Territorial Rebalancing from an Axiological Perspective: A Reaction Capacity Index of Sicily’s Inner Areas

Abstract

The marginalisation of the inner areas due to increased social, material, economic and infrastructural vulnerability is a growing phenomenon affecting many countries today. Although, specific policies, measures, and funding have recently been proposed to address this issue, they have been slow to produce the expected results. Those responsible for decision-makers regarding the prospect of territorial rebalancing need support in identifying the residual value of these marginal areas. This will help them recognise where and how this value can be emphasised in an integrated, long-term redevelopment process. Based on an axiological perspective of territorial capital forms, the research project has developed a “Geo-referenced Value-based Knowledge Model” using Multi-attribute Value Theory (MAVT). It plays a key role in estimating the Reaction Capability Index (I_RCI) of Sicily’s “inner areas”. The results demonstrate the reaction capability of the municipalities in these areas. As a measure of the overall endowment of territorial capital, the I_RCI index can help decision-makers National Strategy Inner Areas (NSIA), promote the efficient use of resources, and encourage the effective implementation of policies aimed at rebalancing the territory.

1. Introduction

In many countries, unpredictable global dynamics and short-sighted national sectoral policies have, have led to significant territorial imbalances in recent years.

On the one hand, these imbalances have led to the concentration of populations, functions, services, infrastructure, technologies and knowledge in urbanised and/or coastal areas [1,2,3,4,5,6]. On the other hand, they have led to the shrinking and marginalisation of urban peripheries, small towns, villages and rural and inland areas, exacerbating the conditions in places characterised by geographical remoteness [7,8,9,10,11,12,13,14].

The complexity and overlap of the phenomena that triggered processes of territorial degeneration encompasses the entire spectrum of territorial and urban values [15].

The 2018 UN World Urbanisation Prospects report revealed that rural depopulation had worsened significantly.

It is estimated that, by 2050, around two-thirds of the global population will live in large cities, primarily in nations like India, China, and Nigeria. People’s choice of location is influenced by greater access to services, education and employment opportunities.

Around 50% of the world’s population lives in cities with fewer than 500,000 inhabitants. Meanwhile, one in eight people live in one of the 33 cities with a population over 10 million.

In Europe, the phenomenon of shrinking regions can be categorised into four types. The first type comprises industrial areas in economic decline, primarily located in Western Europe.

The second type comprises depopulated peripheral areas, which are common in Northern Europe.

The third type comprises areas that have undergone or are currently undergoing political transformations, such as Eastern Europe.

The fourth type comprises structurally weak rural areas in Southern Europe, that are characterised by a sharp decline in fertility rates [16].

In Europe, demographic decline is associated with internal migration. Since the 1960s, the population has become increasingly concentrated in major urban cities, while rural and inner areas have experienced population loss [17,18,19].

Another phenomenon affecting many European cities is “shrinking cities”. These are cities that have experienced, or are currently experiencing, a reduction in population size [20,21].

Given demographic population decline, the old-age index and the birth rate, as well as cancellations from the civil registry, it is evident that the abandonment phenomenon in Europe is serious, and observable at both the territorial and urban levels [20,21].

1.1. Towards a Taxonomy of Territorial Imbalances

The issues of territorial inequalities and the contrast between centres and peripheries are examined in terms of dependency relationships and the conditions of exclusion, as well as the economic, social, spatial and political implications [8,22,23,24].

This relationship of dependency, which is based on an imbalance of economic, political and social power, was originally explained using Friedman’s centre-periphery model [25].

This model identifies the central area as the most urbanised area, characterised by the highest population density and concentration of economic activity.

As opposed, the periphery is characterised by a diverse population and underdeveloped economy.

From this perspective, metropolitan cities are characterised by innovation, infrastructure, services and decision-making functions, a young, active population.

Overall, they have high socio-economic and political-administrative values, which making them highly competitive and attractive. As opposed, areas in decline are characterised by a lack of innovation and infrastructure, low service levels, a lack of decision-making functions, an ageing and shrinking population, abandoned and dilapidated buildings, and a lack of job opportunities.

Overall, the values in these areas are low or negative, which reduces or eliminates their ability to compete.

The latter are unattractive areas because they offer services and opportunities that do not align with the communities’ preferences [26,27,28,29,30,31,32]. In such cases, citizens dissatisfied with a local jurisdiction’s policies can demonstrate their disagreement by moving elsewhere. This decision is made according to Tiebout’s “foot voting” rule [26].

The Spatial Cycle Model (SCM) was proposed by Hall in 1971 [33], subsequently by Klaassen and Paelinck in 1979 [34], Klaassen and Scimemi in 1981 [35], Klaassen et al. in 1981 [36], van den Berg et al. in 1988 [37], Cheshire et al. in 1989 [38], Aydalot in 1985 [39] and van den Berg et al. in 1987 [40], as well as by Kawashima in 1987 [41]. It represents a generalisation of demographic changes in urban and regional systems.

From the perspective of the SCM, demographic change refers to two main sections of the analysed area: the core and the ring.

The core represents the area characterised by a concentration of economic and social functions. The ring represents the territory surrounding the core in all directions. These two sections of the area are closely related to each other in terms of interactions such as trade, commuting, migration and capital movements.

The Spatial Cycle Model (SCM) plays a key role in describing long-term spatial demographic change. The demographic changes occurring in the core and ring define a series of spatial change phases. It is believed that these phases exist in an orderly sequence, forming a cycle.

According to the Spatial Cycle Model (SCM) proposed by Klaassen and Paelinck (1979) [34], four main phases can be identified: I. centralisation and growth; II. decentralisation and growth; III. decentralisation and decline; and IV. centralisation and decline. Each of these phases is divided into two sub-phases, each of which is considered a stage in the cycle. While it is generally accepted that the cycle begins with Phase I, some authors have suggested that it may start with other phases.

In 2015, Kunh [42] emphasised the importance of combining the original meaning of the term “periphery”, which is linked to the economic and social disadvantages experienced by areas located far from Europe’s main centres of economic activity (spatial peripherality), with the concept of socio-economic “marginality” (non-spatial marginality).

In 2017, Copus et al. [43] identified three main types of periphery.

The first type is characterised by low economic potential, and it is located between areas with higher economic potential.

The second type is characterised by poor access to general interest services, which can be attributed to geographical distance, changes in service delivery technologies, austerity-related funding cuts, or other changes in supply, primarily resulting from privatisation.

The third type is characterised by poor socio-economic performance due to the absence any kind of “organised proximity”.

These areas are characterised by a lack of socio-economic and political connections (“connectivity”), which are not necessarily location-specific and are therefore non-spatial.

They also tend to have inadequate governance structures in terms of political influence, which puts them at a disadvantage when competition for public spending resources.

Theoretical approaches to interpreting phenomena of marginalisation are increasingly considering socio-economic performance and political dynamics rather than just location and spatial dimensions, such as distance and access to key services and opportunities [44].

According to these approaches, places that are not geographically remote can still be considered peripheral due to a lack of connectivity and weak interactions [21,43,45], as well as limited influence in governance [46]. This highlights the complexity of territorial disadvantage.

From this perspective, the concept of peripherality is dynamic and characterised by socio-economic factors. This suggests that the phenomenon of marginality may be considered transitory [21,43].

New information and communication technologies are gradually changing the environment in which people, businesses and organisations interact.

Traditionally, “Geographical”, or “Euclidean”, space defined the context for the flow of goods and services that required direct contact.

In this new context, proximity (i.e., physical distance), travel times and travel costs, which characterise the geographical or “Euclidean” spaces in which flows of goods and services requiring direct contact are identified, are no longer the key constraints on interaction.

Information flowing through non-spatial networks can give rise to other forms of organised proximity (social, legal and institutional). These forms of proximity are now more important for identifying new forms of interaction [47].

1.2. Territorial Rebalancing Policies and Funding

Since the 1980s, growing inequalities and disparities between and within regions have led to the implementation of cohesion policies aimed at achieving more equitable and balanced development [48].

The concept of cohesion in European policies addresses the issues of reducing territorial inequalities and promoting more balanced development.

The concept of cohesion was first defined in the Single European Act (1987) [49], in relation to economic and social factors. It was subsequently defined in Article 174 of the Treaty of Lisbon (2007) [50], this time in relation to territorial factors.

The 1989 reform of the EU Structural Funds was a significant step towards strengthening European Cohesion Policy.

It aims was to promote economic and social development in less EU regions by increasing funding and introducing new allocation principles.

This reform shifted the focus of the funds from simply transferring money to addressing structural issues in member states’ regions [51].

The Green Paper on Territorial Cohesion: Turning Territorial Diversity into Strength’ emphasises the need to increase cooperation and partnerships across regional and national borders in order to address governance issues [52,53].

The Treaty on the Functioning of the European Union (2012) emphasises the importance of the Union taking action to strengthen its economic, social and territorial cohesion issues [54].

The year 1988 marked the beginning of the Cohesion Policy. This policy has gained considerable momentum as a result the impetus of Agenda 2030 [55].

Cohesion policy is aimed at the regions and cities of the European Union (EU) and seeks to promote business competitiveness, job creation, economic growth, sustainable development, and an improved quality of life for citizens.

Cohesion policy covers every region in the EU. However, most of the funding is targeted at regions with a GDP per capita of less than 75% of the EU average.

Today, almost a third of the EU’s total budget is allocated to this policy. However, it is a relatively recent policy, having been established in the second half of the 1980s. Since then, its objectives and resources have periodically changed [56,57].

Increasing resources have been allocated to support the objectives of cohesion policies in the various programming cycles. Various funding sources have been identified to support these objectives, and these sources have been integrated and modified throughout the programming cycles.

During the 2014–2020 programming cycle, for instance, grants were provided by the European Regional Development Fund (ERDF), the Cohesion Fund (CF), the European Social Fund (ESF) and the Youth Employment Initiative (YEI).

During the 2021–2027 programming cycle, some funds have been extended and others integrated. This includes the “Investment for Jobs and Growth Goal” (IJG), the “European Regional Development Fund” (ERDF), the “European Social Fund Plus” (ESF+), the “Cohesion Fund” (CF), the “Just Transition Fund” (JTF) and “Interreg: the European Territorial Cooperation Goal”. Almost a third of the total EU budget has been set aside for cohesion policy in the latest programming period (2021–2027) to achieve objectives such as job creation, improving business competitiveness and economic growth, promoting sustainable development, and enhancing citizens’ quality of life, with a view to addressing disparities between regions.

Reducing the gross domestic product (GDP) gap is one of the cornerstones of cohesion policy. The EU’s policy agenda addresses territorial inequality by promoting partnerships between national, regional and local authorities, businesses, and civil society organisations.

In Italy, cohesion policies are implemented and supported by Decree-Law No. 77 of 31 May 2021 [58], Law No. 101 of 1 July 2021 [59], the Decree of 6 August 2021 [60], and Decree-Law No. 13 of 24 February 2023 [61]. Of these, Mission 5—Component 3 of the National Recovery and Resilience Plan (NPRR) is particularly relevant, as it focuses on special measures for territorial cohesion.

The European Commission has proposed various initiatives to revitalise marginal areas, including the “Europe 2020” package—the European Union’s growth and employment strategy [62]—which promotes “smart, sustainable and inclusive growth”. Another initiative is the “Rural Vision” [63], which aims to creates new momentum for rural areas—home to 30% of the EU population —by building on the EU’s green and digital transitions, and the lessons learned from the pandemic.

The European Agricultural Fund for Rural Development (EAFRD) is also involved to a certain extent.

Other funds have been identified in Italy to support marginalised areas, including the Italian Marginal Common Fund [64], the Countering Deindustrialisation Fund [65], the Marginal Municipalities Fund [66] and the Inner Areas Municipalities [67].

In Italy, the National Strategy for Internal Areas (NSIA) has been promoted by the Agency for Territorial Cohesion since 2013.

It provides support for fragile territories affected by structural processes of depopulation and abandonment. These territories encompass 1077 Italian municipalities and a population of just over 2 million inhabitants [68].

The Italian National Recovery and Resilience Plan (NPRR), funded by the European Union’s Next Generation EU initiative, provides of €100–400 million in funding from the Development and Cohesion Fund (FSC) for the implementation of Mission No. 5, “Inclusion and Cohesion”. This funding is intended not only for “special measures for territorial cohesion”, as mentioned above, as well as for the “National Strategy for Internal Areas” and the “Enhancement of Community Social Services and Infrastructure” [69].

Some studies, such as those by Dall’Erba and Le Gallo (2008) [70] and Becker et al. (2012) [71], have highlighted inconsistencies in the positive association between funding provided by cohesion policies and economic prosperity, or the promotion of the revitalisation of disadvantaged areas.

In 2016, Fratesi (2016) [72] identified both empirical and conceptual issues when interpreting the relationship between funding and development. Whether or not policy objectives have been achieved remains an open question in the literature.

In 2020, Fratesi and Perucca emphasised that the implementation and effectiveness of EU policies can vary significantly depending on the specific territorial resources available to EU regions [73].

From this perspective, the territory—and more specifically the regional capital—is not neutral in the mechanism through which implementation of policies generates development. In fact, specific characteristics of the regions mediate the impact of cohesion policy, so these characteristics must be considered when evaluating the policy.

In 2020, Fratesi and Perucca [73] identified three key issues relating to territorial capital: the allocation of cohesion policy funds; the effectiveness of cohesion policy; and regional development.

Firstly, they emphasised the importance of studying the relationship between regional characteristics and allocation funding to different policy needs. This enables us to understand and improve allocation mechanisms.

Secondly, they show that the impact of EU regional policy on regional development depends on the amount of territorial capital available in a specific region.

Thirdly, in addition to considering the direct link between territorial capital and cohesion policy, it is important to understand the role of territory in regional development—that is to say, the broader contexts in which policies are implemented.

1.3. Cognitive Model of Marginal Territories. NSIA Classification

European and Italian cohesion policies form the basis for actions aimed at territorial rebalancing in inner areas.

The implementation of these policies and their outcomes are closely linked to the definition of the action domain. This domain is represented by the territory and, more specifically, its components.

When developing and implementing these policies, it is important to consider the cognitive model of the territory, particularly with regard to the endogenous factors referred to in the literature.

In Italy, the knowledge model that supports the development of policies aimed at territorial rebalancing is the National Strategy for Internal Areas (NSIA) classification, as proposed by the Interministerial Committee for Economic Planning and Sustainable Development (ICEPSD).

The National Strategy for Internal Areas categorises classifies Italian municipalities based on according to their level of remoteness with regard to in terms access to essential services such as health, education, and transport mobility [68].

It classifies municipalities based on their distance from the “pole” municipality, using a primarily a spatial concept of remoteness and, to a certain extent, a non-spatial socio-economic one. Further information can be found in the “Materials” section (Section 3).

The knowledge model used for classifying inner areas according to the NSIA is based on an information structure designed to categorise municipalities according to their spatial remoteness.

The knowledge base informs the identification of municipalities that should be grouped together to form NSIA areas, the definition of territorial rebalancing strategies for these areas, and how funding is allocated to them.

Implementing cohesion policies aimed at territorial rebalancing can be considered a decision-making problem involving the classification, aggregation and financing of municipalities.

In order to implement cohesion policies effectively and efficiently, the decision-making problem must be developed correctly within the context of the action to be undertaken.

1.4. Axiological Domains: Towards a New Cognitive Model of Marginal Territories

The knowledge model underlying the NSIA classification has become the official tool for tackling the marginalisation and abandonment of territories. Italy was the first European country to conduct a survey on the state of territorial marginalisation based on a unified knowledge framework.

The cognitive model proposed by the NSIA Classification is limited in its basis, relying on the criterion of peripherality in terms of access to essential services such as health, education and mobility. The model’s representation of the territory is therefore partial and does not fully address the issues highlighted in the above subparagraph on the taxonomy of territorial imbalances.

It does not provide a comprehensive representation of exogenous factors, which are key to the development and implementation of policies aimed at rebalancing and revitalising territories.

The partial integration of the representation of exogenous factors, and particularly the absence of a metric to measure their level, weakens this cognitive model. As highlighted by Trovato and Nasca (2022) [74], the cognitive model proposed by the NSIA Classification is not capable of fully capturing the values and value potential of these areas.

Furthermore, it fails to highlight the issue of socio-economic marginalisation.

The latter is mainly due to weak or absent “organised proximity”, i.e., weak socio-economic and political connections that are not necessarily linked to geographical location, but rather to governance structures that lack political influence.

To promote the effective and efficient implementation of policies aimed at rebalancing and revitalising the territory, it is necessary to identify a new cognitive model that can integrate exogenous factors and measure their level. This model must also provide a broader understanding of marginality than the spatial type.

In this regard, the study puts forward a cognitive model based on the concept of territorial capital [75]. The concept of territorial capital is used as a reference point for characterising exogenous factors.

The cognitive model is developed from an axiological perspective of territorial capital [76].

This is important for defining a metric that measures the level of exogenous factors.

This model supports the development of a metric that aims to measure the level or value of the endowment of the various territorial capital forms, as well as the overall endowment, based on the estimation of indices [77,78].

The proposed model, the “Geo-referenced Value-based Knowledge Model”, is a hierarchical structure of indices.

This structure has two levels: the first measures the endowment of different forms of territorial capital and the second provides an overall measure of this endowment using the Reaction Capability Index ($I_{RCI}$).

The aim is to determine the total amount of territorial capital available to municipalities in each region or area.

The Reaction Capability Index ($I_{RCI}$) can be used to explore the relationship between territorial capital in weak areas and the effects of cohesion policies [79,80].

The cognitive model and the set of index-based metrics together define the domain of cohesion policies.

The indices are developed based on the Multi-Attribute Theory of Value (MAVT) [81,82].

In the field of literature, MAVT is employed to support the development of policies, as well as the planning and management of related actions [83,84,85,86,87,88,89,90].

In literature, MAVT is also used to support the development of aggregate indicators [91,92,93,94,95]

The $I_{RCI}$ can be used to classify municipalities in inner areas based on their territorial capital endowment [75].

It can also be used to identify groups of municipalities that will constitute the NSIA areas and to measure the effectiveness of cohesion policies aimed at promoting territorial rebalancing.

The study proposes the estimation of territorial capital forms indices, including $I_{RCI}$, for municipalities in the Sicilian region.

The objective is to establish the total amount of territorial capital within the region’s inner areas.

The $I_{RCI}$ index map, developed using QGIS, can help decision-makers identify the most suitable strategies for rebalancing and revitalising the territories [96,97].

The index is developed from an axiological perspective of the territory, i.e., it is based on the values and/or disadvantages of these areas [76,98].

This approach promotes a reinterpretation of the value [74,76,99,100] of the areas’ territorial capital, based on cohesion policies and related funding, and encourages its enhancement in line with the areas’ specific characteristics.

1.5. Structure of the Paper

The paper is organised into the following sections:

• Section 2 introduces the concept of territorial capital;
• Section 3 introduces the NSIA classification and Sicilian NSIA;
• Section 4 illustrates the methodological approach;
• Section 5 reports and discusses the results;
• Section 6 offers reflections on the cognitive model and the application of the I_RCI as an operational tool for supporting territorial rebalancing policies. It also identifies the limitations of the research and its potential future developments.
• Section 7 summarises the key findings of the study.

2. The Concept of Territorial Capital

In 2013, Camagni and Capello emphasised the emergence of a “selective model of regional growth”, whereby the growth of individual regions is differentiated, resulting in a varied mosaic of development stories [101].

In this context, in order to develop growth strategies for regions, territories and cities, it is necessary to understand the various regional resources,—or “territorial capital”—and ensure they are exploited efficiently. The concept of territorial capital was first introduced by the OECD in the context of regional policy [102].

This concept defines “territorial capital” as the stock of resources on which the endogenous development of each city or region is based [75,102].

For the first time, a more holistic and comprehensive analysis of a given location is proposed that take into account its wide range of material and immaterial resources.

Although many researchers have investigated the impact of individual components territorial capital on performance [103,104,105], few have explored how these resources contribute collectively to endogenous development.

In 2008, Camagni [75] provided the first taxonomy of territorial capital, categorising its components according to their level of rivalry and materiality (Figure 1).

The rivalry dimension was categorised as either public goods or private goods, or as an intermediate class of either club and impure public goods.

Meanwhile, the materiality dimension was categorised as tangible goods, intangible goods, or as an intermediate class of mixed goods (hard-soft).

Two criteria are used to classify these components: their degree of materiality (ranging from tangible to intangible) and their degree of rivalry (ranging from private to public).

The four extreme categories, characterised by either high or low levels of rivalry and either material or immaterial assets, represent the traditional sources of territorial capital.

The five intermediate classes highlight the most interesting and innovative elements, which Camagni has defined as the “innovative cross” upon which to focus attention. Along the materiality axis, these categories identify encompass mixed goods characterised by a combination of hard and soft elements, as well as material goods and services.

This provides evidence that the capacity to transform virtual and immaterial elements into effective actions, public/private partnerships and service provision.

It also demonstrates the capacity to transform geographical and cognitive proximity into effective links between economic agents.

In this regard, a more in-depth analysis of the concept of territorial capital is fundamental to understanding how contexts can trigger and guide endogenous sustainable economic development [106].

However, since that not all regions have the necessary prerequisites for this type of approach, and given that favourable conditions are constantly being challenged by globalisation, market changes and technologies [107,108], it is necessary to stimulate endogenous factors exogenously.

This study will analyse the external factors that could contribute to the redevelopment of inner areas. More specifically, these factors will be identified in relation to the categories of elements that traditionally define territorial capital [75], i.e., the components highlighted at the vertices of the rivalry-materiality model square.

3. Materials

The National Strategy for Internal Areas classifies Italian municipalities according to their spatial proximity to a “pole municipality”, which is characterised by the better provision of essential services such as health, education and transport [68].

Municipalities are classified as “belt”, “intermediate”, “peripheral” or “ultra-peripheral” based on their distance from the “pole” municipality (Figure 2).

A municipality’s degree of peripherality is determined by its distance from the “pole” municipality.

Table 1. Number of NSIA areas; number of municipalities per periphery class; population and territorial area for the 2014–2020 and 2021–2027 cycles [68].

Cycle 2014–2020 NSIA Areas	Tot. Municipalities	A—Pole	C—Belt	D— Intermediate	E— Peripheral	F— Ultra- Peripheral	Tot. Population (ISTAT 2020)	Tot. Surface Area (sq.km.)
72 NSIA Areas in 2014–2020 Cyle	1060	1	127	333	482	117	1,938,909	51,205
Cycle 2021–2027 NSIA Areas	Tot. Municipalities	A—Pole	C—Belt	D— Intermediate	E— Peripheral	F— Ultra- Peripheral	Tot. Population (ISTAT 2020)	Tot. Surface Area (sq.km.)
67 NSIA Areas in 2014–2020 Cycle	1105	1	148	358	482	116	2,301,499	54,615
56 NSIA Areas in 2021–2027 Cycle	764		55	236	369	104	2,056,139	38,442
Minor Inland	35		3	1	13	18	213,093	971

shows the 72 NSIA areas that have been identified in Italy in relation to the 2014–2020 NSIA programming cycle. These areas cover 1060 municipalities.

Some areas have been modified and others added to the 2021–2027 NSIA programming cycle. The latest programming cycle involves 1904 municipalities (Table 1).

In the 2021–2027 programming cycle, NSIA areas were selected based on six main criteria [109]:

• Consistency of the new area to be nominated with the Map of Internal Areas proposed for the 2021–2027 programming cycle.
• The existence of a defined and recognisable identity and geomorphological system.
• The area is experiencing demographic difficulties.
• Essential services need to be reorganized.
• The willingness and ability of local authorities to collaborate in order to achieve a shared objective.
• Size of the area.

In Sicily, eleven NSIA areas have been identified, namely (Figure 3): the Calatino area, includes eight municipalities; the Corleone area, includes sixteen municipalities; the Bronte area, which includes thirteen municipalities; the Madonie area, which includes twenty-one municipalities; the Mussomeli area includes eleven municipalities; the Nebrodi area, which includes twenty-one municipalities; the Palagonia area, which includes six municipalities; the Santa Teresa Riva area includes fifteen municipalities; the Terre Sicane area includes twelve municipalities; the Troina area includes fourteenth municipalities; and the Val di Simeto area includes twenty-one municipalities.

Table A1. Characterisation of the eleven Sicilian NSIA areas based on the programming cycle, municipalities, total surface area and total population [68].

NSIA 2014–2020	NSIA-2020	Municipalities	Tot. Surface Area (sq.km.)	Tot. Population (ISTAT 2020)
Calatino	E—Peripheral	Caltagirone	383.37	36,241
Calatino	E—Peripheral	Grammichele	31.02	12,561
Calatino	D—Intermediate	Licodia Eubea	112.45	2794
Calatino	E—Peripheral	Mineo	246.32	4991
Calatino	F—Ultra-peripheral	Mirabella Imbaccari	15.30	4282
Calatino	E—Peripheral	San Cono	6.63	2431
Calatino	E—Peripheral	San Michele di Ganzaria	25.81	2965
Calatino	E—Peripheral	Vizzini	126.75	5772
NSIA 2021–2027	NSIA-2020	Municipalities	Tot. Surface Area (sq.km.)	Tot. Population (ISTAT 2020)
Corleone	F—Ultra-peripheral	Bisacquino	64.97	4203
Corleone	E—Peripheral	Campofelice di Fitalia	35.29	473
Corleone	F—Ultra-peripheral	Campofiorito	22.35	1181
Corleone	E—Peripheral	Castronovo di Sicilia	201.06	2880
Corleone	F—Ultra-peripheral	Chiusa Sclafani	57.40	2611
Corleone	E—Peripheral	Ciminna	56.84	3485
Corleone	E—Peripheral	Contessa Entellina	136.40	1536
Corleone	E—Peripheral	Corleone	228.69	10,580
Corleone	F—Ultra-peripheral	Giuliana	23.64	1730
Corleone	E—Peripheral	Godrano	39.27	1087
Corleone	E—Peripheral	Lercara Friddi	37.27	6340
Corleone	F—Ultra-peripheral	Palazzo Adriano	129.25	1863
Corleone	F—Ultra-peripheral	Prizzi	95.03	4342
Corleone	E—Peripheral	Roccamena	33.32	1388
Corleone	E—Peripheral	Roccapalumba	31.41	2298
Corleone	E—Peripheral	Vicari	85.74	2484
NSIA 2021–2027	NSIA-2020	Municipalities	Tot. Surface Area (sq.km.)	Tot. Population (ISTAT 2020)
Bronte	F—Ultra-peripheral	Cesarò	215.75	2212
Bronte	E—Peripheral	Francavilla di Sicilia	82.11	3636
Bronte	F—Ultra-peripheral	Malvagna	6.90	649
Bronte	F—Ultra-peripheral	Moio Alcantara	8.76	664
Bronte	E—Peripheral	Motta Camastra	25.29	804
Bronte	F—Ultra-peripheral	Roccella Valdemone	40.98	583
Bronte	F—Ultra-peripheral	Santa Domenica Vittoria	19.98	871
Bronte	F—Ultra-peripheral	San Teodoro	13.90	1260
Bronte	E—Peripheral	Bronte	250.01	18,327
Bronte	F—Ultra-peripheral	Castiglione di Sicilia	120.41	2969
Bronte	E—Peripheral	Maletto	40.88	3613
Bronte	F—Ultra-peripheral	Randazzo	204.84	10,324
Bronte	F—Ultra-peripheral	Maniace	35.87	3739
NSIA 2014–2020	NSIA-2020	Municipalities	Tot. Surface Area (sq.km.)	Tot. Population (ISTAT 2020)
Madonie	F—Ultra-peripheral	Alcara li Fusi	62.93	1763
Madonie	E—Peripheral	Caronia	227.26	3097
Madonie	E—Peripheral	Castel di Lucio	28.78	1206
Madonie	E—Peripheral	Castell’Umberto	11.43	2872
Madonie	E—Peripheral	Frazzanò	7.00	601
Madonie	E—Peripheral	Galati Mamertino	39.31	2361
Madonie	F—Ultra-peripheral	Longi	42.11	1348
Madonie	E—Peripheral	Militello Rosmarino	29.54	1197
Madonie	E—Peripheral	Mirto	9.27	913
Madonie	E—Peripheral	Mistretta	127.47	4434
Madonie	E—Peripheral	Motta d’Affermo	14.58	670
Madonie	E—Peripheral	Naso	36.74	3523
Madonie	D—Intermediate	Pettineo	30.62	1240
Madonie	D—Intermediate	Reitano	14.12	733
Madonie	E—Peripheral	San Fratello	67.63	3350
Madonie	E—Peripheral	San Marco d’Alunzio	26.14	1830
Madonie	E—Peripheral	San Salvatore di Fitalia	15.00	1178
Madonie	E—Peripheral	Sant’Agata di Militello	33.98	11,989
Madonie	D—Intermediate	Santo Stefano di Camastra	21.92	4416
Madonie	E—Peripheral	Tortorici	70.50	5908
Madonie	D—Intermediate	Tusa	41.07	2663
NSIA 2021–2027	NSIA-2020	Municipalities	Tot. Surface Area (sq.km.)	Tot. Population (ISTAT 2020)
Mussomeli	E—Peripheral	Cammarata	192.03	5930
Mussomeli	D—Intermediate	Casteltermini	99.51	7473
Mussomeli	E—Peripheral	San Giovanni Gemini	26.30	7590
Mussomeli	D—Intermediate	Acquaviva Platani	14.72	891
Mussomeli	D—Intermediate	Bompensiere	19.74	522
Mussomeli	D—Intermediate	Campofranco	36.06	2758
Mussomeli	E—Peripheral	Marianopoli	12.96	1669
Mussomeli	E—Peripheral	Mussomeli	163.90	10,059
Mussomeli	D—Intermediate	Milena	24.56	2777
Mussomeli	D—Intermediate	Montedoro	14.14	1418
Mussomeli	D—Intermediate	Sutera	35.55	1234
NSIA 2014–2020	NSIA-2020	Municipalities	Tot. Surface Area (sq.km.)	Tot. Population (ISTAT 2020)
Nebrodi	F—Ultra-peripheral	Alcara li Fusi	62.9346	1763
Nebrodi	E—Peripheral	Caronia	227.2587	3097
Nebrodi	E—Peripheral	Castel di Lucio	28.7781	1206
Nebrodi	E—Peripheral	Castell’Umberto	11.4282	2872
Nebrodi	E—Peripheral	Frazzanò	6.9981	601
Nebrodi	E—Peripheral	Galati Mamertino	39.3108	2361
Nebrodi	F—Ultra-peripheral	Longi	42.1099	1348
Nebrodi	E—Peripheral	Militello Rosmarino	29.5356	1197
Nebrodi	E—Peripheral	Mirto	9.2692	913
Nebrodi	E—Peripheral	Mistretta	127.4678	4434
Nebrodi	E—Peripheral	Motta d’Affermo	14.5764	670
Nebrodi	E—Peripheral	Naso	36.7367	3523
Nebrodi	D—Intermediate	Pettineo	30.6164	1240
Nebrodi	D—Intermediate	Reitano	14.1173	733
Nebrodi	E—Peripheral	San Fratello	67.6292	3350
Nebrodi	E—Peripheral	San Marco d’Alunzio	26.1407	1830
Nebrodi	E—Peripheral	San Salvatore di Fitalia	14.9963	1178
Nebrodi	E—Peripheral	Sant’Agata di Militello	33.9764	11,989
Nebrodi	D—Intermediate	Santo Stefano di Camastra	21.9173	4416
Nebrodi	E—Peripheral	Tortorici	70.5008	5908
Nebrodi	D—Intermediate	Tusa	41.0737	2663
NSIA 2021–2027	NSIA-2020	Municipalities	Tot. Surface Area (sq.km.)	Tot. Population (ISTAT 2020)
Palagonia	E—Peripheral	Castel di Iudica	14.14	4352
Palagonia	E—Peripheral	Militello in Val di Catania	62.14	6792
Palagonia	E—Peripheral	Palagonia	57.66	15,805
Palagonia	E—Peripheral	Raddusa	23.32	2875
Palagonia	E—Peripheral	Ramacca	305.38	10,377
Palagonia	D—Intermediate	Scordia	24.26	16,296
NSIA 2021–2027	NSIA-2020	Municipalities	Tot. Surface Area (sq.km.)	Tot. Population (ISTAT 2020)
Santa Teresa di Riva	E—Peripheral	Alì	16.69	702
Santa Teresa di Riva	F—Ultra-peripheral	Antillo	43.4	844
Santa Teresa di Riva	E—Peripheral	Casalvecchio Siculo	33.37	749
Santa Teresa di Riva	E—Peripheral	Fiumedinisi	35.99	1294
Santa Teresa di Riva	E—Peripheral	Forza d’Agrò	11.18	883
Santa Teresa di Riva	D—Intermediate	Furci Siculo	17.86	3205
Santa Teresa di Riva	E—Peripheral	Limina	9.81	735
Santa Teresa di Riva	E—Peripheral	Mandanici	11.65	557
Santa Teresa di Riva	D—Intermediate	Nizza di Sicilia	13.18	3518
Santa Teresa di Riva	D—Intermediate	Pagliara	14.57	1097
Santa Teresa di Riva	E—Peripheral	Roccafiorita	1.14	182
Santa Teresa di Riva	D—Intermediate	Roccalumera	8.77	3953
Santa Teresa di Riva	E—Peripheral	Sant’Alessio Siculo	6.17	1488
Santa Teresa di Riva	D—Intermediate	Santa Teresa di Riva	8.13	9271
Santa Teresa di Riva	E—Peripheral	Savoca	8.80	1660
NSIA 2014–2020	NSIA-2020	Municipalities	Tot. Surface Area (sq.km.)	Tot. Population (ISTAT 2020)
Terre Sicane	E—Peripheral	Alessandria della Rocca	62.2395	2567
Terre Sicane	F—Ultra-peripheral	Bivona	88.5732	3298
Terre Sicane	E—Peripheral	Burgio	42.2307	2532
Terre Sicane	E—Peripheral	Calamonaci	32.8901	1203
Terre Sicane	D—Intermediate	Cattolica Eraclea	62.1638	3364
Terre Sicane	E—Peripheral	Cianciana	38.0815	3177
Terre Sicane	F—Ultra-peripheral	Lucca Sicula	18.632	1730
Terre Sicane	D—Intermediate	Montallegro	27.4124	2385
Terre Sicane	E—Peripheral	Ribera	118.521	18,058
Terre Sicane	E—Peripheral	San Biagio Platani	42.6663	2946
Terre Sicane	F—Ultra-peripheral	Santo Stefano Quisquina	85.5183	4216
Terre Sicane	E—Peripheral	Villafranca Sicula	17.6292	1358
NSIA 2021–2027	NSIA-2020	Municipalities	Tot. Surface Area (sq.km.)	Tot. Population (ISTAT 2020)
Troina	E—Peripheral	Agira	163.11	7756
Troina	E—Peripheral	Assoro	111.50	4892
Troina	D—Intermediate	Calascibetta	88.17	4169
Troina	F—Ultra-peripheral	Cerami	94.87	1859
Troina	D—Intermediate	Catenanuova	11.17	4519
Troina	E—Peripheral	Gagliano Castelferrato	56	3368
Troina	E—Peripheral	Leonforte	84.09	12,583
Troina	F—Ultra-peripheral	Nicosia	217.87	12,947
Troina	E—Peripheral	Nissoria	61.62	2849
Troina	F—Ultra-peripheral	Sperlinga	58.76	691
Troina	F—Ultra-peripheral	Troina	166.95	8699
Troina	E—Peripheral	Regalbuto	169.27	6830
Troina	E—Peripheral	Valguarnera Caropepe	9.32	7163
Troina	D—Intermediate	Villarosa	55.01	4496
NSIA 2014–2020	NSIA-2020	Municipalities	Tot. Surface Area (sq.km.)	Tot. Population (ISTAT 2020)
Valle del Simeto	E—Peripheral	Adrano	83.2208	33,926
Valle del Simeto	D—Intermediate	Biancavilla	70.2753	22,987
Valle del Simeto	E—Peripheral	Centuripe	174.1946	5172

in Appendix A shows the characteristics of the eleven Sicilian NSIA areas based on the programming cycle, municipalities, total surface area and total population.

4. Methods

According to cohesion policies aimed at territorial rebalancing [110] to be identified and implemented effectively and efficiently, it is crucial to define their scope in relation to endogenous factors, as discussed in literature. These endogenous factors can be identified using the concept of territorial capital and the rivalry-excludability model proposed by Camagni in 2008 [75].

Adopting a cognitive model of exogenous factors and developing a metric from an axiological perspective could improve the implementation of cohesion policies.

This study proposes a cognitive model that involves identifying and measuring the level of territorial capital components by constructing a system of aggregate indices that measure various forms of territorial capital and its overall endowment (Figure 4).

The structure of the methodological approach is characterised by the following phases (Figure 5).

The cognitive model covers six forms of territorial capital: human, urban, infrastructure, economic, natural and environmental. These six forms of territorial capital are characterised using indicators for various Sicilian municipalities. Thirteen indicators were identified for human capital, twenty-six for urban capital, five for natural capital, thirty for economic capital, and sixty-seven for infrastructure capital.

These indicators were selected with reference to the datasets of the Agency for Territorial Cohesion [68] and that of the Council of Ministers [111].

These two datasets provide a fairly comprehensive official collection of data on territorial capital indicators. However, the data are not homogeneous, as they refer to different years of collection (2011, 2012, 2013, 2014 and 2015), and are not up to date.

Figure 6 and Figure 7 shows the 127 indicators that were identified to characterise the six forms of territorial capital. The environmental capital indicators highlight the main risks, including seismic activity, flooding and landslides [112,113,114,115]. These indicators are therefore characterised by an opposite polarity to all the other indicators for the different forms of capital.

The model was developed from an axiological perspective [116,117] and is based on the identification a hierarchical structure for estimating indices.

The first level of the hierarchical structure is characterised by the above-identified indicators for estimating the six aggregate indices of the different forms of territorial capital: $I_{HC}$, $I_{UC}$, $I_{IC}$, $I_{EC}$, $I_{NC}$, $I_{EnvC}$ [118,119,120,121]. The second level of the hierarchy is characterised by the indices of the six forms of territorial capital for estimating $I_{RCI}$ index.

The Multi-Criteria Multi-Attribute Value Theory (MAVT) approach was proposed to estimate indices at different levels. This approach is most suitable for the purposes of this study and the chosen analysis perspective.

MAVT is a simplified version of multi-attribute utility theory (MAUT). Unlike MAUT, MAVT does not consider the decision-maker’s risk propensity. Consequently, MAVT is based on simpler elicitation procedures.

The Multi-Attribute Value Theory (MAVT) (Multi-attribute Value Theory) [122,123,124,125] allows decision problems to be structured and analysed using attribute trees (also known as called value trees), and to identify the relative importance of criteria to be identified in the context of the evaluation being undertaken.

The overall objective in an attribute tree is decomposed hierarchically into subordinate objectives (also called criteria) and measurable attributes (also called the lowest level or criteria leaf).

MAVT includes several aggregation models, but the simplest and most widely used is the additive model [126], which can be used when the conditions for preference independence are met. Assuming preference independence, the additive aggregation model can be used. Given this characteristic, it seems that MAVT is a good approach for estimating aggregate indices.

Under these conditions, the value of each index relating to a component of territorial capital or the $I_{RCI}$, is estimated using Equation (1):

$$I_{C_i}={{\displaystyle \sum }}_{j=1}^n{w}_j{v}_j\left(x_j\right)$$

(1)

where $I_{C_i}$ represents an estimate of the aggregate index of a specific component of territorial capital, or $I_{RCI}$, $n$ is the number of indicators or of indices, $w_j$ is the weight of each indicators or indices $j$ and $v_j\left(x_j\right)$ the value taken by the value function for the indicator or index $j$.

Several weighting techniques are described in the literature, which can be grouped into two broad categories: numerical weight estimation and indifference weights [123].

The first category includes ranking, direct rating, ratio estimation and swing weights, and the Entropy Weight Method (EWM). The second category includes the trade-off method [126].

Of the various weighting criteria techniques proposed in the literature, the Entropy Weight Method (EWM) was selected for this case study.

The EWM is a well-researched and widely used model for weighting criteria [127,128].

Compared to other subjective weighting models, main advantage of EWM is that it eliminates the need for human input when allocating weights to indicators/criteria.

This improves the objectivity of the overall assessment results [129,130,131,132,133,134,135,136,137].

For this reason, EWM has become a popular tool for decision-making processes in recent years [138,139].

EWM determines value by measuring the degree of differentiation. The greater the dispersion of the measured value, the greater the differentiation of the index and the more information that can be obtained. According to traditional literature, EWM results are always reliable and effective [140,141].

A spatial analysis of the first-level and second-level results—the estimation of aggregate indices for various territorial capital components and the $I_{RCI}$ index based on quartiles—provides a map showing the level of endowment in Sicilian municipalities [142,143,144]. This analysis was conducted using QGIS.

This mapping provides the knowledge base necessary for evaluating the impact of cohesion policies on the municipalities and NSIA areas involved.

Quartile analysis categorises municipalities into one of four groups according to their level of capital endowment: low, medium, medium-high and high.

These clusters are characterised by the first, second, third and fourth quartiles of the estimated indices, respectively.

Analysing the indices by quartile, province and type of municipality, according to the NSIA classification (i.e., A—Pole, B—Intercommunal Pole, C—Belt, D—Intermediate, E—Peripheral and F—Ultra-peripheral), provides detailed insight into the levels of the various forms of capital.

It shows the number and percentage of municipalities in each province, as well as the number of municipalities in each class NSIA that fall within the different quartiles.

5. Results

With reference to phases 1–5 of the methodological approach proposed in the previous session, the aggregate indices of the six components of territorial capital were estimated, and their values mapped. The level of endowment for various Sicilian municipalities. is highlighted by a map showing the indices based on their quartiles.

The aggregate human capital index was estimated using the value tree shown in Figure 8a.

Figure 8b shows a map of the level of human capital endowment in different Sicilian areas and municipalities.

It shows higher levels of human capital in the eastern areas and lower levels in the central areas.

Figure 9 shows the provinces with the lowest and highest levels of human capital. Agrigento and Caltanissetta have the lowest levels, while Catania and Palermo have the highest. Municipalities in the first quartile in Agrigento and Caltanissetta are mostly classified as D (Intermediate), E (Peripheral) or F (Ultra-peripheral).

This indicates that these areas are more vulnerable, as even those that are not extremely peripheral have low human capital.

In the case of the provinces of Catania and Palermo, municipalities in the fourth quartile are classified under all types of NSIA.

This highlights that some peripheral areas benefit from organised proximity, which limits the impact of their geographical remoteness.

Similarly, the aggregate index for urban capital was estimated using the value tree shown in Figure A1a (see Appendix B).

Figure A1b (see Appendix B) illustrates the level of urban capital endowment in the different Sicilian areas and municipalities.

This map shows that the central areas of Sicily have considerable diversity in urban capital endowment. It shows that both inner and coastal areas have a good supply of urban capital.

This map shows that the central areas of Sicily have a significant diversity in urban capital endowment.

This means that this component of territorial capital is less affected by the typical distinction between inner and coastal areas.

Many Sicilian municipalities have a high level of historical, cultural, architectural and urban heritage [145], as evidenced by the large number of sites included on the World Heritage List.

Quartile analysis of the aggregate urban capital index, by province and municipality type according to the NSIA, shows that approximately one-third of municipalities in the provinces of Catania and Syracuse fall into the first quartile, while one-third of municipalities in the provinces of Caltanissetta, Enna, Palermo, Ragusa and Trapani fall into the fourth quartile (see Figure A2 in Appendix B).

The aggregate infrastructure capital index was estimated using the value tree shown in Figure A3a (see Appendix B).

Figure A3b (see Appendix B) shows a map of infrastructure capital endowment. It highlights the levels of infrastructure capital endowment in different Sicilian areas and municipalities.

It shows that coastal areas, particularly in the provinces of Catania, Palermo, Ragusa, Syracuse and Trapani, have a high concentration of infrastructure capital, while inner areas have medium and low levels.

This infrastructure capital mapping confirms that inner areas are underdeveloped compared to coastal areas. These are characterised overall by a lack of services and transport infrastructure, which exacerbates their socio-economic disadvantage.

The quartile analysis shows that a high percentage of municipalities in the province of Agrigento fall into the first quartile (see Figure A4 in Appendix B).

It also shows that approximately one-third of municipalities in Catania province have the best infrastructure capital endowment (see Figure A4 in Appendix B).

The aggregate index for economic capital was estimated based on the value tree shown in Figure A5a (see Appendix B).

Figure A5b (see Appendix B) shows a map of the level economic capital endowment in different Sicilian areas and municipalities.

This map highlights large with low and medium economic capital endowment, mostly inland, and areas with medium-high and high economic capital endowment, mostly coastal.

Quartile analysis revealed that a high proportion of municipalities in the province of Messina fell into the first quartile, while a high proportion of municipalities in the province of Ragusa fell into the fourth quartile (see Figure A6 in Appendix B).

In the first case, these are economically vulnerable areas characterised by strong spatial and non-spatial marginality.

In second case, the growth of tourism in recent years, partly due to the UNESCO site “The late Baroque towns of the Val di Noto”, has sustained these areas, as has the development of agriculture.

The aggregate natural capital index was estimated based on the value tree shown in Figure A7a (see Appendix B).

Figure A7b (see Appendix B) illustrates the level of natural capital endowment in different Sicilian areas and municipalities.

It highlights areas of high or medium-high natural capital endowment in the east and west of the island, a medium-high endowment in the centre, and a medium or low endowment in the south.

Quartile analysis revealed that a high proportion of municipalities in the province of Caltanissetta fall within the first quartile, while a high proportion of municipalities in the province of Syracuse fall within the fourth quartile. In the latter case, this is due to the presence of several nature reserves spanning multiple municipalities in the province of Syracuse (see Figure A8 in Appendix B).

The aggregate index for environmental capital was estimated based on the value tree shown in Figure A9a (see Appendix B).

Figure A9b (see Appendix B) shows a map of environmental capital endowment.

It illustrates the level in different Sicilian areas and municipalities.

The aggregate environmental capital index provides an integrated measure of the main environmental risks. It takes into account seismic, flooding, landslides, hydraulic s and land consumption risks. This index is characterised by a polarity opposite of that of all other indices.

The environmental capital index map shows which areas of Sicily are most and least exposed to risk. The most exposed areas are in the north and east of the island, while the least exposed areas are in central-south and west.

Quartile analysis shows that a high proportion percentage of municipalities in the province of Enna fall into the first quartile, i.e., they are characterised by low environmental risk (see Figure A10 in Appendix B).

Figure A10 in Appendix B shows that one-third of the municipalities in the provinces of Catania, Palermo and Ragusa fall into the fourth quartile, indicating high environmental risk.

The Reaction Capacity Index ($I_{RCI}$) was estimated using the final level of the hierarchical model structure presented in the “Methods” section (Section 4) and illustrated in Figure 10a.

The $I_{RCI}$ provides an aggregate measure of territorial capital endowment for various Sicilian municipalities.

It can support decision-makers by helping them to identify areas of weakest and measure the territorial rebalancing effects of ex-ante and ex-post cohesion policy strategies.

It can also help to identify of new strategies and financing packages that are tailored to specific contexts and encourage a medium and long-term vision for territorial development.

As can be seen in Figure 10b the level of territorial capital endowment is generally higher in coastal areas, especially in eastern Sicily.

The provincial of Enna, particularly the municipality of the same name, is the only inner area a high level of territorial capital. In general, inner areas are characterised by low to medium levels of territorial capital.

As shown in (see Figure 11), the quartile analysis indicates that a high proportion of municipalities in the provinces of Agrigento, Caltanissetta and Enna are in the first quartile. This highlights the low recovery capacity of these areas, which are generally weaker.

In this case, a strategy aimed at revitalising and enhancing these areas would require significant investment. By contrast, one third of the municipalities in the provinces of Catania, Ragusa, Syracuse and Trapani fall into the fourth quartile, indicating a high level of territorial capital.

Therefore, a strategy aimed at strengthening these municipalities would require less investment, given their greater capacity to react.

$I_{RCI}$ mapping confirmed the traditional contrast between coastal and inner areas. Although inner areas have a low $I_{RCI}$ level, they have good levels of other forms of capital, such as urban and natural capital, as highlighted in the previous paragraphs.

However, these areas are undervalued, primarily due to the low level of infrastructure and economic capital.

The cognitive model, which is based on the estimation of aggregate indices for the six forms of territorial capital and the $I_{RCI}$, has shown that classifying inner areas according to the NSIA based on the criterion of peripherality in relation to the municipal pole does not provide an exhaustive classification for characterising the different municipalities.

Estimating the indices showed that areas classified as F-Ultra-peripheral or E-Peripheral can have higher levels of specific forms of territorial capital or $I_{RCI}$ than areas in other classes. This highlights that the peripherality criterion fails to capture real differences between areas.

6. Discussion

A synoptic analysis of the indicators for the 11 Sicilian NSIA areas reveals how endowed the various forms of territorial capital are, and how well they can react.

Figure 12, Figure 13, Figure 14, Figure 15, Figure 16 and Figure 17 show the levels of the six types of territorial capital and the $I_{RCI}$ for the 11 NSIA areas in Sicily. All Sicilian NSIA areas are characterised by low levels of infrastructure capital, and similar levels human and economic capital. There are minor differences in urban and environmental capital levels, and more significant differences in natural capital levels.

The NSIA areas of Bronte, Madonie, Nebrodi, and Val di Simeto have a greater endowment of natural capital. The Nebrodi Regional Park is located in the NSIA areas of Bronte and Nebrodi. The Madonie Regional Park is located in the Madonie NSIA area, and the Simeto Nature Reserve is located in the Val di Simeto NSIA area.

These areas are of great landscape and natural importance. They represent a valuable natural heritage for the whole of Sicily.

The cognitive model and index structure proposed in this study can support decision-makers in assessing the effects of policies and actions aimed at promoting territorial rebalancing [146], in both the ex-ante and ex post phases.

They can help decision-makers to identify and manage the trade-offs generated by these policies and actions, while also promoting a medium- and long-term development strategy.

For instance, with reference to the NSIA Troina area, we set out below an application of the index system to help decision-makers monitor the impact of cohesion policies over the two programming cycles, 2014–2020 and 2021–2027.

The Troina area’s development strategy is based on three sub-strategies:

• Development and strengthening of essential services.
• Strengthening mobility within the area.
• Promoting heritage identity and strengthening the competitiveness of SMEs.
• Protecting the environment using ecosystem-based approaches.

The first sub-strategy has been developed based on the following actions:

• Measures for the digitisation of local public administration.
• Measures to strengthen and improve school environments.
• Reorganisation and upgrading of local health services.
• Integration of services to promote work-life balance.

The second sub-strategy is supported by action and reference measure 1.3.1 of the 2021-27 ERDF PR Plan Programme, and by action and reference measure 5.2.1.18 of the ERDF 2021-27 Programme Plan. These measures demonstrate the synergy and complementarity of financial sources with the investment lines of the Sicily PR OS 1.3, the M1C2 PNRR, the PNRR M5C2 OS 6, the PN Culture 2021–2027, and the PR Sicily 2021–2027.

It has been developed with reference to the list of proposed actions below:

• Promotion of entrepreneurship through supporting for the creation of new SMEs.
• Promotion of new investments to increase competitiveness.
• Enhancing of public spaces to promote tourist and residential appeal.
• Interventions for the preserving and enhancing of historical and cultural heritage.
• Integrated measures aimed at enhancing and promoting the use of natural areas.
• Redevelopment measures for historic centres and open spaces.

The third sub-strategy has been developed based on the following actions:

• Measures to encourage the creation of energy communities.
• Measures to improve the integrated water service at every stage of the supply chain.
• Emergency management and risk prevention measures.
• Interventions for the construction of infrastructure, equipment, and resources for waste management, collection, reuse, and recycling.
• Protection of areas falling within Natura 2000 sites.
• Green infrastructure and the creation of urban forests.
• Adoption of technological solutions to reduce energy consumption in public buildings and facilities, as well as in public lighting networks.

Figure A11, Figure A12 and Figure A13 in Appendix B show the funding that supported the strategies for Troina’s inner area across the various programming cycles.

The interventions that have been carried out or planned have different impacts on enhancing the various forms of territorial capital.

According to the cognitive model proposed in this study, increases in the endowment of capital forms can be measured and the effects on individual municipalities can be highlighted by estimating the six territorial capital forms indices and the aggregate $I_{RCI}$ index.

The cognitive model allows the identification of changes in territorial capital levels and the responsiveness of various municipalities resulting from the interventions carried out and those planned for the 2021–2027 cycle.

Figure 18 shows the estimated scenarios for the municipalities in the NSIA area of Troina.

The solid line shows the baseline scenario, and the grey dotted line shows the project scenario. The latter incorporates the planned changes to the area’s territorial heritage resulting from the interventions envisaged in the Strategy for this Area (SA).

Some interventions cover all municipalities in the NSIA area of Troina, while others affect individual municipalities. In the former case, the indicators show a minimal increase, as represented by the dotted area in Figure 18.

In the latter case, there is an increase in the endowment of various forms of territorial capital, as well as an increase in the capacity of individual municipalities to react.

The model highlights and measures changes in territorial capital stock both ex ante, to assess the strategy’s effects on the relevant area and municipalities, and ex post phase, to verify the achievement of the strategy’s objectives.

The model can therefore support decision-makers in developing the NSIA strategy [147,148,149,150] and promote a culture of control and monitoring to ensure that the strategy’s objectives are met. This facilitates more effective and efficient management of promoting the revitalisation of these areas and the related funding [151,152].

The mapping of inner areas proposed by NSIA is now the official tool for tackling the marginalisation of these territories. Italy is the first Europe country to address the issue of marginalised areas, developing strategies to revitalise them based on solid knowledge.

This knowledge base provides the foundation for the allocation of financial resources provided for in the various funding packages outlined in the Italian context with reference to cohesion policies developed at European level.

The NSIA’s proposed mapping of inner areas, based on a classification of municipalities according to their remoteness, offers a unified and systematic view of the issue across Italy.

However, it provides limited information, failing to capture the values and potential of these areas, as highlighted Trovato and Nasca (2022) [74].

In other European countries, the issue of marginalised areas is primarily addressed through thematic programmes. Examples include the French programmes “Revitalisation des centres-bourgs” (2014) [153], “Action Cœur de Ville” (2018) [154] and “Petites Villes de Demain” (2020) [155]. In 2008, Switzerland adopted a New Regional Policy (NRP) [156] to support mountain and border regions, rural areas and European territorial cooperation, in line with the Interreg, ESPON, URBACT and INTERACT programmes [157].

In 2016, Spain proposed the “Castilla-La Mancha programme” [158], and in 29 March 2019 approved the General Guidelines for a National Strategy to Address the Demographic Challenge [159].

This strategy maps demographic changes, such as depopulation, and ageing, and the effects of population fluctuation. Spain is also developing a programme called “El plan de España para evitar la despoblación rural” (Spain’s plan to prevent rural depopulation).

This study presents a model that is based on a critical review of the NSIA classification. Referring With reference to the traditional classes of territorial capital components [75], the model supports the development indices that promote an area’s value based on endogenous factors rather than its.

The main limitation of this research relates to the quality of the official databases often provide data from different time periods that is not up to date.

Furthermore, in order to evaluate the strategies and effects generated by funding using index-based metrics, it is necessary to improve data quality and clearly quantify the general and specific objectives for the various thematic areas of the strategies.

This would enable the effectiveness of the adopted strategies and the efficiency of the allocated resources to be measured.

Possible developments of this research will involve further investigation and verification of the proposed cognitive model. In particular, we are developing research on the “Southern question”, and we will therefore extend the implementation of this model to the regions of southern Italy.

The proposed cognitive model can be replicated for all Italian regions, as all the conditions for its applicability are met.

In particular, the datasets used to identify indicators have national coverage, despite the limitations mentioned above.

Further developments in this area of research will focus on implementing the cognitive model in other European contexts, particularly in French municipalities located in the so-called “Diagonale du vide”.

These municipalities are the French equivalents of those in inner areas. It will be interesting to test the cognitive model on this French case to verify its suitability for analysing a context that, while similar, has different characteristics.

Future developments of this research will focus on integrating other components of territorial capital into the cognitive model.

These components can be identified using the five intermediate classes of the rivalry-excludability model proposed by Camagni in 2008 [75].

The research may also address the question of how endogenous factors can be stimulated exogenously.

Integrating other components of territorial capital into the cognitive model, as set out in the five intermediate classes of the rivalry-excludability model proposed by Camagni, could enhance its effectiveness. This could strengthen the model and facilitate its application in complex contexts characterised by local policy heterogeneity, differentiated fiscal flexibility, various potential public/private partnership scenarios, and varying capacities to translate geographical and cognitive proximity into effective links between economic agents.

Currently, the designation of NSIA areas depends on the ability of local authorities to establish a network [160,161]. This mechanism is based on spatial proximity and the convergence of objectives.

Future developments in the Reaction Capacity Index ($I_{RCI}$) and the value-based knowledge approach could facilitate the identification of NSIA areas more effectively [162].

Future developments in this area of research may involve using the Spatial Cycle Model (SCM) [34] to classify inner areas. This model is instrumental in characterising long-term spatial demographic change.

However, this model cannot currently be used due to the lack of time series data for these areas, making it impossible to characterise the phases and sub-phases of the cycle.

7. Conclusions

In recent years, the marginalisation of areas due to progressive social, material, economic and infrastructural vulnerability in urban suburbs, small towns, villages and rural inland areas has become a pressing issue in many countries. This is primarily due to populations, functions, services, infrastructure, technologies and knowledge being concentrated in urbanised and/or coastal areas.

The 2030 Agenda emphasises the importance of identifying measures to support underdeveloped regions, particularly in relation to Goal 11 of the SDGs.

The sustainability goals set out in the 2030 Agenda have raised awareness of this complex issue affecting many regions.

As a result, many researchers and politicians have recognised that this issue cannot be postponed any longer. Research has promoted a renewed awareness of territorial rebalancing and sought to outline measures to tackle this complex problem [163,164]. The concept of peripherality is no longer defined by spatial factors, such as distance and proximity to essential services and opportunities.

Instead, it is now linked to socio-economic performance, political dynamics and the absence of organised forms of proximity—social, legal and institutional—all of which are important factors in identifying new forms of interaction. European cohesion policies promote, and support actions aimed at territorial rebalancing.

The implementation of these policies and their outcomes are closely linked to the definition of the action domain. This domain is represented by the territory and, more specifically, its components.

A key aspect of defining and implementing these policies is the cognitive model of the territory, particularly with regard to the endogenous factors referred to in the literature.

Italy was the first European country to produce a survey of the state of marginalisation in its territories, based on the NSIA classification and a unified knowledge framework.

However, this does not provide an exhaustive representation of exogenous factors, which are a key consideration in the development and implementation of policies aimed at rebalancing and revitalising territories.

This cognitive model is weakened by the partial integration of the representation of exogenous factors, and particularly by the absence of a metric capable of measuring their level.

In this regard, the study proposes a cognitive model based on the concept of territorial capital.

This model identifies the components of territorial capital according to the classic rivalry-excludability approach proposed by Camagni in 2008 [72].

The concept of territorial capital is used to characterise exogenous factors.

The cognitive model was developed from the perspective of an axiological approach to territorial capital.

The proposed model is called the “Geo-referenced Value-based Knowledge Model”. It develops a metric that expresses the level of endowment of various forms of territorial capital, as well as the overall endowment. This is based on the estimation of indices.

The study develops a metric based on the estimation of indices for the different forms of territorial capital ($I_{HC}$, $I_{UC}$, $I_{IC}$, $I_{EC}$, $I_{NC}$, $I_{EnvC}$) and the territory’s capacity to react ($I_{RCI}$). This provides an aggregate measure of the territory’s endowment of territorial capital components.

The indices were developed using the Multi-Attribute Theory Value (MAVT) multi-criteria approach.

The cognitive model was developed for Sicilian municipalities and so can also be used for NSIA municipalities.

Estimating the six indices ($I_{HC}$, $I_{UC}$, $I_{IC}$, $I_{EC}$, $I_{NC}$, $I_{EnvC}$) for the different forms of territorial capital, as well as the aggregate index ($I_{RCI}$), provides an assessment of the level of each form of territorial capital, as well as an overall assessment of territorial capital for all Sicilian municipalities.

Analysis of the quartiles of the indices, developed using QGIS, reveals different clusters of municipalities based on four levels of capital endowment: low, medium, medium-high and high.

These are characterised by the first, second, third and fourth quartiles of the estimated indices, respectively.

Analysing the indices by quartile, province and municipality type, according to the NSIA classification (i.e., A—Pole, B—Intercommunal Pole, C—Belt, D—Intermediate, E—Peripheral and F—Ultra-peripheral), provides detailed support for understanding the levels of territorial capital in Sicilian municipalities, and allows comparisons to be made between the knowledge model based on the NSIA classification and the new model.

The maps of the six indices and the index ($I_{RCI}$) highlight the distribution and level of various forms of territorial capital, as well as the overall endowment of Sicilian municipalities.

The estimate, quartile analysis and map of the human capital index $(I_{HC})$ reveal the level and distribution of human capital in Sicilian municipalities.

The following results were highlighted in relation to human capital:

• The highest levels are found in the eastern areas of Sicily, while the lowest levels are found in the central areas.
• The lowest level is found in the provinces of Agrigento and Caltanissetta. Investment is needed to support the development of municipalities in these provinces.
• The highest level is found in the provinces of Catania and Palermo.

These results, with reference to provinces and types of municipalities according to the NSIA classification, highlight:

• The municipalities with the worst endowment, i.e., those in the first quartile of the index, are in the provinces of Agrigento and Caltanissetta. According to the NSIA classification, most of them are classified as D—Intermediate, E—Peripheral or F—Ultra-peripheral. This shows that even municipalities considered less peripheral under the NSIA classification, such as those in class D (Intermediate), have low human capital endowment. Furthermore, it shows that municipalities in different NSIA classes have the same level of human capital endowment.
• The municipalities with the best endowment, falling within the fourth quartile of the index, are in the provinces of Catania and Palermo. According to the NSIA classification, they fall into all NSIA classes. This suggests that the NSIA classification, and therefore the differentiation into different classes based on the criterion of peripherality, does not accurately reflect their value. In this case, given that these results concern the two main Sicilian provinces, it is likely that some of the more peripheral areas benefit from “organised proximity”, which allows them to mitigate the effects of their geographical peripherality.

The estimate, quartile analysis and map of the urban capital index $(I_{UC})$ reveal the level and distribution of urban capital in Sicilian municipalities.

The following results were highlighted in relation to urban capital:

• A fairly diverse distribution of urban capital among municipalities in central Sicily. Inner areas in central Sicily have a good level of urban capital. This suggests that the classic contrast between inland and coastal areas does not apply in terms of urban capital.
• A better distribution in the southern and central-northern areas.
• Approximately one third of municipalities with a low level are located in the provinces of Catania and Syracuse.
• Approximately one third of municipalities with a high level of urban capital are located in the provinces of Caltanissetta, Enna, Palermo, Ragusa and Trapani.

The estimate, quartile analysis and map of the infrastructure capital index $(I_{IC})$ reveal the level and distribution of infrastructure capital in Sicilian municipalities.

The following results were highlighted in relation to infrastructure capital:

• A high concentration in coastal areas, particularly in the provinces of Catania, Palermo, Ragusa, Syracuse and Trapani.
• A medium and low concentration in inner areas.

These results, with reference to the provinces and types of municipalities according to the NSIA classification, highlight:

• A high percentage of municipalities with a low level in the province of Agrigento.
• Approximately one third of municipalities with a high level fall within the province of Catania.

The estimate, quartile analysis and map of the economic capital index $(I_{EC})$ reveal the level and distribution of economic capital in Sicilian municipalities.

The following results were highlighted in relation to economic capital:

• There are extensive areas with low and medium levels, mostly in inner areas.
• There is a concentration of medium-high and high levels in coastal areas.

These results, with reference to provinces and types of municipalities according to the NSIA classification, highlight:

• A high percentage of municipalities with low levels in the province of Messina.

The estimate, quartile analysis and map of the natural capital index $(I_{NC})$ reveal the level and distribution of natural capital in Sicilian municipalities.

The following results were highlighted in relation to natural capital:

• There are extensive areas with a low level in the southern areas.
• There are extensive areas with a medium-high level in the central areas.
• There are extensive areas with a high or medium-high level in the eastern and western areas.

These results, when considered in relation to provinces and types of municipalities according to the NSIA classification, reveal the following:

• A high percentage of municipalities with a low level in the province of Caltanissetta.
• A high percentage of municipalities with a high level in the province of Syracuse.

The estimate, quartile analysis and map of the environmental capital index $(I_{EnvC})$ reveal the level and distribution of environmental capital in Sicilian municipalities.

The following results were highlighted in relation to environmental capital:

• The areas most exposed to environmental risk are in northern and eastern Sicily.
• The areas least exposed to risk in central, southern and western Sicily.

These results, with reference to the provinces and types of municipalities according to the NSIA classification, highlight the following:

• A high percentage of municipalities in the province of Enna are characterised by low environmental risk.
• One third of municipalities in the provinces of Catania, Palermo and Ragusa are characterised by high environmental risk.

The estimate, quartile analysis and map of the Reaction Capacity Index $(I_{RCI})$ reveal the level and distribution of the overall territorial capital endowment for Sicilian municipalities.

The following results were highlighted in relation to the level and distribution of the overall territorial capital endowment:

• The municipalities with the highest levels of territorial capital are those located on the coast, especially in eastern Sicily, and those in the municipality of Enna.
• The municipalities characterised by low and medium levels of territorial capital are those located inner areas.

These results, with reference to the provinces and types of municipalities according to the NSIA classification, highlight the following:

• A high percentage of municipalities in the provinces of Agrigento, Caltanissetta and Enna fall within the first quartile of the index I_RCI, meaning they are characterised by a low level of territorial capital. These are the most vulnerable areas. A strategy aimed at revitalising and strengthening these areas would require significant investment.
• One third of the municipalities in the provinces of Catania, Ragusa, Syracuse and Trapani fall into the fourth quartile of the index I_RCI, meaning they have a high level of territorial capital. These are the least vulnerable areas. In this case, a strategy aimed at strengthening these areas would require less investment.

A synoptic analysis of the municipalities in the Sicilian NSIA areas was conducted, using the results on the six indices relating to various forms of territorial capital and to for $I_{RCI}$.

This analysis highlighted the following:

• A low level of infrastructure capital.
• A fairly similar level of human and economic capital.
• Some minor differences in urban and environmental capital.
• More marked differences in natural capital.
• There is greater natural capital in the NSIA areas of Bronte, Madonie, Nebrodi and Val di Simeto.

For example, a simulation was proposed for the NSIA area of Troina to show how the proposed knowledge model based on indices metrics could help decision-makers assess the impact of policies and actions aimed at promoting territorial rebalancing, both before and after implementation.