Digitalisation as a Challenge for Smart Villages: The Case of Poland ^†

Abstract

Rural areas face several development challenges. Some lead to rural decline—such as depopulation—and others are intended to counteract this by revitalising the countryside—such as digitalisation. These two processes are on the agenda of the European Union’s new rural development concept, smart villages. The study aims to identify the spatial differentiation of the digitalisation challenge in Polish rural areas. An attempt was made to operationalise two aspects of this challenge—access to fast Internet and digital competence. The subject of the analysis covered rural areas in Poland at the municipal level. The temporal scope of the study is defined by two approaches—static and dynamic. The first aims to show the state of the ‘here and now’, while the second aims to identify the change intensity. Methods of multivariate comparative analysis were used, resulting in hierarchical classifications of municipalities. The results show significant regional differentiation. Municipalities with predominantly ageing populations face greater difficulties in adopting new digital technologies. Overcoming these disparities will be key to improving the quality of life and resilience of rural communities. The study results provide evidence to justify the need for place-based targeted digital investments under the smart villages programmes.

1. Introduction

Rural areas, especially those peripherally located, are prone to the accumulation of unfavourable socio-economic development conditions, leading to the phenomenon known in the literature as rural decline [1,2,3,4]. Administrative units that experience both a decreasing and ageing population face difficulties in completing public tasks and municipal investments [5]. This results in reduced attractiveness for potential investors, which further limits the local labour market, thereby causing a further exodus of residents and a deterioration of the demographic structure, leading to a repetition of this negative cycle, and creating an almost self-perpetuating circle effect [6]. This process can be presented as a diagram of a so-called vicious circle [7] (Figure 1).

Due to their spatial extent, rural areas are subject to the intervention of diverse public policies at different levels of governance. These policies mostly propose tools that would be able to mitigate the effects of rural decline by adapting to the changes taking place. An example of such actions at the supranational level is the ‘Rural Policy 3.0’, popularised for several years by the Organisation for Economic Cooperation and Development (OECD) [8]. It presents a framework for action to address the challenges of the 21st century, which include an ageing population, migration, urbanisation, changes in global production processes, the development of emerging economies, climate change, and increasing environmental pressures, as well as technological advances. The OECD is guided by the fact that rural areas constitute a diverse socio-economic system in which the response to development challenges depends on the specific needs and capacities of local communities, comprised of diverse actors linked by a multi-level governance mechanism. The 11 Principles for the Implementation of ‘Rural Policy 3.0’ [9] indicate that countries’ actions should be based on identifying unique rural assets, optimising the scale of policy implementation, and strengthening the social, economic, environmental, and cultural resilience of local communities [10].

The above assumptions are fulfilled by the new development concept, smart villages, which has been developed since the second half of the previous decade. It is of particular interest in the European Union (EU), where, under the aegis of the European Commission, it has progressed from a slogan/idea to a development instrument within the Common Agricultural Policy (CAP) for 2023–2027 [11]. The EU approach emphasises that local communities in smart villages “build on their strengths and assets and develop new opportunities; traditional and new networks, as well as services, are improved through digital technologies, telecommunications, innovation, better use of knowledge, for the benefit of residents and businesses; digital technologies and innovation can support a higher quality of life, standard of public services, better use of resources, lower environmental impact and new opportunities for rural value chains in terms of products and improved processes; the smart rural concept does not propose a one-size-fits-all solution—it is territorially sensitive, based on the needs, potential and strategy of the territory” [12] (p. 3). The concept particularly emphasises the importance of technological change, i.e., this dimension of rural decline, which is addressed much less frequently in rural research [13,14,15] than the demographic aspect of the phenomenon [16,17,18,19,20]. The aspect of technological changes still needs to be recognised from both theoretical and operationalisation perspectives. This is because in the modern world, they are progressing at a pace that was difficult to imagine just a few decades ago. They have many dimensions and touch virtually every area of human life, bringing many opportunities and benefits. However, their spread is not geographically even.

The technological changes in question are characterised by the fact that they require a constant adaptation of infrastructure to changing digital tools. As early as the 1980s, telecentres, which, in the context of rural areas, were known as telecottages, began to be implemented around the world. These were physical locations where residents could access the ICT tools available at the time, such as landlines, fax machines, and, later, computers and the Internet [21]. This innovation aimed to reduce the geographical isolation of less populated rural areas from the cities by meeting information, educational, and professional needs. Since the beginning of the 21st century, the sense of such places has been diminishing, as fixed Internet, within the reach of individual households, has begun to spread. Today, in the third decade of this century, the Internet is a normality, making it possible to shorten the geographical distances of peripheral areas. It plays a key role in the concept of smart villages, providing the foundation for this approach [19,22,23]. This raises concerns about the risk of digital exclusion for those areas that lack this foundation, both spatially (lack of Internet access) and socio-economically (low digital competences due to unfavourable demographic structure and low education levels) [24].

This article focuses on the digitalisation dimension of rural areas, which, in light of the smart villages concept, is one of the prerequisites for implementing this approach. At the same time, it is a challenge for decision-makers at different levels of governance, as it involves both major infrastructure investments and educational activities. These, however, due to the diversity of rural areas in many socio-economic respects, have so far been implemented with varying degrees of success.

This topic was also chosen because the issue of digitalisation is underrepresented in rural research. One reason for this is the insufficient access to data on digitalisation at the local level. Many times, as in this study, it requires a novel approach and a search for measures that explain a phenomenon or process multidimensionally, often using a proxy with adequate explanatory power in relation to the indicatum. This study is an attempt to operationalise the problem of digitalisation as a rural development challenge in light of the new concept of smart villages. The aim of the study is to identify the spatial differentiation of the challenge of digitalisation in rural areas using the example of Poland. The author poses the questions: (1) Is the digitalisation of rural areas in Poland spatially differentiated? (2) Are any spatial patterns of this differentiation visible? (3) What relation connects the challenge of rural digitalisation with the processes referred to as rural decline?

The subject of the study was located in rural areas in Poland, as the studied country is characterised by strong spatial differentiation of rural development [25], which makes it a good field for observing the intensity of development challenges. The second important argument is that Poland has been one of the most active countries in the work on the smart village concept within the CAP for 2023–2027 [26,27].

2. Literature Review

2.1. Digitalisation of Rural Areas Under the Smart Growth Concept

Due to its short history, the concept of smart villages remains under-explored theoretically. In order to show its essence, the relationship between its assumptions and those of other development concepts can be used. In the case of the issue of digitalisation, the concept of smart rural development can be used as a descriptor.

The concept of smart rural development takes its origins from the much older concept of smart growth, which figured prominently in the economic policies of the EU, as its tenets were based on promoting knowledge and innovation in the context of EU regional development [28,29,30]. According to Naldi and co-authors [31], the concept implies that a move away from a one-size-fits-all approach in regional policies, which have so far mainly targeted areas that are growth centres, towards a territorially differentiated (place-based) approach may better suit the specificities of peripheral rural areas [32,33]. The essence of the concept is contained in the definition, according to which, smart growth is otherwise “the economic progress achieved following innovation, education and research activities in rural areas. In this perspective, the role of public institutions is seen in the creation of conditions for the implementation of the aforementioned types of activity through the creation of an appropriate legal framework, promotion of intersectoral cooperation among economic entities, non-governmental and public institutions or co-financing of various undertakings with public funds” [34] (p. 59). Thus, the concept of smart growth is embedded in a broader context, including not only regional aspects, but also other relevant social processes and phenomena. This includes, among others, the development of human and social capital, demographic changes, and the challenges of technological change—aspects that can provide an advantage as well as drive the vicious circle of rural decline.

In explaining the emergence of negative development processes in rural areas, it is necessary to draw on the spatial regularities of development observed and described on the basis of the theory of polarisation, according to which, it generates inequalities that lead to the concentration of resources in growth centres and their outflow from areas located further from these centres (peripheries) [35]. This is related to the following so-called smart rural development factors [31].

○

Agglomeration effects: They are related to the concentration of population, infrastructure, and economic activity in growth centres, which foster innovation and accelerate development processes. Similar effects can occur in rural areas as long as they reach a sustainable critical mass of development, which enables the generation of innovation and the building of territorial competitive advantages [36,37,38].

○

Amenities and creative rural economies: Rural areas can outperform cities in terms of diversity of assets, which can attract the so-called creative class, enabling them to compete with metropolitan regions [39,40,41]. Although this process is complex and takes a long time, it is achievable with the right configuration of amenities and high-quality digital connections.

○

Collaborative networks: Due to their geographical remoteness and low population density, rural areas have limited access to knowledge, which is crucial for the development of innovation and local specialisation. Overcoming this barrier requires the development of high-speed Internet, especially in peripheral regions, which supports networking and ultimately fosters new products and services.

One concept linked to smart rural development is digital transformation, which is the process of introducing digital technologies into society and the economy. A key tool in this process is digital infrastructure. It shows the potential to break down barriers arising from the centre–periphery relationship by harnessing both internal and external development impulses. The process of rural digital transformation can be divided into five key stages (Figure 2).

The ‘excluded’ stage is characterised by very limited mobile and broadband access and low levels of digital literacy among residents. In this stage, there is a low level of availability of public–private services and low social inclusion. This results in agglomeration effects that strongly polarise space, reinforcing the dominance of metropolitan areas at the expense of peripheral regions.

In the ‘connected’ phase, basic broadband is provided, and residents begin to develop digital competences. Different stakeholders such as local communities, authorities, and the private sector start to work together to set development priorities. This is a formative stage that is highly dependent on external funding, such as from national and regional programmes.

The ‘committed’ stage is characterised by the widespread presence of digital skills in the community, with access to local training, education, and public and business services becoming widely available. In this stage, local stakeholders play a key role and are actively involved in the design, implementation, and monitoring of services. Networks between these stakeholders are being strengthened, resulting in an increase in the quality of human resources and a reduction in dependence on external funding, which was previously needed for large infrastructure investments.

In the ‘experienced’ phase, rural areas are entering a higher level of digital infrastructure with next-generation fibre broadband. Both residents and local businesses show a high capacity to exploit digital and social innovations. The countryside is becoming an active participant in the digital territorial transformation, offering services at a level comparable to those available in metropolitan centres. This level of development enables rural units to remain sustainable while enhancing their attractiveness in the eyes of external stakeholders.

The last is the ‘player’ stage, when the village reaches its full technical, economic, and social capacity to create innovation and the digital economy. The local community becomes a co-owner of its data and the added value generated, transforming residents and entrepreneurs into equal partners in the design of new services and products. Here, the countryside achieves a level of development characteristic of smart regions, capable of creating local agglomeration effects and building resilience to various types of crises.

2.2. Evolution of the Smart Village Idea in the European Union

In the second half of the previous decade, the European Commission started to work on a new approach to rural development with the announcement of the pilot ‘European Union Action for Smart Villages’ [12]. The document, published in 2017 by the European Commissioners—Phil Hogan (Commissioner for Agriculture and Rural Development), Corina Crețu (Commissioner for Regional Policy), and Violeta Bulc (Commissioner for Mobility and Transport)—presented a definitional framework for the concept of ‘smart villages’ and determined directions for further action to support them. Among other things, the document clarified that the concept refers to rural areas and local communities and suggested that these communities should build on their existing assets and make efforts to expand them, particularly with regard to digital technologies, innovation, and knowledge. It was further specified that the objective of a smart village is to improve both traditional and new networks and services, which will translate into a higher standard of living, including public services, a more optimal use of resources, and less negative impact on the environment, and open up new opportunities for rural value chains—in general, it will contribute to improving the quality of life of rural people.

The origins of the concept can be traced back to the 1990s. As presented in the first Cork Declaration [43] entitled ‘A Living Countryside’, which resulted from the European Conference on Rural Development in Ireland, the key role of integrating advances in information technology in sustainable rural development was identified. Alongside this, the importance of adequate infrastructure, training, and education was highlighted. Even then, public participation and bottom–up activities were highlighted as important factors in community-led local development.

Twenty years later, in the second Cork Declaration of 2016 [44] entitled ‘Better Life in Rural Areas’, which set new directions for rural and agricultural policies, one of the ten points was dedicated to investing in the viability and vitality of rural areas. A key action in this respect was to bridge the digital divide between rural and urban areas, both in terms of access to a good quality Internet network and the digital competences of rural residents. This declaration had a huge impact on the subsequent work on the smart villages concept.

Important events for the development of the concept include, for example, a meeting on the future of smart villages in Bled, Slovenia, in 2018. Its outcome was the Bled Declaration [45] for a smart future for EU rural areas. The signatories of this declaration stated that an innovative, integrated, and community-inclusive development of the digital economy in rural areas has the potential to improve the quality of life of their inhabitants and to increase economic and social cohesion between rural and urban areas. The body of the document also lists technological potentials that could be the flywheel of smart development, including precision agriculture, digital platforms, the sharing economy, the closed-loop economy, and rural tourism, among others. The authors of the declaration stressed that while digital technologies are an important tool for the development of smart villages, they are not the only element. Equally important are human resources, which require mobilisation and synergy of action. In order to achieve this synergy, coordinated investments in digital infrastructure and in the competences of rural people are necessary (according to the signatories of the second Cork Declaration of 2016).

Another noteworthy voice in the evolution of the concept under consideration is the European Rural Parliament. Following the deliberations in Candás, Spain, in 2019, the Manifesto of the European Rural Parliament [46] was published. The document updates the content of previous documents developed during the Parliament’s deliberations in 2015 in Austria and 2017 in the Netherlands. The update included, among other things, the addition of the concept of ‘smart villages’ under the headings of policy frameworks, support for rural villages, and the development of communication and digitalisation. For the first time, the concept was addressed in a holistic way, combining the leading technological sphere with the political sphere (financial and systemic support) and the social sphere (responses to the socio-economic challenges of rural areas). The need for multi-level work to concretise the concept in the then-upcoming programming period was emphasised, including through the operation of national working groups on smart villages. The tasks of these groups would be to raise awareness among villagers of their needs and challenges and to develop ways of responding to them, taking into account the use of existing institutional structures such as LEADER. Among these challenges, the out-migration of people, the disappearance of services, and the lag in the development of information and communication technologies were specifically mentioned. Significantly, it places the responsibility for addressing the last-mentioned challenge on actors with real power to do so, such as governments, international funds, and telecommunications providers.

The latest declaration of the European Rural Parliament was announced in 2022 after deliberations in the Polish city of Kielce, taking place at a time when the assumptions of the CAP national strategic plans were already accepted by the European Commission or were in the final stages of being finalized. The declaration was made in a new geopolitical reality, including the COVID-19 pandemic and Russia’s aggression against Ukraine, which was reflected in the additional role assigned to smart villages. Indeed, the activities of the concept are intended to provide local communities with greater resilience to crises, assuming that they will implement unique solutions to local problems within the framework of civil society instruments—local development strategies, smart villages concepts, etc. [47].

The cited declarations and manifestos were mainly declarative and guided the activities undertaken in relation to the smart villages concept. However, they were not the only important points in the work on the concept. Among the most important initiatives is the activity of the Thematic Group on Smart and Competitive Rural Areas at the European Network for Rural Development (ENRD), which was active from 2017 to 2020. This group was a centre for the exchange of views and experiences from different EU countries and one of the main advisory bodies for the design of solutions for smart villages support within the CAP. The results of the Group’s work can be attributed to five key spheres of smart villages [11]: infrastructure, institutions, services, communities, and initiatives. This shows that the practical activities in relation to the concept were in line with its theoretical assumptions and took into account current socio-economic processes in rural areas. Indeed, it has been indicated that the drivers of smart villages are the following [7].

• Response to demographic change, particularly depopulation;
• Seeking local solutions to relieve the burden on the public budget;
• Using and strengthening the links between villages and cities;
• Enhancing the contribution of rural areas to the shift towards a low-carbon, closed-loop economy;
• Supporting digital transformation.

The recommendations developed by representatives of different sectors—social, governmental, and scientific—during an international workshop entitled ‘Smart villages as a solution to key rural challenges’, which took place in 2019 in the Polish capital Warsaw, played a key role in the discussion on creating support for smart villages. The recommendations mainly focus on the practical implementation of smart villages, rather than defining what exactly should be introduced. According to them, key actions should include the following.

• Build on existing experiences, such as those gained through the LEADER initiative and activities related to Village Renewal programmes. It is important to avoid creating new bureaucratic structures, as it is better to focus on developing existing solutions and involving local leaders in decision-making processes.
• Start with small projects (e.g., in one locality) and then expanding to work with external partners to address wider issues that may be difficult to tackle on a limited scale.
• Draw attention to the need for the intensive digitalisation of rural areas and the development of digital competences of the population, treating this as a priority and long-term task.
• Take advantage of the experience of local leaders and available human resources and, where these are lacking, seek to recruit and engage them.
• Promote innovative and unusual solutions to local problems that can serve as models for other communities and reward active and committed groups.

These recommendations are an important element in the design of support for smart villages, emphasising the need for a practical approach to implementing the smart villages concept, based on past experience, involving local leaders and resources and promoting innovative solutions to local problems.

A key strategic step towards supporting smart villages is the document, “A long-term vision for EU rural areas—towards stronger, better connected, resilient and prosperous rural areas by 2040”, published by the European Commission. It identifies four key areas for action to achieve the goals of building stronger, connected, resilient, and prosperous rural areas by 2040 (Figure 3). Within this vision, smart villages are seen as the foundations of community-based local development. Particularly important is to engage in research and innovation aimed at supporting rural communities, which is essential for creating connected (digitally and transportally) rural areas, especially in the context of developing services and public systems [48].

3. Materials and Methods

3.1. Definitions

Several key terms are used in the article and need to be defined. The first is the term, “challenge”, which is “a difficult task, a new situation, etc. requiring someone’s effort, dedication, etc., being a test of someone’s knowledge, resilience, etc.” [49]. This definition conceals the following three layers that allow a deeper characterisation of the term described.

• The challenge is subject to diagnosis, and, therefore, its identification is the result of monitoring the phenomenon in question (new situation);
• A challenge requires having the right resources to take action to respond to it (effort, dedication, knowledge);
• An effective response to a challenge depends on the circumstances in which these actions are taken (difficult task, resilience).

Another term is ‘digitalisation’, understood in the article as a process of providing easier access to information and services, which reduces the distance between urban and rural areas, enabling some activities to be conducted virtually. Digitalisation makes it possible to introduce innovative solutions and to compensate for the disadvantages of the rural periphery. This allows for the active use of human resources and the promotion of a diversified rural economy and provides new opportunities for work, education, and social life in rural areas [50].

Thus, the author assumed that the ‘challenge of rural digitalisation’ is a situation (phenomenon, process, trend) that requires urgent, appropriately measured action using local resources, as the absence of such action or its inadequate alignment threatens the transformation of the countryside into a resident-friendly environment.

3.2. Spatial Scope

The spatial scope of the analysis covers rural areas in Poland, defined by the administrative criterion, i.e., everything that lies outside the administrative borders of a city. The study is conducted at the local level—according to the Eurostat nomenclature of local administrative units (LAUs) (Figure 4). The analysis covers 2175 local government units, i.e., all urban–rural municipalities (including towns and villages: 652 units) and rural municipalities (including only villages: 1523 units) existing in 2021. All 302 urban municipalities, which, by definition, do not contain rural areas in their territories, are excluded from the analysis.

3.3. Time Snapshots

The study uses two complementary temporal snapshots. The first snapshot, a static one, aims to show the state of digitalisation at a specific point in time. The empirical indicators in the static approach correspond to the latest locally obtainable data on digitalisation. In the case of Internet access, they are taken from the Polish Office of Electronic Communications (UKE) and refer to the status as of 31 December 2019. In the case of the measure of digital competence, the status as of 31 December 2021 was examined according to the Statistics Poland (GUS).

The second approach—dynamic—is intended to show the direction of the changes taking place, i.e., the change between the two assumed time points. The time horizon in the case of Internet accessibility concerns the years 2019–2023. Such a scope results from the availability of data at the base point (state as of 31 December 2019) and at the end point—the value of this indicator projected as of 31 December 2023 by UKE. For the measure of digital competence, data from the base year of 2005 were taken into account. The use of a dynamic approach makes it possible to capture processes that are not discernible in a static approach. It also has its drawbacks, such as the association of indicator values with the baseline situation (base effect), occurring when indicator values are either very high or very low in the base year relative to the set average. The author is aware of these drawbacks and takes them into account when analysing the empirical data.

3.4. Operationalisation of Variables and Data Sources

There is a clear difference between rural and urban areas in terms of broadband access. Data for 2019 show that, on average, 30% of buildings in rural municipalities in Poland had Internet access of more than 30 Mbps, compared to 62% in urban municipalities [51]. This problem is particularly difficult to solve, as low population density in rural areas is associated with high infrastructure maintenance costs, and the lack of adequate economic incentives can limit the development of the local economy. This, in turn, leads to increased migration, which negatively affects the population structure, which is becoming older, and the spatial distribution of the population. The operationalisation of this challenge required looking not only at the infrastructural aspect, but also, as already highlighted in the Cork 2.0 Declaration cited earlier, the digital competence aspect. Together, the two are complementary to each other, as on the one hand, it is difficult to realise the potential of digitalisation without Internet skills, and on the other hand, even high digital competences are not enough in the face of poor Internet access. Both challenges are named as follows.

• Provision of digital infrastructure;
• Acquisition of digital competences.

The availability of data sources on the studied issue at the LAU level should be assessed as low. The only data that can be publicly obtained come from the Broadband Atlas of the UKE and concern the infrastructural aspect (connection speeds in municipalities). For the static snapshot, the building penetration rate of fixed Internet with a speed of at least 30 Mbps in 2019 was used, calculated as the ratio of the number of buildings within reach of the new generation network with a speed of 30 Mbps (NGA30) to the number of all buildings in the municipality. For the dynamic snapshot, UKE’s projected change in building penetration of fixed Internet with a speed of at least 30 Mbps from the end of 2019 described above to the end of 2023 was taken into account.

Data on the digital competence aspect are available at most at regional levels, with no urban or rural breakdown. The difficult-to-measure nature of this phenomenon has led to a search for a contextual, proxy measure. According to the methodology of social studies represented by Nowak [52], the researcher must sometimes infer the occurrence of certain phenomena from a so-called latent characteristic, which is directly unobservable. In such a situation, the existence of an indicatum cannot be ascertained by direct observation. Still, this existence must be justified indirectly based on measured correlations and theoretical assumptions. This is also the case with the measurement of digital competence at the municipal level. The chosen indicator is based both on certain theoretical assumptions as well as on the results of other studies presented below, which make it possible to demonstrate its relationship to the issue under study. Indeed, as Wickens [53] (pp. 760–761) proved, “it is better to use even a poor proxy than to use none at all and omit the unobservable variable”.

According to ‘Social Diagnosis’ data for 2015, an average of 92% of 16–44-year-olds used the Internet (16–24-year-olds: 97.5%; 25–34-year-olds: 92.9%; 35–44-year-olds: 85.6%), while in older groups, the percentage was already significantly lower at 60% for 45–59-year-olds; 40.8% for 60–64-year-olds; and 17.9% for older ones [54] (p. 383). Data from the Centre for Public Opinion Research for 2022 [55] (p. 2) present a similar distribution in the use of this medium: 18–34 years: 100%; 35–44 years: 96%; 45–54 years: 86%; 55–64 years: 65%; 65–74 years: 41%; and older ones: 22%. The Polish Ministry of Digital Affairs [56] calculates that as many as 78% of those not using the Internet in the 16–74 age group are over 60. Half of the non-users cite a lack of skills as the main reason. Moreover, according to Woźniak-Jęchorek [57], from the perspective of data on the use of the Internet, the groups with the highest digital exclusion are the oldest people and, to a lesser extent, people with a low level of education and people with a poor material situation. This regularity is also confirmed by the results of the Information Society Survey conducted by the Statistics Poland [58]. Overall, numerous studies confirm that the older the age group, the less frequently people in this group use the Internet; they are also likely to have lower digital skills and to generally perform fewer activities using the Internet [54,55,56,57,58,59,60].

It was decided to take into account the relationship between two subpopulations within the working-age group. The first subpopulation consists of individuals aged 18–44, referred to in statistics as the mobile age group. The second includes individuals aged 45–64 for men and 45–59 for women, referred to as the non-mobile age group. Therefore, a measure reflecting digital competence was considered to be the ratio of the number of people of immobile age to the number of people of mobile working age in 2021. This substitute measure is intended to show the share of people who, according to the above-mentioned research, have the lowest digital competences and constitute the lowest percentage of Internet users. Moreover, this measure shows a relatively high correlation with demographic indicators: (1) the proportion of the population in the post-working age group (r = 0.629) and (2) the ratio of people in the post-working age group to people in the pre-working age group (the so-called child–elderly ratio) (r = −0.701). Its explanatory value can be assessed as high, as it applies to the entire population (it is not based on a sample). In order to illustrate the dynamic approach, the indicator of change in the ratio of the number of people at immobile age to the number of people at mobile working age from 2005–2021 was used. The data were obtained from the Local Data Bank of the Statistics Poland (BDL GUS).

3.5. Statistical Analysis and Presentation Methods

Four empirical indicators were analysed, two for each challenge, including one for each time snapshot. Basic descriptive statistics are presented in

Table 1. Basic characteristics of digitalisation indicators.

Name of the Challenge	1. Provision of Digital Infrastructure		2. Acquisition of Digital Competences
Indicator number a	1.1 (IS)	1.2 (ID)	2.1 (IS)	2.2 (ID)
Indicator name	Building penetration of fixed Internet of at least 30 Mbps	Projected change in building penetration of fixed Internet of at least 30 Mbps	Ratio of people of immobile age to people of mobile working age	Change in the ratio of people of immobile age to people of mobile working age
Years	2019	2019–2023 (2019 = 100)	2021	2005–2021 (2005 = 100)
Unit	%	%	Person	%
Area b	2 & 3	2 & 3	2 & 5	2 & 5
Direction	Stimulant	Stimulant	Destimulant	Destimulant
Source c	UKE	UKE	BDL GUS	BDL GUS
Median	34.84	142.29	0.65	118.07
Average	39.75	196.70	0.65	119.56
Minimum	0.51	100.00	0.41	71.07
Maximum	98.91	647.00	1.03	189.58
Range	98.40	547.00	0.62	118.52
Standard deviation	23.61	129.50	0.06	14.26
Coefficient of variation	59.39	65.83	9.47	11.93
Correlation coefficient d	0.930	0.769	−0.480	−0.601

^a IS—Static indicator. ID—Dynamic indicator. ^b 2 and 3—Indicator calculated for the entire municipality (in the case of urban–rural municipalities including the city). 2 and 5—Indicator calculated for the rural area of the municipality (in the case of urban–rural municipalities without a city). ^c Acronyms for names of institutions explained in Section 3. ^d Spearman’s rho correlation coefficient between the indicator value and the value of the synthetic measure (in static or dynamic terms, depending on the type of indicator—IS or ID). Source: own study.

.

Indicator 1.1 (IS) is a stimulant, as higher values indicate better access to high-speed Internet, which positively contributes to digitalisation and technological progress. This indicator’s variability is relatively high, with a coefficient of variation of 59.39%, reflecting significant differences across the units analysed.

Indicator 1.2 (ID) is also a stimulant, as the projected increase in Internet penetration signifies ongoing infrastructure development and digital transformation. It also exhibits considerable variability, with a coefficient of variation of 65.83%, indicating notable differences in the projected growth rates across the regions.

Indicator 2.1 (IS) is a destimulant, as higher values suggest lower digital competences. The variability of this indicator is relatively low, with a coefficient of variation of 9.47%, indicating less variation between the observed units.

Indicator 2.2 (ID) is also a destimulant, as an increase in this ratio suggests a worsening demographic trend, which influences the achievement of digital competences across the local communities. The variability of this indicator is low, with a coefficient of variation of 11.93%, reflecting some differences in the phenomenon’s changes across the areas. In the case of indicators 2.1 (IS) and 2.2 (ID), low statistical variability does not reduce explanatory power, as the research approach involves an analysis of the entire population.

For the digitalisation challenge, summary indices were created to illustrate the intensity of the digitalisation challenge, relating to both time snapshots. Here, one of the taxonomic standardisation methods, known as zero-based unitarisation, was applied. The aim of this method is to transform the analysed data in such a way that they are comparable [61,62] according to the following formula:

$${{\mathrm{a}}^{\prime }}_{\mathrm{i}}={\displaystyle \frac{{\mathrm{a}}_{\mathrm{i}}-{\mathrm{a}}_{\mathrm{m}\mathrm{i}\mathrm{n}}}{{(\mathrm{a}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{a}}_{\mathrm{m}\mathrm{i}\mathrm{n}})}}$$

where:

a′_i—the ith normalised variable;

a_i—the ith normalised variable;

a_min—the minimum value of the variable in the set;

a_max—the maximum value of the variable in the set.

This allows for the obtaining of variables whose values lie within the range [0;1] and are characterised by relatively small information loss compared to the pre-normalisation data. The basis for the calculations is the difference between the maximum and minimum values of a variable in the dataset (range). For stimulants, i.e., features wherein an increase indicates an increase in the level of the studied phenomenon, normalisation is calculated by subtracting the smallest value in the dataset from the normalised value and dividing the difference by the range. For destimulants, i.e., features wherein an increase indicates a decrease in the level of the studied phenomenon, normalisation is calculated by subtracting the normalised value from the maximum value in the dataset and dividing the result by the range.

The normalised values in the interval [0;1] were used to develop synthetic indices in two time snapshots, synthetic and dynamic, according to the formula:

$${\mathrm{W}}_{\mathrm{i}}={\displaystyle \frac{1}{\mathrm{n}}}\sum _{\mathrm{j}=1}^{\mathrm{m}}{\mathrm{m}}_{\mathrm{i}}{{\mathrm{a}}^{\prime }}_{\mathrm{i}\mathrm{j}\mathrm{n}}$$

where:

a′_ij—the normalised value of the j-th characteristic in the i-th object;

n—number of objects, where: i—weighting factor of feature number i; $\sum _{j=1}^m{m}_i$ m = 1.

In the method used, it is possible to use varying weights for the variables, but it was considered that in this case, the strength of the variables is the same and equal weights were used—which is methodologically correct [58,61,63,64,65]. The synthetic measures developed take the form of a hierarchical classification, allowing the municipalities to be ordered on a scale from the highest to lowest position (from 1 to 2175). In the next step, these classifications were divided into five equinumerous classes, meaning that each class contains the same number of units (municipalities). This division was based on the level of intensity of the digitalisation challenge, distinguishing the following categories.

(1)

Very small challenge: the municipality’s position in the classification from 1 to 435;

(2)

Small challenge: position from 436 to 870;

(3)

Moderate challenge: position from 871 to 1305;

(4)

Big challenge: position from 1306 to 1740;

(5)

Very big challenge: position from 1741 to 2175.

The final result of the statistical analyses is an index of the accumulation of digitalisation challenges in the municipalities. It was created by superimposing a static synthetic measure and a dynamic synthetic measure, and then identifying the municipalities that fell into classes of ‘big’ or ‘very big’ levels of digitalisation challenge intensity in both of them.

The results of the analyses were presented using cartographic methods of presentation. In this case, a cartogram was used, which allows the intensity of a given phenomenon to be presented in administrative units. The thematic map and spatial analysis system MapViewer version 8.7 licensed by Golden Software LLC (Golden, CO, USA) was used for this purpose.

4. Results

4.1. Provision of Digital Infrastructure

Indicator 1.1 (IS), building penetration of fixed Internet of at least 30 Mbps, provides an assessment of rural areas’ access to relatively fast Internet. At the end of 2019, an average of 34.8% of buildings in the country were in NGA30 coverage. The reported availability values ranged from the lowest—0.51%—to the highest—98.91%. The best access to this network was in rural areas in western and southern Poland. Outside these areas, high-speed Internet is available in spatially limited clusters of municipalities in other parts of the country (Figure 5).

Poor NGA30 coverage mainly applies to central and eastern Poland. Exceptions are units located close to large cities, although this is not the rule.

Indicator 1.2 (ID), projected change in building penetration of fixed Internet of at least 30 Mbps, was used to measure the dynamics of digital infrastructure provision between 2019 and 2023, i.e., until the end of the disbursement of European funds under the Digital Poland Operational Programme 2014–2020.

On average, the increase in NGA30 access was projected to be 42.3%, reaching a median of 61.7% for rural areas at the end of 2023 (compared to 34.8% in 2019). However, in 182 municipalities (or 8.4% of the total set), no change is expected due to a lack of investment. Among these municipalities, there are 55 that perform below the national median in 2019 and where further investment in digital infrastructure seems to be a necessity.

In general, the highest growth rates are projected in the areas with the lowest penetration rates at the end of 2019 (cf. Figure 5 and Figure 6). However, this does not imply a full levelling off of spatial differences, as the distribution of penetration for 2023 will still be somewhat similar to that of 2019. This is evidenced by the relatively high correlation coefficient r = 0.669.

4.2. Acquisition of Digital Competences

The use of indicator 2.1. (IS), the ratio of the number of people of immobile age to the number of people of mobile working age, was justified in chapter 3. As it is a destimulant, it is desirable to keep its values as low as possible, which indicates a favourable relationship between the analysed age groups. The national average was 0.65 in 2021, meaning that for every 100 people of mobile working age, there were 65 people of immobile age. The lowest number recorded was 41 people, while the highest was 103 people.

The least favourable relations in the analysed age groups are found in the municipalities of eastern and central Poland, especially in the extreme regions of this part of the country (Figure 7). Relatively unfavourable indicators were also noted in units in north-western Poland.

In contrast, the most favourable relationship is found in most of the area in a belt from northern Poland, through the Wielkopolska region (western part), to the south-east of the country (Małopolska region).

Indicator 2.2. (ID), which illustrates the ratio of the number of people of immobile age to people of mobile working age between 2005 and 2021, was used to measure the dynamics of the challenge under study. Higher values of this indicator show an undesirable direction of change. The average change in the value of the indicator was 18.1% for the whole country, with a maximum increase of 89.6% and a maximum decrease of −28.9%. Only 114 municipalities (5.9% of the whole set) recorded a decrease in the number of immobile persons in relation to mobile ones, which mainly concerned areas located near the largest cities, as a result of the migration of young people to these areas (Figure 8).

The largest increase in the number of immobile persons was observed in north-eastern Poland and in the southern provinces (Śląskie and Opolskie), where in 2021, the indicator reached the highest values in the country. Significant growth dynamics also characterised south-eastern Poland, where in 2021, the ratio of the analysed age groups, despite dynamic growth, remained favourable.

4.3. Summary Index of the Intensity of the Digitalisation Challenge

The biggest challenges regarding digital infrastructure and digital competences in total, according to the static approach, are concentrated mainly in the areas of central, eastern, and northern Poland. The smallest problems in this respect are found in the southern and western parts of the country, as well as in some small areas of central and eastern Poland. The spatial distribution of the synthetic measure shows a high positive correlation (r = 0.930) with the indicator of access to fixed Internet of at least 30 Mb/s (cf. Figure 5 and Figure 9).

The synthetic measure for the dynamic approach, which corresponds to the analysed changes over time, shows a moderate negative correlation with the static measure (rs = −0.524) (cf. Figure 9 and Figure 10). The highest intensity of digital infrastructure problems was observed in Warmia and Mazury (north-eastern part of the country) and in a belt of municipalities stretching from the eastern part of the Opolskie province (central–south) to the south-eastern parts of Poland. Rural areas in central and eastern Poland are relatively in the best situation in this respect.

Municipalities facing intensive digitalisation challenges, both statically and dynamically, are mainly concentrated in the north-eastern part of Poland (Figure 11). The largest number of such municipalities is located in the Warmińsko-Mazurskie province, where as many as 42% belong to this group. Other provinces with a relatively high percentage of municipalities with an accumulation of digital challenges are Mazowieckie (18.3%), Podlaskie (18.1%), Opolskie (17.6%), and Kujawsko-Pomorskie (13.4%). In other regions, the share of these municipalities is minimal, and no municipality with such an accumulation of the studied challenge was found in Wielkopolskie.

5. Discussion

The concept of smart villages does not operate in isolation from the wider socio-economic context. In this respect, some aspects of it can be interpreted through the lens of the concept of smart rural development. The element that connects these two concepts is the so-called vicious circle of rural decline described by Weber and Fischer [6], which generates negative processes in these areas that are interconnected along the line of cause and effect, mainly under the influence of demographic changes. Breaking these negative trends is a difficult challenge, but theoretically achievable in five stages. The process implies a transformation: from a digitally excluded and socially passive village, through an active village with access to digital infrastructure and engaged residents, to a ‘player’ village with full technical, economic, and social capabilities to create innovation and exploit local resources and assets. Indeed, as the study by Valentín-Sívico et al. [66] suggests, having adequate Internet access at home improves the quality of life of residents. In turn, Townsend et al. [67] argue that broadband can make a significant contribution to addressing social and physical isolation in rural communities. These findings are in line with the central tenet of the smart villages concept.

The methodology used made it possible to assess the intensity of this challenge in the local dimension in two time snapshots—static and dynamic. The analysis shows that access to high-speed Internet in rural areas in Poland varies greatly. Although the projected investments until 2023 were to significantly improve this situation, regional differences will remain visible. This differentiation is reinforced by the polarisation of development. According to a study by Rosner and Stanny [68], the differential factor of the spatial arrangement of rural areas linked to the historical borders from before World War II and the entrenched division between the better developed west of the country, with a multifunctional economy, and the less developed east of the country, with a mono-functional economy, is still strong. This arrangement is the dividing line for many socio-economic processes and becomes apparent in aspects such as the process of population ageing, much more advanced in the east of the country than in the west of the country, and the process of the de-agrarianisation of the local economy with an exactly inversely proportional distribution. This implies that the low rate of digitisation is linked to the relationship that the less multifunctional the local economy (the more dependent on agriculture as the dominant economic sector), the greater the process of ageing (and depopulation) [20,69]. This, in turn, translates into digital competence and the ability to adopt new technologies. As research by Batorski [54], Woźniak-Jęchorek et al. [57], Lelkes [59], and Barbosa-Neves et al. [60] shows, older people are digitally excluded more than young people. So, both low levels of socioeconomic development and demographic old age overlap and accumulate the scale of the challenge.

Similar observations regarding the links between digitalisation and demographics are noted by researchers from other countries. The influence of Internet access on migration from rural areas is multifaceted, with evidence suggesting both positive and negative impacts. Enhanced Internet connectivity facilitates access to information about urban opportunities, thereby increasing migration intentions among rural youth. However, the effectiveness of this influence varies based on the type and quality of Internet infrastructure available. Internet access allows rural youth to gather information about job prospects and living conditions in urban areas, significantly increasing their migration intentions [70]. According to Goodman [71], young individuals, who are typically more adept at using technology, are particularly influenced by Internet availability, leading to higher rates of rural-to-urban migration. Ruiz-Martinez and Esparcia [72] add the Internet opens avenues for e-commerce and remote work, which can either encourage migration or provide alternatives to leaving rural areas. Previous study by Mahasuweerachai et al. [73] indicates that rural areas with robust broadband options experience different migration patterns compared to those with limited access. The COVID-19 pandemic has further underscored the importance of Internet access for economic resilience, suggesting that improved connectivity could mitigate out-migration by enhancing local opportunities [72]. Conversely, while Internet access can stimulate migration by providing information and opportunities, it may also exacerbate rural decline if individuals leave without contributing to local economies.

Digital competence measured by a proxy measure based on the relationship of the economic age groups of the population is also challenging. In many regions, especially in the east and north of the country, the proportion of older people relative to the working-age population is growing relatively rapidly, further hindering the acquisition of digital competence by this group. This distribution is linked to the historically established dividing line for many processes, described in the previous paragraph. A study by Komorowski and Stanny [19] provided similar conclusions regarding Internet accessibility. Also helpful in explaining spatial patterns are the analyses of Janc and Jurkowski [74] (p. 78), who found that, although the quality of the Internet in rural areas is generally lower than in urban areas, the spatial distributions of this phenomenon by municipality “do not directly relate to the core-periphery system”. This would imply that the development of the Internet network is taking place within some other, less polarised order.

It is worth mentioning, however, that despite the steady improvement of digital access in Poland (e.g., the installation of new fibre-optic subscriber network systems with an average speed of 231 Mbps was reported in 67% of all municipalities in Poland in 2022), the challenge is still the relatively high gap in Internet capacity between urban and rural areas [75,76]. This conclusion applies across the EU, as demonstrated by de Clercq et al. [77] in a study of the urban–rural digital divide in 1348 EU regions. A similar gap in the development of fibre-optic networks is noted in the United Kingdom by Gerli and Whalley [78], and it is indicated that only a combination of public intervention and private action can improve this situation. Some studies indicate that the digital economy is crucial for rural modernisation [79]. In general, the cited publications confirm the desirability of directing public funds for digitalisation to areas with low levels of development of this infrastructure. Thus, planned further infrastructure investments are a necessity in the current EU programming period—in the case of Poland, under the European Funds for Digital Development 2021–2027, as well as under the CAP funds for 2023–2027.

In summary, the smart villages concept represents an innovative approach to countering rural decline, integrating digital technologies as a key element in maintaining the vitality of rural areas. Digitalisation, understood as the development of broadband infrastructure and raising the digital skills of the population, is an essential condition for the implementation of smart village strategies. The lack of adequate digital infrastructure and low digital skills may exacerbate existing developmental disparities, leading to further exclusion of peripheral rural communities. At the same time, investment in digital technologies in rural areas can break the vicious circle of rural decline, stimulating the development of the local economy, improving access to public services, and strengthening social capital. Therefore, coherent policy actions targeting the development of digital infrastructure and support for technological competencies are key to transforming rural areas into smart, sustainable, and resilient communities.