Spatial Differentiation of EU Countries in Terms of Energy Security

Abstract

Global discussion on energy security remains deeply embedded in social, political, and economic discourse, especially in light of ongoing geopolitical instability and disruptions in supply chains. The aim of this study is to assess the degree of differentiation in the energy security of EU countries and to distinguish typological groups of the studied facilities based on the level of this phenomenon in 2023. This article uses a three-stage research procedure to assess the energy security of EU countries. In the first stage, statistical data were collected for 21 diagnostic features belonging to three groups: energy production and consumption, energy imports and exports, and economic and social factors. Next, using the TOPSIS method, three synthetic measures were constructed: separately for each group of features, taking into account features from the first and second groups, and taking into account features from all three groups. Based on these measures, typological groups of countries were identified using the three-median method. In the final stage, the impact of socio-economic characteristics on energy security was assessed. The results presented in this paper confirm the varied level of energy security in EU countries and indicate that it is linked not only to categories directly related to the energy economy but also to the level of socio-economic development of a given country. The top places in the ranking are occupied by countries such as Sweden, Finland, the Netherlands and Austria, while the last places in the ranking include Malta, Greece, Cyprus and Ireland.

1. Introduction

Security as one of the basic needs of man is an indispensable element of his life, activity and development. It is most often understood as a situation characterized by the lack of loss of something (e.g., health and material goods), i.e., as a state of no threat and peace. In scientific discourse, the ambiguity and complexity of the issue are indicated, which is the subject of interest of many disciplines, including security sciences, legal sciences, political and administrative sciences, economics and finance, sociology, ecology.

The issue of energy security is inherently complex, influenced by many different factors, which are subject to constant change. Therefore, it is difficult to indicate one clear definition of this concept [1]. Strojny [2] notes that the majority of researchers addressing energy security tend to highlight its various dimensions. The core topics of discussions surrounding energy have remained largely consistent since the crises of the 1970s—it is still predominantly focused on individual countries, with particular emphasis on oil and nuclear energy [3]. The sharp increase in oil prices caused by supply disruptions is still perceived as the main threat to energy security. Other factors disrupting security include armed conflicts, internal instability and nuclear proliferation. As an important tool for protecting the energy environment and promoting national economic transformation, it is advisable to study the relationship between environmental protection regulations and energy security to ensure energy security [4].

Renewable energy is a stable and safe alternative to fossil fuels that is not affected by price volatility or geopolitical risk [5]. However, it should be remembered that moving away from fossil fuels poses serious challenges in terms of the security of necessary livelihoods in mining-related communities [6]. Such changes require, above all, support from governments, which are obliged to take thoughtful actions and investments related to energy transformation.

According to some researchers, energy security issues should be considered not only in the context of affordable energy availability, but also from the perspective of political and social sciences [7]. The authorities of countries are obliged to guarantee access to energy for society. This is related to the quality of life of residents on the one hand, and to economic stability on the other. Access to affordable energy affects economic growth and improves people’s well-being.

The aim of this study is to assess the degree of differentiation of the energy security in EU countries and to distinguish typological groups of the studied countries due to the level of this phenomenon in 2023. The choice of the research period and its point-by-point nature results primarily from the desire to obtain the most up-to-date status of the analyzed phenomenon for the 27 EU countries and is inextricably linked to the availability of comparable statistical data. This article proposes an original methodological approach to assessing energy security based on 21 diagnostic features belonging to three groups: energy production and consumption, energy import and export, and economic and social factors. Specifically, this article addresses the following research questions:

• What characteristics are most frequently used in energy security research?
• In which countries is energy security at risk?
• How do socio-economic characteristics influence the level of energy security?

The theoretical and methodological added value of this article results from the development of the authors’ approach to assessing energy security using an aggregated indicator based on the TOPSIS method. The empirical added value is related to demonstrating the varying levels of energy security across European countries, which can be used to shape energy policy both in individual countries and across the EU.

Although energy security has been extensively covered in the literature (e.g., [1,2,3,4,5,6,7]), only a few authors analyze this phenomenon in the context of its level in EU countries (e.g., [8,9,10,11,12,13,14,15,16]). A significant number of publications focus solely on a selected country, region, or economic sector, thus losing the value of comparability. This article attempts to fill this gap by analyzing and comparing the security levels of all Member States. This article should also be considered a contribution to the discussion on choosing the direction of energy transformation in the context of building a sustainable economy.

This article consists of seven sections. Section 1 is an introduction, which explains the most important motivations of the authors to conduct research on energy security, presents the aim and indicates the method used in this study. Section 2, based on a review of literature, presents the concept of energy security and proposals for its measurement, as well as its assessment in European countries. The next section of this study contains the characteristics of the statistical material and the research method. Section 4 discusses the obtained results and Section 5 compares the results with research by other authors. Section 6 proposes policy implications, and the final section (Section 7) presents conclusions and directions for future research.

2. Literature Review

2.1. Definitions of Energy Security

Energy security is a multifaceted concept, interpreted in various ways, which complicates efforts to define it precisely. Researchers are split between those who view energy security concept through an economic lens and those who focus on its political and strategic dimensions. Nonetheless, the economic and political interpretations represent two facets of the same issue and both are necessary to explain the challenges as well as to propose solutions for dealing with the security of energy supplies in the EU [8].

Energy security as a guarantee of energy supply is usually defined by the resistance of the energy system to exceptional and unpredictable events that may threaten the physical integrity of energy flows or lead to an unstoppable increase in its prices regardless of economic fundamentals. It is therefore part of the national security system, because reliable and constant access to energy sources, at a cost that can be borne by society, is an essential element of any modern economy. More generally, energy security is a state of the economy that ensures that current and future demand for fuels and energy is met, in a technically and economically justified manner, with minimal negative impact of the energy sector on the environment and living conditions of society [17,18,19].

Energy security can be considered in the short or long term. In the short term, it is the intermittent availability of energy sources at an affordable price. In contrast, the long-term approach concerns the security of energy supply and other environmental issues related to energy demand and supply [20,21,22].

In the literature—in addition to specific definitions of this concept—the elements shaping energy security are also listed in order to analyze and evaluate this phenomenon on this basis. Franki and Višković [23] indicate that the basic dimensions of energy security are: energy cost, import dependence, and sustainable development. Elkhatat and Al-Muhtaseb [24] emphasize the key role of renewable energy in combating climate change and influencing energy security. Zhao et al. [4] extended the definition of energy security to include aspects such as resource availability and environmental availability and acceptability and Thanh et al. [20] added affordability to the above-mentioned aspects. The diversity of factors influencing energy security is also confirmed by the results of Fang et al. [25], who used economic, energy, environmental, and demographic data to assess sustainable energy security in China. Factors from similar areas (energy supply, environment, economy, society) were also identified by Kruyt et al. [26] and Augutis et al. [27], for example. Vivoda [28] adopted an even more comprehensive framework for evaluating energy security, incorporating a wide range of quantitative and qualitative indicators related to both traditional concerns and emerging factors—such as environmental, socio-cultural, and technological dimensions—within energy policy. Another dimension, largely absent from previous analyses, was the consideration of existence of an energy security policy. Yang et al. [29] proposed an index system consisting of five dimensions such as energy security drivers, energy security pressures, energy security status, energy security effects, and energy security responses. Kuzior et al. [15], when assessing energy security as the ability of the energy system to continuously, economically and environmentally safe meet the needs of consumers in EU countries, took into account: resource supply, resource availability, consumption, possibility of compensation, efficiency, safety and innovation.

2.2. Methods of Measuring Energy Security

The considerations presented above indicate that energy security is a very important, yet also very complex, issue. Therefore, its measurement is particularly important, as evidenced by numerous studies in which the authors proposed a wide range of energy security indicators to compare the performance of individual countries in the world, including European countries, or to track changes in a given country’s performance over time [9,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53].

When analyzing the results of research on energy security in detail, the diversity and differences in the indicators used by individual authors are clear. Martchamadol and Kumar [54] distinguish energy security indicators: disaggregated (simple) and aggregated (complex), where Kruyt et al. [26] point out the gap in research that would use aggregated indicators over long periods. In their opinion, a long-term assessment of energy security allows for taking into account a country’s development strategy, as well as the effects of unexpected events. The importance of composite indicators representing different dimensions of energy security is emphasized by Ang et al. [55]. According to them, in order to obtain an index that is a complex indicator of energy security, each of the simple indicators should be assigned a specific weight according to its importance and then an appropriate aggregation technique should be applied.

An example of the use of an aggregate energy security indicator is the study conducted by Alemzero et al. [56], who determined it based on thirteen variables (features) and used it to compare the energy security status of African countries. Their research results indicate the need to develop renewable energy sources and invest in energy infrastructure in Africa. These conclusions are also confirmed by the study by Ofosu-Peasah et al. [57], who assessed energy security based on an aggregate indicator constructed from twenty-four variables (features).

Another division of energy security indicators was made by Löschel et al. [43] who divided them into ex post and ex ante energy security indicators. Ex post indicators refer to the past, answering the question of whether the energy system has caused any serious disruptions in the economy. On the other hand, the ex ante indicator answer the question whether serious disruptions can be expected in the energy system in the future.

2.3. Assessment of the Energy Security of European Countries

Many authors have undertaken research to assess the energy security of European countries. Some of these studies focus on selected European countries, while others cover the entire EU.

Table 1. Selected scientific papers on the assessment of energy security in the EU, sorted by the number of countries covered by this study.

Countries	Authors and Research Objectives
UE	Checchi, Behrens, and Egenhofer [8]: Identification and assessment of existing and potential threats to EU energy supply based on a sectoral approach; Le Coq and Paltseva [9]: Measuring the exposure of EU Member States to energy supply risks; García-Verdugo, San-Martín [10]: Risk assessment of energy security; Rodríguez-Fernández, Carvajal, and Ruiz-Gómez [11]: Assessment of changes in gas supply security in EU Member States in the context of the 2006 and 2009 gas crises; Tutak et al. [58]: Assessment of energy sustainability using 13 characteristics of this sector from an energy, economic, and environmental perspective LaBelle [13]: Assessment of the energy relations of Germany, Poland, and the entire EU with Russia; Rosiek [14]: Identifying threats and challenges to ensuring energy security; Kuzior et al. [15]: Assessing the energy system’s ability to continuously, economically, and environmentally safe meet consumer needs; Osuma and Yusuf [16]: Determining the optimal renewable energy mix to enhance energy security and sustainable development
Austria, Bulgaria, Croatia, Czech Republic, Estonia, Lithuania, Latvia, Poland, Romania, Slovakia, Slovenia and Hungary (countries belonging to the Three Seas Initiative)	Tutak and Brodny [59]: Assessment of the energy security level using 17 features in the field of: energy, economics, environmental protection and social phenomena
Spain, Italy, France, Germany, the Netherlands, Poland and Sweden	Prina et al. [60]: Defining the optimal energy mix, limiting supply risk, necessary to achieve the 2030 and 2050 energy targets
Poland, Czech Republic, Slovakia and Hungary	Brodny and Tutak [61]: Assessment of sustainable energy security taking into account energy, climate, economic and social aspects characterized by 14 features
Estonia, Latvia and Lithuania	Augustis et al. [27]: assessment of achieving the level of energy security based on 67 features from the following areas: technical, economic, geopolitical and socio-political
Croatia	Franki and Višković [23]: Quantitative measurement and comparison of energy sector quality taking into account three different dimensions of energy security: costs, reliability and sustainability

Source: own elaboration.

provides a summary of selected studies, along with the authors and main research objectives.

The publications presented in Table 1 indicate that the conducted studies differ not only in terms of the countries and research period, but also in the set of features used to analyze and evaluate the discussed phenomenon. Furthermore, these studies employed various methods to achieve their research goal. For example, Tutak and Brodny [59] used Gray Relational Analysis (GRA) in their study, while Kuzior et al. [15] employed nonlinear (piecewise linear) normalization and multiplicative convolution.

In the publications mentioned above, the authors drew the following conclusions regarding energy security in the EU countries studied:

• Estonia and Latvia are characterized by a higher level of energy security than Lithuania, as they have large domestic energy resources (oil shale and hydropower), while Lithuania relies heavily on electricity imports [27];
• Due to the strong causal relationship between energy and pollutant emissions, the goal of the EU’s energy policy should be sustainable energy development, which will enable energy security to be achieved [58];
• The main recommendations for the EU in terms of ensuring energy security include introducing transitional programs (to eliminate energy poverty), developing prosumer and distributed energy, developing energy storage technologies, creating a common energy market, diversifying energy sources and suppliers, and modernizing transmission networks [59];
• EU energy security is threatened by the EU’s growing dependence on natural gas (primarily from Russia). When the study took into account geopolitics (economic, political, and social factors) and vulnerability (dependence on external energy, connections with energy exporters), it was shown that in terms of the level of gas supply security in EU countries, Denmark ranked best, while Lithuania ranked worst [11]. A threat to the energy security of EU countries related to natural gas imports was also pointed out by Hartvig et al. [62];
• The threats associated with the EU’s increasing dependence on energy imports from external suppliers (over 60% of EU economies depend on external energy supplies [59]) require the EU and individual Member States to increase diplomatic efforts and develop predictable, reliable, and transparent regulatory frameworks for energy policy [8].
• The exposure of EU Member States to energy supply risks varies depending on the type of energy [9].
• It would be more effective to promote energy security by attempting to reduce energy insecurity [10].

Although the research objectives, sets of characteristics, and methods for assessing energy security used in scientific studies vary, it is crucial that national or coalition governments use these results to develop appropriate policy standards on this important and strategic issue. This is supported by research by Sovacool [63], who notes that energy security should be focused on as a multidimensional concept, allowing for a shift from a narrow perception of energy security solely as the security of fuel supplies or appropriately priced energy services. Developing a systematic approach to measuring energy security can serve as a foundation for energy policy and strengthen institutional capacity, enabling the assessment of national energy security performance over time and facilitating targeted improvements where needed. As noted by Ang et al. [55], the construction of an energy security index is highly subjective, as the selection of features and their weights can be arbitrary. The selection of features stems from issues related to data availability and quality, as well as the use of information in the form of surveys or expert opinions. Narula and Reddy [64], however, believe that assessing energy security based on existing indicators has its limitations, as individual countries are beginning to look at this phenomenon more broadly, incorporating aspects related to environmental protection into their energy policies. It is worth mentioning that the literature already includes proposals for aggregate indicators that include an environmental dimension (e.g., [4,15,20,27]).

3. Research Methodology

The realization of the research objective led the authors to formulate the following research questions, presented in the Introduction:

• What characteristics are most frequently used in energy security research?
• In which countries is energy security at risk?
• How do socio-economic characteristics influence the level of energy security?

The answers to these questions were obtained by implementing the research procedure presented later in this chapter.

3.1. Stages of the Applied Research Procedure

In this article, a three-step research procedure was used to assess the energy security of EU countries, the subsequent stages of which are presented in Figure 1. In the first stage, statistical data on feature related to energy security were collected, then they were characterized and divided into three groups. In the next stage of this study, synthetic measures were constructed:

• Separately for each distinguished group of features,
• Taking into account features from the first and second groups, and
• Taking into account features from all three groups.

In the final stage, the impact of socio-economic features on the level of energy security was examined. For this purpose, the spatial distribution of EU countries was compared, obtained on the basis of a synthetic measure taking into account only the features from the first and second groups and a synthetic measure determined for features from all groups.

3.2. Characteristics of the Research Material

Despite the great importance of energy security for all countries in the world, there is currently no consensus on the exact interpretation of this concept and there is no universal methodology that could be used to calculate it [26]. Based on the various dimensions of energy security described in the previous part of this article and the features characterizing these dimensions (e.g., [15,26,27,59]), as well as due to the full availability of data for all EU countries, the authors decided to group the features they proposed into three groups: energy production and consumption, energy import and export, and economic and social factors (Table 1). The inclusion of a group of factors characterizing the level of socio-economic development of the country in the research results from the interrelationships between energy security and this development. As many authors point out, energy security affects the level of development of societies [65,66] and is an important element building the competitive advantage of states on the international market [67]. At the same time, the transition to more sustainable socio-economic development undoubtedly affects the level of energy consumption [68,69] and the security of the entire energy system [70].

The first group includes diagnostic features that characterize production capabilities and final energy consumption according to the sources of its acquisition. In determining their nature, it was assumed that the greater the energy production, the greater the energy security of the country. In the case of an increase in the consumption of energy obtained from traditional sources, energy security decreases, while an increase in the consumption of renewable energy increases this security. The authors are aware of the fact that an increase in energy consumption is an inherent feature of socio-economic development; however, this study assumed that energy obtained from renewable energy sources should be assessed positively in terms of energy security, because these sources are the basis for achieving energy independence. In turn, significant energy consumption from traditional sources results in a threat to energy security, especially when a given country has to import coal, gas or oil. The literature review shows that the diversification of energy sources has a significant positive impact on energy security; therefore, two features that characterize this phenomenon were included in this group. These are: the number of energy sources from which the country produces energy and the number of energy sources from which the country consumes energy. Both features were considered as stimulants.

The second group of features contains values that indicate the level of energy import and export. All features characterizing import were considered destimulants, because their high level indicates the country’s dependence on external energy supplies. On the other hand, features related to export were considered stimulants, because their high level indicates a surplus of energy produced in the country, and thus at a time of threat to energy security, this energy can be directed to meet domestic needs.

The third group is the most numerous, in which both features characterizing the level of economic development of the country and features characterizing energy prices and the situation of households in terms of meeting energy-related needs were proposed. In this group, three features were considered stimulants. These are features whose high values indicate a high level of economic development of the country, and the higher the level of development, the easier it is to ensure energy security, as there are funds for both energy transformation and potential energy import. The remaining seven features were considered destimulants and included energy prices and features characterizing the level of threat to energy security of households. Both the characteristics of this group and their nature are confirmed by the research of Demski, Poortinga, Whitmarsh et al. [71], who demonstrated, among other things, that greater concerns about energy security occur in countries that are performing worse economically and where social well-being is lower. Furthermore, their research indicates that affordability and the volume of energy imports also influence the sense of energy security. It is worth mentioning that the HICP characteristic included in this group takes into account the change in electricity, gas, and other fuel prices in 2023 compared to the previous year, thus allowing for an assessment of energy affordability. This characteristic was considered a destimulant, as the higher its value, the lower the affordability of energy for households.

In total, the study used twenty-one features characterizing the energy security of EU countries in 2023. Nine of them (43%) are stimulants, i.e., an increase in the value of this variable has a positive effect on the level of the phenomenon being studied (

Table 2. Groups of diagnostic features used to assess the energy security of EU countries.

Symbol	Name of Diagnostic Feature	Unit	Properties
Group 1. Energy production and consumption
G1_X1	Primary production of energy	ToE per 1000 people	Stimulant
G1_X2	Number of energy sources from which a country produces energy	number	Stimulant
G1_X3	Share of solid fossil fuels in final consumption of energy	%	Destimulant
G1_X4	Share of oil and petroleum products (excluding biofuel portion) in final consumption of energy	%	Destimulant
G1_X5	Share of renewables and biofuels in final consumption of energy	%	Stimulant
G1_X6	Number of energy sources from which a country consumes energy	number	Stimulant
Group 2. Energy imports and exports
G2_X1	Imports of Energy	ToE per 1000 people	Destimulant
G2_X2	Number of energy sources which were imported	number	Destimulant
G2_X3	Energy imports dependency	%	Destimulant
G2_X4	Exports of Energy	ToE per 1000 people	Stimulant
G2_X5	Number of energy sources which were exported	number	Stimulant
Group 3. Social-economic situation
G3_X1	Disposable income	Euro per capita	Stimulant
G3_X2	Gross domestic product at market price	Euro per capita	Stimulant
G3_X3	Electricity prices (excluding taxes and levies) for household consumers (consumption from 2500 kWh to 4999 kWh)	Euro	Destimulant
G3_X4	Electricity prices (excluding taxes and levies) for non-household consumers (consumption from 500 MWh to 1999 MWh)	Euro	Destimulant
G3_X5	Average annual rate of change in the Harmonized Index of Consumer Prices (HICP) for electricity, gas and other fuels	%	Destimulant
G3_X6	Gross domestic expenditure on R&D (GERD)	Euro per capita	Stimulant
G3_X7	Housing cost overburden rate	% of people	Destimulant
G3_X8	Expenditure on social protection related to covering the costs of housing	% of GPD	Destimulant
G3_X9	Inability to keep home adequately warm	% of household	Destimulant
G3_X10	Share of housing and energy costs in household expenses	%	Destimulant

Source: own elaboration based on the Eurostat database [72].

).

3.3. Research Method

A wide range of multivariate statistical methods can be used for linear ordering of objects, such as TOPSIS, Grey Relational Analysis (GRA), Principle Component Analysis (PCA), COPRAS, and Hellwig’s taxonomic development measure. All of them, at one stage, require comparability of the diagnostic feature matrix and are therefore sensitive to the standardization/normalization method used. Analyses conducted by other authors (e.g., [73,74,75]) show that the results obtained using these methods are comparable in most cases, and none of these methods demonstrates significant advantages over the others. Therefore, the choice of the TOPSIS method was dictated by its systematic and consistent approach to assessing and comparing multi-feature objects and its well-established position in the literature.

The TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) method [76] is based on the concept of synthetic measures. It is derived from the group of MDNA (Multi-criteria Decision Analysis) techniques and is based on the distances of an object, whose position in k-multidimensional space is described using a set of diagnostic features, from two artificial objects called the pattern and the anti-pattern. The pattern coordinates correspond to the ideal object, which in each analyzed dimension (diagnostic feature) achieves the most desirable value, and the anti-pattern coordinates correspond to the object, which for each diagnostic feature achieves the least desirable value. Its calculation procedure includes six steps (Figure 2):

Step 1—Assigning weights to individual diagnostic features in order to distinguish their impact on the analyzed phenomenon according to the following principle:

$${\displaystyle \sum }_{k=1}^j{w}_k=1$$

(1)

where

w_k—weight for the k-th diagnostic feature (k = 1, 2, …, j), and

j—number of diagnostic features.

To exclude the possibility that the different number of diagnostic features in each group influenced the value of the individual synthetic measures, it was decided to assign individual weights to each feature. Their values are presented in

Table 3. Unit weights by diagnostic feature groups.

Group	G1	G2	G3
Number of features	6	5	10
Unit weight	$\frac{1}{6}$	$\frac{1}{5}$	$\frac{1}{10}$

.

In the case of measures M_i_G1 + G2 and M_i_G1 + G2 + G3, in order to maintain the summability principle at 1, the unit weights were divided by 2 and 3, respectively, i.e., by the number of feature groups used to determine a given synthetic measure.

Step 2—Making the set of diagnostic features comparable. A wide range of techniques can be used for this purpose (see: [77]). In this work, the zero unitarization method was used [78,79].

$$m_{ik}=\left\{\begin{array}{c}{w}_k·{\displaystyle \frac{x_{ik}-\min x_k}{\max x_k-\min x_k}} \mathrm{for} \mathrm{stimulants}\\ w_k·{\displaystyle \frac{\underset{i}{\max } x_{ik}-x_k}{\max x_k-\min x_k}} \mathrm{for} \mathrm{destimulats}\end{array}\right.$$

(2)

Step 3—Determination of the coordinates of the pattern and anti- pattern. Due to the specificity of the zeroed unitarization method used in step 2, these coordinates are constant for all dimensions (diagnostic features) and are 1 for the pattern and 0 for the anti-pattern.

Step 4—Calculating the distance of the analyzed object from the pattern and anti-pattern [80,81] according to the formulas:

$$d_i^{+}=\sqrt{{{\displaystyle \sum }}_{k=1}^j{\left(m_{ik}-1\right)}^2} ,$$

(3)

$$d_i^{-}=\sqrt{{{\displaystyle \sum }}_{k=1}^j{\left(m_{ik}\right)}^2},$$

(4)

Step 5—Calculation of the synthetic measure value (M_i) for the i-th tested object according to the formula:

$${\mathrm{M}}_{\mathrm{i}}={\displaystyle \frac{d_i^{-}}{d_i^{-}+d_i^{+}}}.$$

(5)

In the additional sixth step, based on the values of synthetic measures for all objects, they can be assigned to homogeneous typological groups. In this work, the three-median method [82] was used, which allows for the separation of four typological groups of objects with values of the (M_i) measure from the following ranges:

$$\mathrm{group}1:{\mathrm{M}}_{\mathrm{i}}>{\mathrm{med}}_1\left(\mathrm{M}\right)$$

(6)

$$\mathrm{group}2:\mathrm{med}\left(\mathrm{M}\right)<{\mathrm{M}}_{\mathrm{i}}\le {\mathrm{med}}_1\left(\mathrm{M}\right)$$

(7)

$$\mathrm{group}3:{\mathrm{med}}_2\left(\mathrm{M}\right)<{\mathrm{M}}_{\mathrm{i}}\le {\mathrm{med}}_1\left(\mathrm{M}\right)$$

(8)

$$\mathrm{group}4:{\mathrm{M}}_{\mathrm{i}}\le {\mathrm{med}}_2\left(\mathrm{M}\right)$$

(9)

The TOPSIS method was used to classify multi-feature objects [83,84,85,86], including for energy research [58,87,88,89,90].

4. Research Results

The study of energy security among EU countries began with an analysis of their situation in terms of the level of features from individual groups. The starting point for this analysis were synthetic measures, the values of which are presented in

Table 4. Values of synthetic measures for individual groups of features (G1, G2, G3), ranks and typological groups for EU countries in 2023.

Country	Mi_G1	Rank	Group	Mi_G2	Rank	Group	Mi_G3	Rank	Group
Austria	0.668	3	1	0.491	18	3	0.551	6	1
Belgium	0.490	20	3	0.432	22	4	0.559	5	1
Bulgaria	0.541	12	2	0.529	6	1	0.476	17	3
Croatia	0.525	15	3	0.529	7	2	0.532	8	2
Cyprus	0.332	25	4	0.365	26	4	0.409	26	4
Czechia	0.577	8	2	0.513	12	2	0.426	24	4
Denmark	0.492	18	3	0.554	4	1	0.504	13	2
Estonia	0.524	16	3	0.580	2	1	0.519	10	2
Finland	0.787	2	1	0.531	5	1	0.530	9	2
France	0.622	4	1	0.512	13	2	0.423	25	4
Germany	0.557	10	2	0.498	15	3	0.444	21	4
Greece	0.398	23	4	0.509	14	3	0.393	27	4
Hungary	0.582	6	1	0.495	16	3	0.434	23	4
Ireland	0.365	24	4	0.385	25	4	0.506	12	2
Italy	0.537	13	2	0.477	21	4	0.471	20	3
Latvia	0.577	7	2	0.558	3	1	0.475	19	3
Lithuania	0.446	22	4	0.517	11	2	0.482	16	3
Luxembourg	0.332	26	4	0.286	27	4	0.686	1	1
Malta	0.310	27	4	0.386	24	4	0.593	4	1
Netherlands	0.489	21	4	0.520	10	2	0.594	3	1
Poland	0.491	19	3	0.491	17	3	0.485	15	3
Portugal	0.535	14	3	0.397	23	4	0.510	11	2
Romania	0.619	5	1	0.583	1	1	0.476	18	3
Slovakia	0.560	9	2	0.485	20	3	0.440	22	4
Slovenia	0.505	17	3	0.526	8	2	0.543	7	2
Spain	0.541	11	2	0.486	19	3	0.487	14	3
Sweden	0.801	1	1	0.524	9	2	0.627	2	1

Source: own elaboration.

. When assigning ranks to individual EU countries and separating typological groups, the fact that high values of the synthetic measure are desirable in the TOPSIS method was taken into account.

The information in Table 2 indicates that Sweden had the highest level of energy production and consumption (G1). This was due to the highest values of the features among the EU countries: G1_X₅ (Share of renewables and biofuels in final consumption of energy) and G1_X₆ (Number of energy sources from which a country consumes energy), as well as the lowest value of the feature G1_X₄ (Share of oil and petroleum products in final consumption of energy). Finland was also in the same group, with the highest level among the Member States concerning primary energy production per 1000 inhabitants (G1_X₁), as well as a high level of energy consumption from renewable sources (G1_X₅) and low values of the features of energy consumption from solid fossil fuels (G1_X₃) and from crude oil and petroleum products (G1_X₄). In addition, the first group was joined by: Austria, where, similarly to Sweden, as many as 13 types of sources were identified from which the country consumes energy (G1_X₆), as well as France with the highest number of sources from which the country produces energy (G1_X₂) among the Member States. The next countries in this group are: Romania and Hungary, distinguished by high levels of features: G1_X₂ (the number of sources from which the country produces energy) and G1_X₆ (the number of sources from which the country consumes energy).

The lowest levels of the synthetic measure in the group of energy production and consumption were recorded in seven countries, among which Malta was the weakest, which was mainly due to the lowest values of features: the G1_X₁ (Primary production of energy) and G1_X₂ (Number of sources from which a given country produces energy) among the EU countries. In addition, the fourth typological group (the worst) also included Luxembourg, Cyprus, Ireland, Greece, Lithuania and the Netherlands.

The situation of EU countries is different in terms of energy imports and exports (G2). In this case, the ranking is headed by: Romania, Estonia and Latvia. The first of these countries stands out positively among EU countries due to the lowest value of the feature concerning energy imports per 1000 inhabitants (G2_X₁). The second of them stands out positively due to the feature G2_X₃ (Energy imports dependency). In turn, Latvia has low levels of features: G2_X₁ (Imports of energy) and G2_X₂ (Number of energy sources which were imported). In addition, the first typological group (the best) also includes Denmark, Finland and Bulgaria, countries with positive values of most of the features adopted for this study in the group of features characterizing energy imports and exports.

In the last, weakest typological group, similarly to the first group of features, again included Ireland, Cyprus, Luxembourg and Malta, joined by: Portugal, Belgium, Italy. These three countries were assessed better in terms of the first group of features (energy production and consumption), occupying, respectively: 14th position (typological group II), 20th position (typological group III) and 13th position (typological group II).

The last group of features used in this study characterizes the socio-economic situation of EU countries (G3). In this case, the highest levels of the synthetic measure were recorded by six countries: Luxembourg, Sweden, the Netherlands, Malta, Belgium and Austria. Luxembourg, ranked first, is a country characterized by the highest level of disposable income per capita (G3_X₁) and GDP per capita (G3_X₂) among EU countries, low expenditure on social protection related to covering housing costs (G3_X₈), and the lowest percent of households reporting the inability to maintain adequate heat in the home (G3_X₉) among the Member States. Sweden, ranked next, has the lowest electricity prices for non-household consumers in the EU (G3_X₄) and the highest gross domestic expenditure on research and development (G3_X₆). The remaining countries in this typological group are characterized by positive values of features from this group, i.e., the values of the features—stimulants are above the EU average, and the values of the features—destimulants are below the EU average. Malta is worth noting with the lowest share of housing and energy costs in household expenditures among the countries studied (G3_X₁₀), as well as Belgium with the lowest feature G3_X₅ (average annual rate of change in the Harmonized Index of Consumer Prices (HICP) for electricity, gas and other fuels).

The least favorable values of the synthetic measure in this group of features concern Greece, Cyprus and France. These are countries, except for France, with low disposable income per capita (G3_X₁) and GDP per capita (G3_X₂), with Greece having the highest housing cost overburden rate among EU countries (G3_X₇), Cyprus a high electricity price index for households (G3_X₃) and for non-household consumers (G3_X₄), and France a high share of housing and energy costs in household expenditure (G3_X₁₀).

It should be noted that the leaders in each of the above rankings are: Sweden, Finland and Austria (except G2). The worst situation was observed for: Malta and Luxembourg (except G3), Cyprus, Greece (except G2) and Ireland (except G3).

In the next stage of research on energy security in EU countries, synthetic measures were determined by combining features from the first and second groups (G1 + G2), and then combining features from three groups (G1 + G2 + G3). The values of these synthetic measures are presented in

Table 5. Values of synthetic measures for groups of features G1 + G2 and G1 + G2 + G3, ranks and typological groups in EU countries in 2023.

Country	Mi_G1 + G2	Rank	Group	Mi_G1 + G2 + G3	Rank	Group	Difference in Position *	Difference in Group **
Austria	0.581	4	1	0.566	3	1	−1	0
Belgium	0.466	22	4	0.513	10	2	−12	−2
Bulgaria	0.535	10	2	0.502	14	3	4	1
Croatia	0.527	12	2	0.530	7	2	−5	0
Cyprus	0.348	25	4	0.377	27	4	2	0
Czechia	0.543	8	2	0.485	20	3	12	1
Denmark	0.519	14	3	0.511	11	2	−3	−1
Estonia	0.548	7	2	0.535	6	1	−1	−1
Finland	0.664	1	1	0.593	2	1	1	0
France	0.572	5	1	0.507	12	2	7	1
Germany	0.529	11	2	0.491	17	3	6	1
Greece	0.445	23	4	0.420	26	4	3	0
Hungary	0.538	9	2	0.487	19	3	10	1
Ireland	0.374	24	4	0.438	25	4	1	0
Italy	0.511	17	3	0.493	15	3	−2	0
Latvia	0.569	6	1	0.521	9	2	3	1
Lithuania	0.479	20	3	0.480	22	4	2	1
Luxembourg	0.313	27	4	0.482	21	4	−6	0
Malta	0.344	26	4	0.455	24	4	−2	0
Netherlands	0.504	18	3	0.540	4	1	−14	−2
Poland	0.491	19	3	0.488	18	3	−1	0
Portugal	0.471	21	4	0.491	16	3	−5	−1
Romania	0.601	3	1	0.537	5	1	2	0
Slovakia	0.523	13	2	0.480	23	4	10	2
Slovenia	0.515	16	3	0.529	8	2	−8	−1
Spain	0.518	15	3	0.504	13	2	−2	−1
Sweden	0.650	2	1	0.640	1	1	−1	0

* Negative value indicates improvement in position. ** Negative value means moving to a better group. Source: own elaboration.

.

The leaders of the ranking, which takes into account features related to energy production and consumption (G1) and energy import and export (G2), are the following countries: Finland, Sweden, Romania, Austria, France and Latvia. In turn, the ranking created after adding features reflecting the socio-economic situation (G3) shows as leaders: Sweden, Finland, Austria, the Netherlands, Romania and Estonia. Countries arranged in non-decreasing order according to the value of the M_i_G1 + G2 + G3 measure were presented in Figure 3.

The last two columns of Table 5 show the differences in rankings and affiliations to typological groups calculated by taking into account the features belonging to the first and second groups and the three groups together. As can be seen from Table 5, adding the third group related to the socio-economic situation to the analysis caused significant changes in the case of some Member States. A significant improvement in position was noted by countries such as Netherlands (improvement by 14 positions), Belgium (improvement by 12 positions) and Slovakia (improvement by 8 positions). On the other hand, a drop in the ranking concerned, among others, countries such as Czechia (drop by 12 positions), Hungary and Slovakia (drop by 10 positions).

Typological groups of EU countries created on the basis of the values of the M_i_G1 + G2 measure are presented in Figure 4, and those created on the basis of the M_i_G1 + G2 + G3 measure in Figure 5.

As can be seen from Figure 4, the highest energy security (first typological group), when only energy-related features (G1 and G2) were taken into account, was characteristic of the Scandinavian countries, Austria, France, Romania and Latvia. The second typological group is dominated by the countries of Central Europe. The worst situation (fourth typological group) concerns countries located in the northern part (Benelux countries, Denmark and Ireland) and southern Europe.

Adding the socio-economic features to the measure changed the spatial layout of the analyzed countries (Figure 5). It is difficult to detect a clear pattern, although similarly to the M_i_G1 + G2 measure, southern Europe (except Spain) is marked with a shade of red. Once again, however, the top places in the classification are, among others, Northern European countries, Estonia and Romania, and at the lowest: Ireland, Greece, Malta and Cyprus.

5. Discussion

Energy security is a current and very important topic for every country in the world. Research in this area has been, is and will be conducted, because without energy neither households nor businesses can operate. Some of them concern selected energy aspects, e.g., analysis of the supply of one energy product, and some are of a general nature, taking into account all factors influencing the problem under study. Our research is an attempt to indicate a method of assessing the level of energy security of countries, taking into account both features directly related to energy and socio-economic features.

As in any empirical research, current and comparable data are of crucial importance. Thanks to the experience of other researchers (e.g., [4,20,23,24,26,27]), we decided to group the diagnostic features selected for the construction of the synthetic measure into three groups, which are discussed in detail in the “Research methodology” section of this study. The studies of the above-mentioned authors had different scopes, so in some cases it is difficult to clearly state the similarity of the obtained results.

The results of our research showed that the highest level of energy security was observed in: Sweden, Finland, Austria, the Netherlands, Romania and Estonia. Three of these countries (Austria, Romania and Estonia) were also at the top of the ranking in the study by Brodny and Tutak [61]. It should be emphasized that the subjects of the research by these authors were not all EU countries, only those that belong to the Three Seas Initiative. Estonia’s high position is also confirmed by the results of Augutis et al. [27], indicating the high level of energy security of this country in the form of its own resources. The ranking of countries in terms of the assessment of sustainable energy development [58]—established similarly to our research using the TOPSIS method—starts with France and Sweden. In the classification obtained in this publication, France was only in 12th position, and the difference in the results may be caused by the number of indicators included in the calculations. In our research, we used 21 diagnostic features, and Tutak et al. [58]—13. In Sweden and Finland, more than half of energy production comes from renewable sources, which largely secures their needs. The results of Kuzior et al. [15] are optimistic for the studied group of countries, indicating that most EU countries have a sufficient or moderate level of energy security.

The countries with the highest level of energy security problems according to our research results were: Cyprus, Greece and Ireland. Zachariadis et al. [91] attempted to determine the causes of these problems in Cyprus by assessing the effects of the national energy and climate plan for this country. Compliance with energy and environmental obligations by Cyprus may bring benefits, but it requires a strong will from the authorities of this country in implementing individual provisions. High energy costs and dependence on natural gas and oil supplies from Russia have influenced the energy crisis in Greece [92]. According to Glynn et al. [93], the main threats affecting energy security in Ireland are: dependence on imports, low fuel diversification and volatile energy prices, and these are the reasons why in our research, in which such features were taken into account, this country is ranked low. Malta, in turn, although an ideal location for photovoltaics, the vast majority of its electricity comes from imported fossil fuels [94]. This situation is not favorable for this country, therefore the solution would be to switch to low-emission renewable energy (sun, wind and biogas), which would provide energy independence [95]. However, the problem of implementing renewable energy technologies in this country is the lack of space and high landscape value [96].

The results of the EU countries ranking obtained using the energy security measure constructed by the authors were compared with the classification of EU countries obtained on the basis of three independent measures (Energy Trilemma Index, Energy Sovereignty Index, Energy Security Risk), which are used to assess various aspects related to the energy economy.

The Energy Trilemma Index (ETI) [97] is published annually by the World Energy Council (WEC) and is one of the most extensive measures, as it takes into account three areas that can be characterized as follows [98]:

• Energy security—reflects a country’s ability to meet current and future energy needs, recover quickly from disruptions, and maintain a steady supply of energy;
• Energy equity—indicates the availability and affordability of energy for all citizens;
• Environmental sustainability—refers to the efficiency and productivity of energy generation, industry, and distribution, as well as efforts to decarbonize and improve air quality.

Comparing the results of the classification of EU countries based on the M_i_G1 + G2 + G3 measure (Table 5) and the ETI for 2023, the top three in both cases are Sweden and Finland. According to the ETI, the leader of the ranking is Denmark, which took only 11th place in our study. The last place in both classifications went to Cyprus. The consistency of the classification was assessed by calculating the Pearson linear correlation coefficient between the values of these measures, which was 0.722. This result means a quite strong positive correlation, so there is quite a large similarity between the country rankings obtained using the ETI and M_i_G1 + G2 + G3. The observed discrepancies in the rankings are a consequence of a different methodological approach and a set of features adopted by WEC when calculating the ETI [98]. Although the measure constructed by the authors of this study did not focus on the analysis of the impact of energy security on sustainable development or the state of the natural environment, the adoption of features describing the structure of energy production by type of carrier (renewable and non-renewable) resulted in a moderate positive correlation between the environmental sustainability score (ETI component) and M_i_G1 (M_i_G1 + G2 + G3 component)—the Pearson linear correlation coefficient between these measures was 0.502.

In the case of the Energy Sovereignty Index (ESI) [99], published by the European Council on Foreign Relations, similarly to the ETI, Denmark was at the top of the ranking, while Malta was ranked lowest. The Pearson linear correlation coefficient between the ESI values and the M_i_G1 + G2 + G3 values was 0.525, which indicates a moderate positive relationship, so the differences in these rankings are greater than in the previous comparison (ETI with M_i_G1 + G2 + G3). The reasons for this state of affairs can be sought both in a different set of diagnostic features and in the timeliness of the data (the ESI measure covered 2023, but a lot of data came from earlier years).

In the Energy Security Risk ranking [100] published by the Center for the Study of Democracy for 2021 and for five countries (Germany, Italy, Bulgaria, Greece, Romania), we can also find similarities to the results obtained on the basis of the measure proposed by the authors in this paper. Among the five countries assessed, the lowest level of risk was recorded for Romania (5th place in the ranking based on M_i_G1 + G2 + G3), which was the result of, among others, a significant decrease in energy consumption due to the transition to a more sustainable economy. At the same time, there was a rapid increase in the production of energy from renewable sources in this country, and the reduction in the prices of crude oil and natural gas had a positive impact on the level of energy poverty. It is also worth noting that Romania is moving away from dependence on supplies of the above-mentioned carriers from Russia. In turn, Greece (26th place in the ranking based on M_i_G1 + G2 + G3), due to the large share of emission sources in electricity production and increasing energy consumption per capita, was identified as the most threatened in terms of energy security among the five countries analyzed.

6. Political Implications

The research conducted in this article, together with the information obtained from the literature review, allows us to suggest the following policy implications to policy makers, both in individual EU countries and in the entire community:

• Reducing dependence on imported energy by increasing the share of renewable energy sources. Due to the high variability of energy production from renewable sources, in order to improve the resilience of the energy system and increase the security of energy supply for all EU countries, it would be necessary to invest in interconnections with neighboring networks [101]. Renewable energy sources as an instrument for increasing domestic energy production improves energy security, but one should caution against policies that limit the use of coal, resulting in a decrease in total energy production and an increase in the share of natural gas. The justification for the above thesis is the frequent dependence on gas imports [102].
• Development of energy storage system, as the underdeveloped system blocks the efficient use of energy from renewable sources [103].
• Implement advanced air quality monitoring and control systems, and create air quality maps. The use of fossil fuels in industry, transport, and households is associated with the emission of air pollutants and greenhouse gases. Therefore, it is natural to protect the environment and transition to a low-emission economy, which in turn leads to a change in the energy mix, with conventional sources being replaced by renewables. However, this change will not happen overnight, and in many countries, the use of conventional energy sources, which pollute the air, still dominates and will continue to dominate. Therefore, air quality measurements should be continued to identify areas with high levels of pollution [104]. Investing in effective measurement equipment will enable the creation of air quality maps [105], and technological advances in this area can help people protect their lives.
• Establishing air quality standards and strictly enforcing them, as well as supporting businesses and residents in using green technology. However, it should be borne in mind that not all pro-ecological activities declared by individual economic organizations have pure intentions; therefore, it is important to remember to systematically counteract the “greenwashing” phenomenon [106].
• Financing innovative ways to reduce air pollution, through the use of advanced technologies, e.g., the effective method proposed by Zheng et al. [107] for removing gaseous elemental mercury from coal-fired exhaust gases.
• Investing in clean energy through the issuance of green bonds, which aim to finance projects that contribute to environmental protection and fight with climate change. Policy makers are advised to integrate carbon pricing with green energy investments [108]. The importance of carbon bond issuance and its enforcement is crucial due to the inflow of foreign investment. As Walter and Ugelow [109] assume, some processes causing intense pollution are transferred via foreign direct investment to host economies with less restrictive environmental regulations, contributing to a deterioration of environmental quality. Other researchers, such as Ahmad et al. [110], argue that foreign direct investment protects environmental quality in host economies.

A common approach to EU energy security would contribute to solving the problems of Member States in this regard. However, they oppose the creation of such a policy because short-term national interests prevail—perceived as a guarantee of this security [21]. Although an integrated solution would give EU countries a stronger negotiating position than each country can achieve individually (both politically and economically), solidarity is noticeable only in selected aspects. As Grimmel [111] emphasizes, the crises faced by the EU may, paradoxically, be an opportunity to solve specific problems, including energy problems, in solidarity.

7. Conclusions

Energy security has become a central focus of economic policy across all nations, and addressing the threats it faces involves a complex process that demands both immediate measures and a long-term strategic approach. In the case of Europe, the main directions of action in this area should be primarily: diversification of energy sources and suppliers, development of renewable energy sources, strengthening of energy infrastructure, and increasing the common energy policy at the EU level. Many studies emphasize the role of increasing the share of renewable energy sources and biofuels in the energy mix for achieving energy security, energy independence and achieving climate goals, while reducing dependence on imported fossil fuels. Discussions and establishing a common EU energy policy are necessary, but considering the differences in the level of energy security in individual countries, the geopolitical situation and preferences of Member States, this is not an easy task. Keeping their finger on the pulse and monitoring changing internal and external factors should be a priority for the authorities of the EU countries.

This article concerns the analysis of energy security of EU Member States, and its assessment was carried out on the basis of the authors’ synthetic measure obtained by the TOPSIS method using 21 diagnostic features characterizing energy production and consumption, energy import and export and the socio-economic situation.

Based on a literature review, the most frequently used characteristics in energy security research were identified. Based on this, and taking into account the availability of statistical data at the EU country level, a set of features was selected for this study, grouped into three groups. This answers the first research question stated in the introduction.

To answer the second research question, synthetic measures were constructed, which identified countries with a high level of security, as well as those at risk. It turned out that the top positions in the ranking (the first typological group—high security) were occupied by countries such as Sweden, Finland, the Netherlands, and Austria, while the fourth typological group (the lowest level of security) included Malta, Greece, Cyprus, and Ireland.

The results presented in this paper confirm (the third research question) that energy security is linked not only to categories directly related to the energy economy (production, consumption, energy imports, and exports), but also to the level of socio-economic development of a given country. The addition of socio-economic characteristics to the metric changed the ranking of some countries. This is particularly evident in the case of the Netherlands, Belgium, the Czech Republic, Hungary, Slovakia, and Slovenia. However, once again, the highest positions in the ranking are occupied by countries from Northern Europe, Estonia, and Romania.

Proposing a measure that allows for the comparison of countries’ energy security levels, which takes into account not only energy-related diagnostic features but also characteristics of socio-economic development, constitutes the added value of this article. The results obtained can be used in the development of energy policies both at the national level and within the European Union.

It is also necessary to point out certain limitations related to the conducted analysis:

• Firstly, the analysis was based primarily on aggregated values of characteristics related to the levels of energy production and consumption and its import and export. This prevented their in-depth analysis in the context of individual energy carriers.
• Secondly, due to the availability of data and the need to ensure their comparability, only electricity prices in individual EU Member States were taken into account, without taking into account information on the prices of other energy carriers.
• Thirdly, the analysis was limited to only one year, which makes it impossible to answer the question about the rate and direction of changes in the analyzed phenomenon over time. At the same time, the time scope of the analysis covered the period of significant turmoil related to energy supplies, resulting from the war in Ukraine.

According to the authors, an attempt at a dynamic approach to the analyzed phenomenon would be an interesting direction. In this context, an attempt to assess the impact of the level of energy security on the implementation of sustainable development goals, or the level of energy production from renewable sources, will be possible subjects of our future research.