The Implementation of the European Green Deal Strategy as a Challenge for Energy Management in the Face of the COVID-19 Pandemic

Abstract

This article aims to present changes in the use of electricity by service companies, resulting from regulations within the framework of increasing energy efficiency from the perspective of the implementation of the European Green Deal strategy. To achieve the above goal, the following research question was formulated: to what extent did the COVID-19 pandemic affect the implementation of energy transformation and electricity consumption among the surveyed group of recipients? It should be noted that, so far in the global environment, more and more electricity has been used every year, and this tendency is still continuous and growing. Therefore, in European Union countries, measures have been taken to balance demand and its rational use, resulting from the implementation of the European Green Deal strategy. According to the strategic goal of the indicated policy, EU countries are obliged to implement a sequence of actions enabling their transformation into a modern, resource-efficient, and competitive economy. In particular, the strategy aims to achieve three main goals: 1. Achieving climate neutrality by reducing net greenhouse gas emissions to zero in 2050; 2. Decoupling economic growth from the consumption of natural resources; and 3. Striving for an equal standard of living in all countries. Therefore, the behavior of individual countries should focus on providing the required amount of energy to ensure socioeconomic growth while reducing pollution and environmental devastation caused by traditional methods of energy production and use. There are numerous talks and debates about the defined tasks and mechanisms leading to the achievement of the indicated goals, in which hardly any mention is made of the methods of monitoring the progress and evaluation of individual projects at the stage of building a new green deal. This study aims to fill the research gap observed in the current state of knowledge on energy management in EU countries from the perspective of the European Green Deal strategy and changes in its management resulting from the conditions of the COVID-19 pandemic. Therefore, in line with the aim of the article: 1. Factors influencing electricity consumption in service enterprises operating in EU countries were indicated; 2. Energy consumption variability in these entities was determined; and 3. The correlation between electricity consumption and GDP growth in the service sector located in EU countries was indicated. Therefore, a hybrid research method was used to analyze the data obtained from the databases of Eurostat and Statistics Poland, which consisted of the following analyses: diagnostic-descriptive, main components, and wavelet transform. Based on the conducted research, it should be concluded that energy consumption among service companies operating in the EU market is conditioned by three sources of causes along with the relevant consumption factors. The first group includes energy and technical premises, along with technological determinants. The second is shaped by financial and economic motives, which include socioeconomic factors. The third group is related to environmental sources represented by the natural environment and geographic and meteorological factors. The increase in electricity consumption in service enterprises is related to the average GDP growth of EU countries following a two-way cause-and-effect relationship implemented until 2019. Nevertheless, since 2020, there has been a noticeable decrease in energy consumption by 14.01% by service entities, which results from the limitations caused by the COVID-19 pandemic and the implementation of the European Green Deal strategy. Meanwhile, the structure of electricity consumption growth was dominated by industrial enterprises (increased by 37.7%) and individual consumers (increased by 7.8%). The results of the research may motivate the managers of EU countries and enterprises to analyze the factors of energy consumption, its variability, and dependence on economic growth, which contribute to determining the forecast of future energy demand, in connection with the ongoing energy transformation resulting from the implementation of the European Green Deal strategy, and economic slowdown caused by the COVID-19 pandemic. The issues presented in this article are an attempt to fill the gap indicating practical experience related to the process of electricity management and management in the political, economic, and technological dimensions from the perspective of implementing the European Green Deal strategy and the conditions resulting from the COVID-19 pandemic.

1. Introduction

Service companies operating in the market today are largely subordinated to the environment and especially to determinants originating from the state of the economy in which they operate. A typical parameter is cyclicality, with the following processes of economic growth and decline, where crisis is an integral component. Various types of crises (economic, social, natural, and the present, so-called pandemic), which permanently take control of the economies of countries, have emphasized the need to implement real and rapid structural changes in the economy. To overcome the threats resulting from, inter alia, climate change, the European Green Deal (EGD) was introduced, i.e., a growth strategy with the role of transforming the economic system of the European Union (EU) into a competitive and sustainable economy.

The European Green Deal strategy concerns, inter alia, reducing the consumption of energy generated from non-renewable (primary) sources and supplied to individual consumers and service companies from the electricity grid (final) in terms of neutralizing climate change. This study aims to present the impact of various economic, social, technological, and environmental factors on the energy efficiency of institutions located in the 28 countries (pre-Brexit) of the European Union (EU28). In addition, the study intends to analyze the above-mentioned factors together with an indication of the ones that should be of strategic importance in the process of reducing energy consumption by service companies and increasing the energy efficiency of EU28 countries.

On the other hand, the main goal of the article is to present the changes in the use of electricity by institutional consumers (service companies) resulting from regulations within the framework of increasing energy efficiency from the perspective of implementing the European Green Deal (EGD) strategy.

Based on the specific goal and question, a hypothesis was formulated that EU countries support the process of energy transformation through policies promoting energy efficiency in the political, economic, and technological dimensions. This was destabilized under the influence of the pandemic crisis, with it being perceived as a strategic opportunity for economic development and the possibility of changes in the structure of energy consumption management among institutional recipients. In connection with the above thesis, the statistical data obtained from the databases of Eurostat and Statistics Poland were analyzed. The results were interpreted using the diagnostic-descriptive method with nonlinear linearized regression and the multivariate method with the use of principal components analysis (PCA) and wavelet transform (WT), which were verified with the use of Statistica and MATLAB & Simulink software.

The research procedure carried out for this article enabled verification of the hypothesis and implementation of the assumed research goal, which was set out in the introduction to the article. This study attempts to present the changes in the use of electricity by service companies from EU countries, which result from the application of the European Green Deal strategy and constitute a challenge for energy management.

In the process of making decisions on increasing energy efficiency, many variables are taken into account that should not be underestimated. Although the essence of some factors is subject to reduction, special attention should be paid to the technological, socioeconomic, and environmental determinants influencing energy consumption.

Changes taking place in the environment of enterprises resulting, inter alia, from the global COVID-19 crisis, and the fast pace of market and enterprise development affect the dependence of the level of energy use on the size of GDP achieved in the analyzed EU countries. Thus, the demand for energy and the level of consumption in service entities largely depend on the strength and pace of the economic development of a given country. The main benefit of the research carried out for this study is that it provides information to energy managers that can be used to achieve energy efficiency and the rational implementation of the European Green Deal strategy.

The issues analyzed in the article have not been the subject of research for the service sector so far. For this reason, it intends to supplement the information in the research carried out so far on energy use and increasing energy efficiency resulting from the implementation of the EGD strategy from the perspective of the COVID-19 pandemic.

1.1. The European Green Deal

The current model of a linear economy connected with shaping economic growth was associated with the use of non-renewable natural resources. Achieving climate neutrality, which is the main goal of the European Green Deal strategy, imposes a gradual resignation from this type of model, with the assumption of its replacement by the circular economy paradigm. To achieve the goal of climate neutrality, universal access to cheap energy and reliable recycling systems is necessary, given the availability of critical minerals from the perspective of energy efficiency.

The EGD strategy, introduced by the European Commission on 11 December 2019, is considered one of the most important in the field of EU climate policy. It covers the overall economic strategy and many areas of EU policy [1]. Among others, it covers energy, industry, construction, agriculture, services, transport, biodiversity, and the elimination of soil, air, and water pollution. The main goal of the strategy is to transform the EU into a fair and prosperous society, living in a modern, resource-efficient, and competitive economy with net zero greenhouse gas emissions in 2050 and where economic growth is decoupled from the use of natural resources [2]. Thus, the idea of EGD is also expressed in the protection, maintenance, and improvement of natural capital, and ensuring health security by counteracting the threats and negative effects of climate change.

According to the adopted plan, the transformation of the EU economy toward a sustainable future will be implemented through seven strategic goals: 1. Development of ambitious climate goals for 2030 and 2050. 2. Clean, safe, and affordable energy. 3. Competitive industry and a circular economy are recognized as key factors in reducing greenhouse gas emissions. 4. Energy-saving, emission-free, and resource-conserving construction. 5. Zero pollutant emissions for a non-toxic environment. 6. A fair, healthy, and environmentally friendly food system. 7. Sustainable and intelligent mobility [3,4,5].

The implementation of the EGD goals related to achieving climate neutrality by 2050 and tasks resulting from the protection of the natural environment obliges EU countries to change their methods of energy production and consumption [6]. Nevertheless, the implementation of the EGD provisions was interrupted as a result of the spread of the COVID-19 pandemic. The pandemic crisis contributed to the correction of the targets, which currently prioritized the acceleration of the EU energy transformation in connection with the forecast recession, in line with the presented “Fit for 55” package [7,8].

1.2. Factors Influencing Energy Consumption

In recent years, the interests of entrepreneurs and scientists in the field of energy consumption have focused especially on issues related to energy management from the perspective of improving energy efficiency. The process of rational use of energy affects the entrepreneurship of entities operating in a given environment due to its fundamental impact on the energy, economic, social, and environmental security of individual EU countries. Recently, the reduction of energy consumption and the related improvement in energy efficiency have been recognized as the priorities of the green economy concerning European and global requirements aimed at sustainable development. Nevertheless, there are ongoing discussions on the identification of factors contributing to changes in energy consumption (energy saving), which are also a premise for recognizing the composition of instruments in the approach to measuring their effectiveness. Identifying the relevant factors influencing energy consumption is of strategic importance for regions and sectors.

The indication of factors, including those influencing the level of energy consumption, is a complicated process that requires thorough analysis and knowledge of the essence of the issues under consideration [9]. On the basis of the research conducted so far among individual consumers, the following factors that affect energy consumption should be identified: economic and financial situation [10,11,12,13,14], socio-demographic conditions [12,15,16,17,18,19], physical characteristics of the dwelling [20,21,22,23,24], location of the apartment [10,19,25,26,27], environmental and climatic conditions [20,28,29,30,31], and charges for energy use [25,32,33,34]. Meanwhile, in the case of countries and enterprises, the diagnosis may simultaneously cover political [35], technological [36], socioeconomic [37], meteorological, and climatic [38,39] factors, which are presented in Figure 1 [40].

Another economic condition that has recently influenced energy consumption is the SARS-CoV-2 coronavirus, which caused the global COVID-19 pandemic crisis. Since the end of 2019, countries around the world have also been struggling with the health crisis caused by the above-mentioned virus, which has spread rapidly to every country [41].

In Europe, the first cases of infection with the SARS-CoV-2 virus were recorded on 24 January 2020 in France. In turn, from March of the same year, numerous restrictions were introduced (including lockdown, closing the operations of enterprises from selected sectors of the economy, remote work, social distancing, and restrictions in movement), which slowed down the further spread of the COVID-19 coronavirus [42]. More than half of the global community (over 4 billion people) has been subject to the above-mentioned restrictions since the beginning of the pandemic. A situation of this kind caused a sharp decline not only in economic output, but also in the economic growth rate of most countries. In addition, restrictions imposed by governments on the functioning of economic activities also reflect the level of energy consumption. All countries where the restrictions were imposed saw a sharp decline in GDP and energy consumption. Thus, the COVID-19 pandemic disrupted operations in the global energy system. As a consequence, global energy consumption has undergone a process of sharp deceleration compared with the upward trend recorded since 2009, i.e., the previous global shock [43].

Due to the above-mentioned conditions, the international energy market has been characterized by instability. The COVID-19 pandemic has had a major impact on energy consumption in many sectors, including agriculture, production, financial, education, health, sports, hotels, tourism, catering, and food [44]. According to the forecasts of the International Energy Agency (IEA), the demand shock phenomenon in 2020 was to be distinguished by the largest reduction in energy demand in 70 years. Thus, it was estimated that global energy demand would be reduced by 6% compared to 2019. This type of situation represents a limitation that is seven times greater than the 2009 financial crisis [45]. In turn, in China, which is considered the epicenter of the pandemic, the lockdown began in late January and ended in late March 2020, during which the highest level of energy reduction of 6.5% was recorded. The restrictions introduced in connection with the situation resulted in a further decline in energy demand, which was felt in the following weeks by the world’s largest economies, including France, India, Italy, Spain, Great Britain, and the US West Coast. In these countries, daily energy consumption decreased by a minimum of 15% [46].

In addition, restrictions caused by lockdowns and company closures contributed to a global recession, resulting in the GDP falling by 4.4% in 2020. In turn, global electricity demand decreased by 5% [47]. Nevertheless, the process of increasing energy demand was initiated in individual countries, along with the partial lifting of restrictions.

2. Materials and Methods

In the undertaken research, the subject was the analysis of the dependence resulting from the level of electricity consumption in service enterprises concerning the GDP per capita growth among 28 countries belonging to the European Community in the years 2000–2020 and the impact of the global COVID-19 pandemic on this process. In the process of diagnosing the indicated issues, data from statistics for European Union countries collected by the Statistical Office of the European Communities (Eurostat) were used. The analysis was conducted for 28 countries of the European Community: Austria (AT), Belgium (BE), Bulgaria (BG), Croatia (HR), Cyprus (CY), Czechia (CZ), Denmark (DK), Estonia (EE), Finland (FI), France (FR), Germany (DE), Greece (GR), Hungary (HU), Ireland (IE), Italy (IT), Latvia (LV), Lithuania (LT), Luxembourg (LU), Malta (MT), the Netherlands (NL), Poland (PL), Portugal (PT), Romania (RO), Slovakia (SK), Slovenia (SI), Spain (ES), Sweden (SE) and the United Kingdom (UK—data until 2019 due to the UK leaving the EU structures). The database was for: 1. Energy consumption factors among service enterprises, a set of 10 years (2010–2019); 2. Determination of variability in electricity consumption, a summary of 25 years (1995–2020); and 3. Definition of the correlation between energy consumption electricity and GDP per capita growth, a list of 21 years (2000–2020).

The collected data were analyzed statistically using Statistica and MATLAB and Simulink software. In the first stage, Principal Component Analysis (PCA) was used, in the second stage, Discrete Wavelet Transform (DWT), and in the third stage, Linearized Nonlinear Regression (LNR).

This article uses a hybrid method of statistical data analysis, which consists of principal components analysis, wavelet analysis, and linearized nonlinear regression. Energy consumption testing can also be performed using competing quantitative and qualitative methods. The first is based on preliminary qualitative research, the aim of which is to explain the reasons for the behavior of the analyzed phenomena and to comprehensively understand and correctly interpret the activities undertaken without referring to any research techniques and tools. The analysis is made following the isolation of phenomena and the components that can be displayed qualitatively in them, or the relationships and dependencies between them and the functions they fulfill [48]. In turn, during quantitative research, data are collected and then subjected to statistical and econometric analysis. Thus, the application of this approach is related to assigning certain measures to the analyzed phenomena. The main purpose of this method is to collect, analyze, and present measurable data [49]. Qualitative and quantitative research differ in the size of the research sample, the scope of the issues, the type of research questions, methods, techniques, research tools, and ways of interpreting and formulating conclusions using the obtained results [50].

Integrated research methods are developed on the basis of qualitative and quantitative approaches. Combining different concepts during research requires the use of various research methods in practice, which allows relevant and real results to be obtained. There are ten procedures used in the process of combining research methods in the so-called hybrid method: 1. Triangulation, the aim of which is to authenticate the research results; 2. Complementarity of selected approaches; 3. Integration of methods to generate a broader research horizon; 4. Research on the structural features of the phenomenon using quantitative methods, and process circumstances using a qualitative method; 5. Presentation perspectives of the respondents (qualitative methods) and researchers (quantitative methods); 6. Supplementing the qualitative results with quantitative ones reduces the problem of the research sample being representative; 7. The use of qualitative results for the interpretation of relationships between variables in quantitative datasets; 8. Showing the relationship between the micro and macro levels of the studied phenomenon; 9. Application of selected approaches during various stages of the research process; and 10. Shaping integrated forms.

Research conducted using the hybrid method differs from competing research approaches in terms of complexity, complementarity, polymethodicity—triangulation, and coherence.

2.1. Principal Component Analysis

In the initial stage of the research, Principal Component Analysis (PCA) was used to analyze the collected data. It is considered to be one of the best methods of factor analysis. The analysis consists of the orthogonal transformation of the n-dimensional data set describing the properties of a given phenomenon into a new structure of variables, the so-called principal components (PC), with a dimension less than n and constituting a linear combination of the original variables [51]. Such a transformation consists of the fact that the variances of successive variables take a smaller and smaller form, and the total variance of all analyzed data creates an identical sum of variances of the main components [52]. Thus, the role of PCA analysis is to determine subsequent PCs.

The first component (PC1) is formed by recognizing the vector $a_1=\left[a_{11},a_{21},\dots ,a_{p1}\right]$, which has a unit length $\left({\sum }_{i=1}^p{a}_{i1}^2=1\right)$, so that the linear relation is the structure:

$$Z_1=a_{11}{X}_1+a_{21}{X}_2+\dots +a_{p1}{X}_p$$

(1)

with the maximum variance between the set of this type of linear combination, where $X_i\left(i=1,2,\dots ,p\right)$ are the vectors of the observed values of random variables.

On the other hand, the second main component (PC2) is created by identifying the unit vector $a_2=\left[a_{12},a_{22},\dots ,a_{p2}\right]$, which is orthogonal to the vector PC1 $a_1=\left[a_{11},a_{21},\dots ,a_{p1}\right]$, therefore it implements the formula ${\sum }_{i=1}^p{a}_{i1}·a_{i2}=0$, and the equation formed by it:

$$Z_2=a_{12}{X}_1+a_{22}{X}_2+\cdots +a_{p2}{X}_p$$

(2)

which provides the maximum variance between a community of such linear combinations [53].

Reducing the space dimension of the analyzed features and grouping them into subsets allows a graphic explanation of the relations between the studied variables to be created, and the determination of their significance. The first few components are assumed to concentrate a significant part of the variability in the original dataset [54]. Thus, the goal of PCA is: 1. A reduction of variables to a smaller number of represented and uncorrelated dimensions, i.e., new variables; 2. An indication of relationships between the analyzed data; and 3. Classification of variables and cases.

In the process of analyzing the collected data, the normalized varimax rotation (maximum of the variance) was used, which enabled the maximum differentiation of loads within the factor’s range. As regards individual factors, the variables concentrating the maximum factor loadings concerning the dependent factor were determined (using the adopted value of ≥0.7). The graph of coordinates generated as a result of the analysis shows the relationship between the objects for the principal components. Meanwhile, the values of factor loadings remain the correlation coefficients between the original data and the main components [55].

For this study, the analysis of the analyzed issues included 17 factors that affect the consumption of electricity among service companies operating in the European Community. The following conditions were taken into account in the research: X1—Energy productivity; X2—Key indicators of physical energy flow accounts. Net domestic energy use for energy purposes; X3—Key indicators of physical energy flow accounts. Energy input and output; X4—Physical energy flow accounts for totals bridging to energy balances totals. Gross inland energy consumption—territory principle; X5—Exports of electricity and derived heat by partner country. Electricity; X6—GDP per capita; X7—Imports of electricity and derived heat by partner country; X8—Electricity production capacities by main fuel groups and operators; X9—Summary energy balance and indicators as % of GDP; X10—Innovation Index; X11—Key indicators of physical energy flow accounts. Extraction of natural energy inputs; X12—Total number of electricity retailers to final consumers; X13—Supply, transformation, and consumption of electricity; X14—Energy efficiency; X15—Final energy consumption; X16—Cooling and heating degree days by country; and X17—Production of services. Volume index of production.

The PCA was carried out due to the possibility of: 1. A reduction of variables to a smaller amount of represented and uncorrelated data, thanks to which new variables were obtained (the so-called main components); 2. Determination of the relationship arising between the variables; and 3. Grouping of variables. The use of PCA for the exploration of the above-mentioned data enabled the implementation of an intensified analysis of the primary variables.

2.2. Discrete Wavelet Transform

Wavelet analysis is the decomposition of a signal into components that function with a specific location in time and a defined scale. Therefore, it is referred to as “time-scale” analysis because there is an analogy between low scale and high frequency and between large scale and negligible frequency [56]. The components are shifted and scaled versions of the so-called basic wavelet as a continuous, square-integrable function $\Psi \left(t\right)\in L^2\left(R\right)$, which realizes the following assumptions of admissibility:

$${{\displaystyle \int }}_0^{+\infty }\frac{\left|\Psi \left(\omega \right)\right|{ }^2}{\omega }d\omega <\infty$$

(3)

from which, it follows that:

$${{\displaystyle \int }}_0^{+\infty }\Psi \left(t\right)dt=0$$

(4)

where Ψ(ω) is the Fourier transform of the function Ψ(t).

In the form of a wavelet, functions that are approximately consistent with the formula defined using Formula (4) are used. Wavelet analysis is based on the Continuous Wavelet Transform (CWT) or the Discrete Wavelet Transform (DWT). Although it is not an autonomous prognostic technique, its parameters make it possible to use it to forecast time series. In particular, series that are distinguished by being non-stationary and represent short-term oscillations of uneven amplitude, for which the preceding links in the causal chains are connected with a time scale (decision horizon) [57].

Usually, the analysis performed with DWT is used in the assessment of prognostic processes, where the results are the wavelet coefficients determined for the octaves of frequency. The recognition of only the octaves themselves may be justified in the case of an investigation of economic processes. DWT results in splitting the signal into two components, called approximation and detail, as a result of the separation by a low-pass (a_j) and high-pass (d_j) filter and the operation of re-sampling the signal by selecting only even samples. The discrete wavelet transform of signal x(t) is as follows [58]:

$$x\left(t\right)={\displaystyle \sum }_k{a}_{j0}\left(k\right){\omega }_{j0,k}\left(t\right)+{\displaystyle \sum }_k{\displaystyle \sum }_{j0}^{\infty }{d}_j\left(k\right){\Psi }_{j,k}\left(t\right)$$

(5)

where Ψ(t)—wavelet function, ω(t)—scaling function, djk—high-frequency coefficients that make up the set of the so-called details, ak—the coefficients contain low-pass information and a constant component, i.e., they constitute the so-called signal approximation.

During the analysis, the approximation is subordinated to the next decomposition into approximation and detail, which are not subject to subsequent divisions. As a result, the signal is presented as a sum of approximations of the last level of the total details from all levels. For this article, a discrete Meyer wavelet (the so-called dmey) was used to shape the diagnostic signals.

2.3. Linearized Nonlinear Regression

Regression examines the interaction of one or more independent variables (X₁, …, X_k) on the dependent variable (Y) using an appropriate mathematical function. In a situation where the explained variable is influenced by several explanatory variables, this type of regression is called multiple. It makes it possible to present the dependencies of variables in a functional form, among others, in regression models [59]. Regression analysis refers to the determination of the regression function using the appropriate selection of its structure and the calculation of the evaluation of parameters. A commonly used regression equation is the function [60]:

$$y=f\left(x\right)+\epsilon$$

(6)

where y—dependent variable, f(x)—linear regression function, ε—random component.

A regression model follows the type of function on which it is based. As a result, it can take the form of a linear or nonlinear function. For linear regression in the empirical database, a straight line is adjusted using the least squares method, which optimally illustrates the relationship between the variables. In turn, the linear multiple regression model can be written by equation [61]:

$$y={\beta }_0+{\beta }_1{X}_1+{\beta }_2{X}_2+\cdots +{\beta }_k{X}_k+\epsilon$$

(7)

where y—dependent variable, β₁—model parameters (regression coefficient, estimators), showing the influence of successive independent variables (for i = 1, …, k) on the dependent variable, X_i—i-th dependent variable, $\epsilon$ Ε—random component.

Nevertheless, in some circumstances, the relationships between the phenomena studied do not have a rectilinear form but a curvilinear one. Therefore, a nonlinear regression model differs from a linear model in terms of the type of mathematical function and the method of estimating the parameters of the function. Thus, in the nonlinear regression model, a curve is fitted to the variables that adequately show the relationship between the analyzed data. The general structure of a nonlinear function can be represented by the formula [62]:

$$y_i=ɳ\left(x_i,\beta \right)+{\epsilon }_i$$

(8)

where, $ɳ$(x_i,β)—any differential function, continuous and differentiable, x_i = [x_1i, x_2i,…,x_ki]^T—T—vector of k independent variables, β = [β₁, β₂,…, β_m]^T—vector m of the coefficients, ε_i—random deviation.

In the case of nonlinear regression, it is recommended to perform linearization of the relations between variables at the beginning. This involves transforming a nonlinear trend into a linear one using the principle of logarithmic pairs of numerical values of the dependent and independent variables to reduce the data to a linear function to obtain its linearized distribution.

3. Results

3.1. Determinants of Energy Consumption Variability among Service Enterprises in the EU

In the first stage of the research, the factors that influence and condition changes in energy consumption by service entities operating in EU countries were identified following the regulations adopted in the field of increasing energy efficiency. To present the determinants influencing the non-systematic use of energy, PCA was used. The method implemented to reduce the dimensionality of the variables made it possible to present the determinants contributing to the transformation of energy consumption while maintaining the maximum scope of information. As a result of minimizing the amount of data necessary to explain a specific variable, PCA improved the analysis of the obtained results. In the analyzed set of 28 countries making up the European Community in 2020, the X1–X17 variables included in the chapter on the methodology of principal components analysis were used (see point 2).

The technique of graphic PCA projection allowed the dependencies and disproportions between the examined energy consumption factors in the juxtaposition of the first two main components to be highlighted (see Figure 2).

In 2010, the first component (see Figure 2a) explained 63.08% of the variability in the data, while the second component was shaped by variables with opposite signs. This informs the high value of the GDP per capita factor, which corresponds to a negligible number of electricity sellers to end consumers concerning the total energy balance of the EU countries and the % of GDP indicators, and the opposite. At the same time, the second component explained 14.46% of the data variability. Thus, both components explained 77.54% of the variance of the original variables.

According to the PCA results, it should be concluded that the factor loadings that have a positive correlation are responsible for the way of using the factors influencing the variability of energy consumption in the EU countries. In particular, they include factors such as energy productivity and the supply, transformation, and consumption of electricity; gross inland energy consumption (the territory principle)—production in services (volume index of production); electricity production capacities by main fuel groups and operators—final energy consumption; electricity production capacities by main fuel groups and operators—extraction of natural energy inputs; energy input and output—energy efficiency; and GDP per capita—innovation index.

Among the negatively correlated factors, one should distinguish the conditions influencing the variability of energy consumption in EU countries: imports of electricity and derived heat by partner country—cooling and heating degree days by country, as well as energy productivity—summary energy balance and indicators as % of GDP.

The analyzed variables are also characterized by uncorrelated determinants, which include GDP per capita—supply, transformation, and consumption of electricity; innovation index—net domestic energy use for energy purposes; energy productivity—final energy consumption; exports of electricity and derived heat by partner country (electricity)—summary energy balance and indicators as % of GDP.

In 2015 (see Figure 2b), the factors determining changes in energy consumption among EU countries were represented in 62.07% by the first component and in 15.47% by the second component, accounting for a total of 77.54% of the variance of primary variables. Based on the determination of the factor loadings, the positive correlation is represented by the following factors influencing the use of energy: supply, transformation, and consumption of electricity, as well as final energy consumption; net domestic energy use for energy purposes—energy efficiency; electricity production capacities by main fuel groups and operators—net domestic energy use for energy purposes; energy input and output—energy efficiency; exports of electricity and derived heat by partner country (electricity)—total number of electricity retailers to final consumers; and extraction of natural energy inputs—total number of electricity retailers to final consumers.

Among the factors that indicate a strong negative correlation in energy consumption, the following should be distinguished: cooling and heating degree days by country—energy productivity; and summary energy balance and indicators as % of GDP—innovation index.

In addition to the above-mentioned correlated factors, there are also recognizable features that are not related, e.g., GDP per capita—final energy consumption; innovation index—supply, transformation, and consumption of electricity; energy productivity—gross inland energy consumption (the territory principle); and imports of electricity and derived heat by partner country—cooling and heating degree days by country.

In 2019 (see Figure 2c), the components accounted for 61.31% (first) and 11.72% (second) of the variability of the analyzed factors. In total, they represent 73.03% of the variability of the studied data, which affects the variability of energy consumption in EU countries. Due to the analysis of the distribution of individual determinants responsible for the circumstances accompanying energy consumption in the chart of factor coordinates, it should be stated that the most strongly correlated are: Exports of electricity and derived heat by partner country (electricity) and the total number of electricity retailers to final consumers; final energy consumption—supply, transformation, and consumption of electricity; gross inland energy consumption (the territory principle)—energy efficiency; and energy input and output—energy efficiency.

The graphical list of the analyzed variables also shows that the summary energy balance and indicators as % of GDP—energy productivity; cooling and heating degree days by country—innovation index are negatively correlated.

Nevertheless, the considered data also show that there is no correlation between the following energy consumption factors: GDP per capita—total number of electricity retailers to final consumers; extraction of natural energy inputs—summary energy balance and indicators as % of GDP; energy productivity—production in services (volume index of production); and innovation index—exports of electricity and derived heat by partner country (electricity).

The last stage of the research enabled the results to be obtained for projecting the location of EU countries (objects) on the planes of the main factors to identify the determinants for the variability of energy consumption (see Figure 3).

According to Figure 3a, in 2010, we can distinguish three compact clusters of points (countries) where the consumption of energy was influenced by identical factors. The first group is represented by Croatia (HR), Malta (MT), Estonia (EE), Lithuania (LT), Latvia (LV), Bulgaria (BG), Slovakia (SK), Hungary (HU), Czechia (CZ), Romania (RO), and Poland (PL). The second group consists of Austria (AT), Sweden (SE), the Netherlands (NL), Finland (FI), Belgium (BE), Ireland (IE), Cyprus (CY), Slovenia (SI), Portugal (PT), Denmark (DK), and Greece (GR). The third group is formed by Spain (ES), the United Kingdom (UK), Italy (IT), and France (FR). In addition, when analyzing the distribution of points on the map, two countries can be noticed, Luxembourg (LU) and Germany (DE), which are different from the others.

The linear maps for 2015 (see Figure 3b) also make it possible to distinguish three homogeneous groups of countries representing homogeneous factors determining energy consumption. The first community is created by Malta (MT), Slovenia (SI), Slovakia (SK), Portugal (PT), Greece (GR), Czechia (CZ), Hungary (HU), Lithuania (LT), Romania (RO), Bulgaria (BG), Croatia (HR), Latvia (LV), Estonia (EE) and Cyprus (CY). The second set is formulated by Luxembourg (LU), Denmark (DK), Ireland (IE), Finland (FI), Belgium (BE), the Netherlands (NL), Austria (AT), and Sweden (SE). The third group consists of the United Kingdom (UK), Italy (IT), Spain (ES), and Poland (PL). In the list of countries, two countries stand out in terms of energy consumption from the rest of the surveyed population: France (FR) and Germany (DE).

In 2019, the same circumstances determining changes in energy consumption by service entities were also observed among the three clusters of EU countries (see Figure 3c). The first of them represents the following countries: Ireland (IE), Greece (GR), Belgium (BE), Finland (FI), Czechia (CZ), Portugal (PT), Romania (RO), Hungary (HU), Slovakia (SK), Croatia (HR), Estonia (EE), Latvia (LV), Malta (MT), Cyprus (CY), Lithuania (LT), and Slovenia (SI). The second group is designated by Luxembourg (LU), Denmark (DK), Austria (AT), Sweden (SE), and the Netherlands (NL). The third group consists of Spain (ES), the United Kingdom (UK), Poland (PL), France (FR), and Italy (IT). The linear map prepared for 2019 shows two EU countries, Bulgaria (BG) and Germany (DE), which differ from the surveyed group.

The conducted research shows that ”he d’terminants of energy consumption in individual EU countries are influenced by three groups of causes, which can be classified as follows: 1. Energy and technical, 2. Financial and economic, and 3. Environmental. The first group includes technological factors (electricity production capacities by main fuel groups and operators; energy productivity; the total number of electricity retailers to final consumers; production in services, energy efficiency, innovation index, energy input and output, and extraction of natural energy inputs). The next set is created by socioeconomic factors (GDP per capita; exports of electricity and derived heat by partner country; and the total number of electricity retailers to final consumers). The third table represents environmental factors (cooling and heating degree days by country) that do not affect the level of energy consumption among service companies operating in the EU.

3.2. Energy Consumption in EU Countries

This stage of the research was aimed at obtaining a forecast of the energy consumption index in European Union countries. The analysis was carried out for the time series, which represented the energy consumption index in the sector of service enterprises operating in the European Community in the years 1995–2020. Discrete wavelet transform (DWT) was used to identify volatility tendencies in the level of energy consumption. The rationale for using this type of method is the ability to analyze changes in the amplitude of oscillation or decomposition of variables, both in terms of time and frequency. The study used the series of annual energy consumption for the period 1995–2020 (25 observations). Each series underwent a four-level wavelet decomposition for the sample, log₂31 = 4.95 ≈ 4 (see Figure 4). The four-level wavelet decomposition was performed in the MATLAB & Simulink environment using a regular, orthogonal, symmetrical, discrete Meyer wavelet (dmey).

The multidimensional decomposition of the series that presents the intensity of energy consumption among EU service companies enables their analysis in terms of both short (annual) and long (five-year) periods. The results of the decomposition of the wavelet coefficients (s, a₄, a₃, a₂, a₁) and the corresponding components (cfs, s, d₄, d₃, d₂, d₁) are presented in Figure 5.

The presented components have divergent scale values and different corresponding frequencies. Gradually, slowly changing signals are waveforms of wavelet coefficients marked as d₄ (see Figure 5b). This type of situation means a large-scale and low-frequency consumption of energy among service companies operating throughout the EU. In turn, the detailed components (d₃, d₂, d₁) have a scale smaller than d₄. The marginal scale among the detailed components is available for the d₂ courses. In turn, the d1 distribution is characterized by the lowest frequency, which indicates the signals with the highest frequency. When analyzing energy consumption among service enterprises operating in the EU at the approximation level at the fourth resolution level with a frequency of 1.5 (a₃—see Figure 5a), a downward trend can be clearly stated.

The course of the approximation and details obtained as a result of the decomposition of the diagnostic signal emitted by the energy consumption variables are shown in Figure 5c. The first, second, and fourth decomposition levels using a discrete Meyer wavelet were adopted for the analysis. In each of them, the tested signal is the sum of the detail approximation from this level and details representing all previous levels.

Based on the exploration of the wavelet series (see Figure 5b) and the sequences obtained by applying the Meyer wavelet decomposition, it should be concluded that energy consumption decreased in the final research period at the turn of 2019/2020. Thus, the obtained distribution of the series of the analyzed variables in the period 1995–2020 indicates the pandemic crisis, which influenced not only the health condition of employees but also the economic situation of service enterprises due to limitations in their functioning and the related reduction in energy consumption (see Figure 5c).

Using the data obtained from the Eurostat databases, it should be concluded that the structure of energy consumption during the COVID-19 pandemic (2020) was dominated by industrial enterprises, which recorded an increase in energy consumption by 37.7%. The same tendency was noticed among individual customers, with the consumption intensity reaching a level of 7.8%. In the case of service entities, there was a decrease in energy use by 14.1%.

In the next stage of the research, nonlinear linearized regression was used to analyze the relationship between the level of electricity consumption among service enterprises and the growth of GDP per capita of 28 EU countries in the years 2000–2020. The data analysis was carried out with the use of a linearized model because it enables the reduction to a linear model using an optimal transformation of the variables.

The performed regression analysis (see Figure 6) shows that the regression equation (p for the F test is 0.149) is significant, and the regression coefficient is significantly different from 0 (p < 0.001). On the other hand, the value of the correlation coefficient R = 0.91030 indicates a positive relationship with a very good straight line fit. It should be concluded that the residuals have a normal distribution and that the deviations of the points from the fitted line are random. The conducted nonlinear linearized regression analysis shows the relationship between the increase in energy consumption by service companies operating in EU countries, which is accompanied by an increase in average GDP per capita values. This assumption is confirmed by the R correlation coefficient, which indicates a strong correlation between the analyzed variables. In turn, the coefficient of determination R² explains 91% of the dependence of GDP growth on the increased level of electricity consumption by service enterprises in the EU. Thus, it should be concluded that by the cause-and-effect relationship in a bidirectional configuration, energy consumption strengthens GDP. A process of this kind contributes to increased energy consumption. Nevertheless, from 2020 onwards, a decrease in consumption is visible among EU service companies due to the global COVID-19 pandemic and its constraints and the implementation of the EGD strategy.

On the basis of the conducted research, it is necessary to confirm the hypothesis that EU countries support the process of energy transformation through policies promoting energy efficiency in the political, economic, and technological dimensions, which was de-stabilized under the influence of the pandemic crisis, and have recognized in it a strategic opportunity for economic development and a possibility of changes in the structure of energy consumption management among institutional recipients.

4. Discussion

Nowadays, energy is one of the key elements determining the development of civilization because the management process requires energy support resulting from access to and dependence on electricity in all areas of life and sectors of the economy. They operate with the use of electrical devices and objects made with machines powered by electricity or operated by industrial automation systems that are not able to operate without access to energy. Such a tendency is an irreversible process, and thus electricity consumption will create an upward trend.

The energy management system is characterized by a qualitative change concerning the issues related to the generation, transmission, distribution, and consumption of energy. Moreover, it plays an important role in the economic development of countries. Identifying energy suppliers and buyers in a selected geographical area allows for the analysis of energy supplies and the level of their consumption by individual groups of recipients. A systematic supply of energy is a necessary condition for the functioning of the modern global economy and of households. Adjusting the pace of economic growth to the obligation to prevent climate change is one of the key dilemmas of energy efficiency and sustainable development policy.

In the modern conditions of the economic downturn, investment slowdowns shaped by the COVID-19 pandemic crisis and its economic consequences, as well as difficulties resulting from forecasting the demand for individual energy sources, the implementation of programs and strategies is more complicated than in pre-pandemic times. Therefore, it is necessary to identify the key factors of energy consumption among service companies to determine the level of its consumption depending on the GDP per capita growth from the perspective of the pandemic crisis and the implementation of the European Green Deal strategy.

The research conducted so far in the area of the analyzed issues is especially focused on the analysis of factors affecting, or with the possibility of influencing, energy consumption. Most authors only focus on a few factors (usually from one to a maximum of three or four) that are verified in terms of the intensity of electricity consumption or total use of electricity. Nevertheless, much of the research deals with the relationship between total energy consumption and gross inland production.

The indicated relations were discussed in studies [35,37,63,64,65], in which economic growth was identified for the factors influencing energy consumption among individual and institutional consumers. It is an integral component of the increased use of energy. Although the cause-and-effect relationship between energy consumption and economic growth has been included in the subject matter of the research conducted so far, no clear agreement has been obtained regarding the indicated relationship. The nature of linkages viewed from the perspective of causation is a key factor taken into account by company and country managers. If such a relationship is from consumption to growth, then the strategy of saving (reducing) energy may adversely affect economic growth. Moreover, on the basis of the research results presented by F. Zhang, it should be concluded that economic growth is considered a permanent factor in increasing energy efficiency [66].

The studies carried out in 2000–2011 indicate a different type of relationship between economic growth and energy consumption. They aimed to check the correlation between the analyzed variables, for which the dynamic panel models were used according to the Blundell-Bond method and the system GMM estimator [67]. As a result of the statistical analysis, it should be noted that this type of relationship is one-way in nature, running from GDP per capita to energy consumption. In the opposite direction, the verified impact was statistically insignificant [68].

From the research conducted so far, it should be concluded that if there is a relationship between energy consumption and economic growth, this type of situation indicates an energy-dependent manner of economic growth. Thus, energy is a generator of economic growth, and its shortage harms economic growth. The unidirectional cause-and-effect relationship between the analyzed issues means that economic growth contributes to the increase in energy consumption. Thus, it should be recognized that economic growth is not based on energy consumption. Therefore, the energy-saving strategy will have a lower impact on economic growth, or its impact may be completely negligible. Meanwhile, the cause-and-effect relationship in a bidirectional form indicates that energy consumption supports economic growth, which results in a more significant use of energy. Conversely, if the research confirms the lack of a causal relationship between the level of energy consumption and economic growth, then the concept of energy saving will not result in economic growth [69].

The PCA carried out for this study indicates three groups of factors that have a major impact on energy consumption. The first group consists of political factors resulting from the adaptation of the law in force in individual countries to EU regulations in the field of energy efficiency. The second group is represented by technological factors, such as production in services, innovation index, and electricity generation capacity. The third group consists of socioeconomic conditions resulting from economic development (GDP per capita) and the total number of electricity sellers to end consumers. During the research, the fourth group of geographic and meteorological factors was also distinguished. Nevertheless, it does not have a direct impact on energy consumption in EU countries due to the analyzed degree-days of cooling and heating.

Due to the high share of industrial and service enterprises in the structure of gross domestic product (GDP), there is a greater energy demand. This type of process is intensified when economic growth manifests itself in the economic development of a given country and the increased purchasing power of the population. Therefore, among the countries that are currently developing, one should expect an intensification of the energy consumption process [66,70,71] from the perspective of the relationship between consumption and economic growth [72,73,74].

The PCA linear maps prepared for this study (Figure 2a–c) indicate groups of countries with identical energy consumption factors. The first is shaped by countries located in the eastern part of the EU. The second is made up of countries located in the northern part of the European Community. The third is constituted by countries located in the Mediterranean basin.

When analyzing the research conducted so far, it is necessary to indicate four divergent hypotheses posed as a consequence of the analysis of dependencies resulting from the studied issues [75,76,77,78,79]. The first hypothesis states that energy consumption is an essential element of economic growth, viewed directly or indirectly from the perspective of capital and labor supplementation as factors of production. Consequently, reducing energy consumption generates a decline in the real GDP. The second thesis is based on a unilateral causal relationship, which shows that low energy consumption measures may ultimately have an unnoticeable or limited impact on the real GDP. In turn, the third hypothesis results from a two-way dependence of the nature of feedback on the basis of which energy consumption and real GDP interact simultaneously. However, the fourth hypothesis, the so-called neutral one, proves that reducing energy consumption has no effect on economic growth, and vice versa. According to this type of thesis, the energy-saving strategy does not have any impact on real GDP.

The analysis of the results of 48 studies on the relationship between electricity consumption and the impact of GDP confirmed the above hypotheses. In the case of 29% of research works, the impact of energy consumption on domestic production was found. On the other hand, 27% confirmed the hypothesis regarding feedback, and 23% emphasized the statistically significant impact of GDP on energy consumption. Meanwhile, in 21% of the analyses, no relationship was found between the analyzed issues [79]. Moreover, such issues were compared among the populations of developed and developing countries. In the first group of countries, the link has a two-way form, i.e., the impact of GDP per capita on energy consumption is noticeable, as is the degree of energy use on GDP. However, in the group of developing countries, a one-way relationship was observed in which an increase in GDP per capita contributes to an increase in energy consumption [80].

The research conducted for this article with the use of linearized linear regression confirms the cause-and-effect relationship with a two-way structure, indicating the relationship that determines the increased energy consumption in service enterprises, which contributes to the growth in GDP per capita.

In addition to the assumptions mentioned above, the literature on the subject provides information that higher GDP per capita also affects other energy efficiency factors. It enables investments in new technologies and the multiplication of capital [66]. Nevertheless, various types of energy-saving solutions make it easier for countries, especially countries in transition, to transfer the level of energy necessary to guarantee economic growth and social development.

Previous studies on the impact of the COVID-19 pandemic on energy consumption paid special attention to the problems resulting from the level of energy demand, which, due to its impact, causes significant variability in the sphere of energy demand and supply [45,74,81,82]. Nevertheless, the current data showing the actual economic impact of the COVID-19 pandemic are partial and difficult to estimate due to the unique situation of the continuing successive waves of its influx. In such a situation, future economic forecasts to a large extent depend on electricity consumption, which informs us about the variability of its demand [75,83,84].

The research carried out for this study using the wavelet transform shows that in the final stage (2019–2020), a reduction in energy consumption in EU service enterprises is noticeable. Thus, it should be stated that the socio-economic consequences of the crisis caused by the COVID-19 pandemic are borne to this day not only by entities operating in the service sector, but also by countries belonging to the European Community.

During a pandemic, energy demand should be stabilized, e.g., broken down into its systematic demand for enterprises, institutions, and individual recipients as well as complimentary, but necessary for the needs of coping with the COVID-19 pandemic (e.g., energy consumption in materials, research and development sector, supply chains).

5. Conclusions

The use of energy is an important problem for service companies operating in the European Union, which are obliged to increase energy efficiency and reduce greenhouse gas emissions resulting from the implementation of the European Green Deal strategy. The inefficient use of energy generates a significant portion of the costs of business operations. For this reason, theoretical, methodological, and empirical research is carried out on the ways, methods, and strategies of effective energy management. In addition, opportunities to reduce electricity consumption through its rational use are sought by analyzing factors, considering changing trends, and examining the cause-and-effect relationships arising between energy consumption and GDP per capita growth in EU countries.

In line with the adopted research objective, this article presents changes in the use of electricity by service companies that are a consequence of the implementation of the EGD strategy. To achieve the set goal and obtain an answer to the research question, hybrid methods of statistical data analysis (PCA, DWT, and nonlinear linearized regression) were used. The study made it possible to identify the factors influencing energy consumption, among which the following sources were distinguished: political, technological, socioeconomic, and geo-meteorological. These determinants are used by three groups of EU countries located in the eastern and northern parts of the European Community and the Mediterranean.

In addition, the analysis carried out for the years 1995–2020 showed variability in consumption. Particularly noteworthy is the last year of research, in which the reduction in energy use was influenced by the economic situation of service enterprises resulting from the restrictions introduced by EU governments during the COVID-19 pandemic and the EGD strategy presented at the end of 2019. On the other hand, the diagnosis of the cause-and-effect relationship between energy consumption and GDP indicated the use of a two-way structure in the form of a feedback relationship. Thus, it should be noted that strong economic growth is correlated with the natural process of increasing energy demand. Nevertheless, from 2020, a 14.1% reduction in its use can be noticed among service enterprises operating in the EU. The obtained distribution of the series of analyzed variables indicated the initial phase of the implementation of the EGD strategy and the pandemic crisis, which influenced not only the economic situation of the studied entities but also limitations in their operation and the associated reduction in energy consumption.

The conducted research justified the hypothesis, which showed that EU countries support the process of energy transformation with the use of policies promoting energy efficiency in the energy-technical, financial-economic, and environmental dimensions by the EGD. In addition, they strive to use effective energy management methods that have been destabilized by the pandemic crisis. They are seen as a strategic opportunity for economic development and the possibility of changing the structure of energy consumption management among service entities.

The results of the conducted research provide the managers of service enterprises and countries located in EU countries with information that can be used in the modification and implementation of the EGD strategy in the scope of reducing energy consumption, improving energy efficiency, and the impact of the economic growth rate on energy use.

This article is a multi-faceted approach to the analysis of the European Green Deal strategy from the perspective of challenges to energy management in service companies located in the European community. Therefore, future research should focus on identifying the factors influencing energy efficiency from the point of view of energy use in service provision processes and management by individual buyers. Due to the current pandemic situation, further analyses of the circumstances affecting the level of energy consumption are particularly important. Individual EU countries are struggling with the COVID-19 pandemic, hence the importance of coordination and cooperation in the field of regular access to energy, energy security, and energy justice.