents and reducing the unemployment rate when their value in the territory is over 40 euros/capita.

If the study of [66], carried out for the period 2011–2015 (first programming period of LEADER in Romania), concluded that “statistical evidence points to a non-significant difference in economic indicators, suggesting that action undertaken by LAGs did not achieve the goal of promoting economic development of non-agricultural activities in rural areas of their pertinence”, our research, based on data related to the second programming period (2017–2023), led to the conclusion that LEADER funding has an effect on the dynamics of economic statistical indicators, especially if the funds exceeded a certain amount.

Our research completes the study of [55], which, using a cross-section regression model for the value of the projects submitted using 2010 data of 80 LAGs, remarked on a “negative coefficient associated with the VDI showing that the value of projects submitted is greater in less developed counties. Basically, this means that the less developed a village is, the higher the community involvement will be”.

Using social and economic statistics and data about agriculture, tourism, emigration, income, and the labour force in Romania for the years 2011 and 2017, [48] “found an indirect link between people at risk of poverty, RDP subsidies, and CAP total subsidies. In this latter case, an increase in CAP’s second pillar financial subsidies might lower poverty risk, implying financial subsidies’ positive role in reducing the socioeconomic marginalisation in rural areas in all Romanian counties”. LEADER is part of the second pillar of CAP; thus, corroborating the results of our study and those previously made, suggests the need for greater attention in the future from European and national policymakers towards the financing of LEADER.

In the future, interesting research could be carried on using the rural development index proposed by [67], in which it was observed “that, in addition to the economy, social welfare is one of the areas that contributes most to the development of rural areas, thus contradicting “traditional” indices based solely on economic and demographic indicators”. LEADER has proved to be more than a “rural laboratory” [68], confirming its role not only as an social and innovation driver [35], but also as an economic means to sustainable rural development, involving the community.

Future research could examine the complete programme dataset because the programming period is expected to conclude in 2025 [43]. Furthermore, in order to verify whether the effects are still present after a few years, we could pick up the study again. This research, along with others, has practical implications because it may give the managing authority the proof needed to focus future programs on particular operations or areas of interest that have been shown to add value to the programme and help communities achieve their goal of sustainable rural development.