Do Businesses Protect the Environment Through Appropriate Decisions in the Context of Choosing Information and Communication Technologies?

Abstract

Technological progress, digitalization and globalization of economic activity contribute to the growth of the use of information and communication technologies in enterprises. On the one hand, modern information technologies support pro-environmental activities, but on the other hand, they are a source of waste themselves. For this reason, their use in enterprises should be analyzed and controlled in the context of their multi-faceted impact on the natural environment. This article focuses on: (1) analyzing the relationship between variables defining the level of actions taken in the field of ICT to protect the natural environment and the level of digitalization of the enterprise and (2) identifying a synthetic measure of development defining the level of involvement of enterprises in digitalization and environmental protection. The analysis will be performed on data describing actions taken by enterprises grouped by European Union countries in the context of the use of ICT equipment for environmental protection and environmental values for 2022. Based on the chi-square test, it was found that statistically significant relationships are observed only for medium and large enterprises. The synthetic measure of development allowed for the indication of model countries due to the actions taken aimed at pro-environmental behavior in relation to ICT services and equipment. There were also no significant linear relationships between a high level of digitalization of enterprises and thinking in environmental categories in the context of actions taken in relation to ICT services and equipment.

1. Introduction

The selection, implementation and application of information and communication technology (ICT) tools and system components impact the environment; hence, they require thorough consideration from an environmental perspective. The ICTs have a twofold impact on the natural environment. The use of modern information technologies promotes pro-environmental activities; however, they are a source of waste.

The economy and society are shaped by sustainability and digitalization according to major trends [1]. The economic and environmental impacts of ICTs are expected to be favorable, to stimulate growth and to mitigate adverse environmental effects; however, the ICTs’ role in the economy and environment is found to be ambiguous and is controversially discussed [2,3]. Economic growth and energy consumption increase the ecological footprint [4]. Technology diffusion, as described by ICT mobile, and internet and non-renewable energy consumption degenerates environmental quality in the long run, while by contrast, renewable energy and technological innovation improve environmental quality [5]. As for industrial manufacturing processes, research has found that their chemical oxygen demand (COD) emission has been significantly reduced as a result of industrial digital transformation [6]. The ICTs reshape the economic production process and residential life. They are able to control the rising energy demand through improving energy-use efficiency or energy productivity; potentially, this indirect impact of ICT can save much more money than the direct impact of ICT itself [7]. When the relevant systems and their interaction effects are ignored, the evaluations of ICT implications result in biased estimates [2]. Consequently, the impact of ICTs should be analyzed in a context that captures both the positive and negative impacts and direct and indirect aspects in the short and long run.

1.1. Theoretical Foundation

The first stage in examining the relationship between digitalization and environmental sustainability is to identify the meaning of digitalization itself and the concepts associated with it. This is essential because, although these concepts have been with us for decades, even recent publications [8,9] have struggled to define what is meant by this term or have simply treated it as a common term and shied away from a more detailed definition or conceptual delineation [10]. Therefore, following the approach of some authors, we briefly summarized the most relevant characteristics of digitization [11]:

• Using the data collection capabilities of ICT infrastructure, they are better integrated into production and/or service processes (IoT) [12];
• ICT infrastructure’s decision-making possibilities allow better integration into production and/or service processes (Artificial Intelligence) [13];
• The control capabilities of the ICT infrastructure enable AI decisions to be better integrated into production and/or service processes (IoT) [12];
• Direct-to-device (M2M) communication [14].
• As a consequence of these characteristics they have:
• Access to much more and better-quality data than before (Big Data) [11];
• More intensive and broader information representation than before (virtual reality and augmented reality) [11].

And last but not least, there is a significant transformation in the nature of the ICT infrastructure underpinning services:

• The locally deployed ICT infrastructure has become much more complex: a much more sophisticated ICT infrastructure in terms of the quantity, and the purpose of use (network devices, congestion devices, etc.) provides the needs presented earlier [14];
• With the advent of cloud-based solutions, it has become possible to outsource the tasks related to the operation of the ICT infrastructure to a specialized service provider so that some of the services of the ICT infrastructure can be accessed independently of geographical location [11];
• And logic requires one to mention here also the vehicles involved in production, operating without human intervention, typically used in precision agriculture [15].

These technologies are collectively referred to as INDUSTRY 4.0 technologies [11]. In this context, the concept of “digital transformation” should be mentioned, whereby the servers involved change the way they operate from the previous analogue technologies to digital technologies [16]. This entails a radical transformation of previous business models or the creation of new business models [17]. These technologies are expected to reduce business costs and increase results, but this approach excludes the consideration and evaluation of externalities [18].

1.2. Approaches in Analyzing the Relationship Between Digitization and the Environment

Reviewing the previous literature on the relationship between digitalization and environmental sustainability, one must conclude that several approaches should be used in the analysis. The first approach is the nature of the impact. Within this we can differentiate the following [19]:

• Indirect effects;
• Direct effects.

The direct impact can be assessed if the integration of INDUSTRY 4.0 solutions enables the company concerned to implement a more efficient production process. Of course, it is possible to identify direct impacts with positive and negative connotations. Such positive indirect effects could be the reduction of production waste or the further diffusion of JIT production processes with minimum or zero production stock [20]. Negative direct effects include safety risks due to the interaction between robots (or automated devices) and humans, or higher energy requirements in production/service processes [18].

For indirect effects, we can also identify positive and negative impacts. Positive impacts include the wider use of data generated during the manufacturing/service process [16], and negative indirect impacts include the reduction of the need for a live workforce [18]. Unfortunately, negative indirect impacts include the increase in the production of digital assets that are otherwise difficult to recycle as a result of the use of INDUSTRY 4.0 solutions [19,21].

The second approach is the impact mechanism:

• Impact on the configuration of production [22];
• Effects on product [22];
• Impact on the ICT Infrastructure as a whole [23].

The impact on the mode of production is relatively easier to see; this includes the so-called smart factories (or precision agriculture, as mentioned earlier), where more efficient production processes can be developed by applying the elements in the ICT infrastructure described earlier at the system level [16].

The same is true for the effects on products. Many authors point to the positive impact of INDUSTRY 4.0 when analyzing, for example, their product life cycle. In addition, the effect that a digital product can be a perfect substitute for a physical product has been demonstrated. A good example is the relationship between printed books and e-books [19].

The impact on the ICT infrastructure as a whole: there are more sub-questions to consider than this, but this is where we are in biggest trouble, and the problem is multifaceted. In practice, the issue of “green ICT” was already raised before the INDUSTRY 4.0 era; the production and operation of ICT infrastructure should be practiced in a more environmentally friendly way. If we observe that these studies [23] raise a number of problems (non-exhaustive: life cycle of some components of ICT infrastructure, energy consumption, etc.), we can see that these problems have not been solved but only transformed since the INDUSTRY 3.0 era [19]. In the cited study, we can find a number of data on the extent to which the production of certain ICT devices consumes the natural environment. The source cited quotes the fact that ‘the extraction and refining of one tonne of rare earths can produce 60,000 m³ of waste gas containing liquid hydrogen fluoride, resulting in 200 m³ of acidic wastewater’. In this context, it is also worth considering the fact that—in addition to the fact that the production of ICT devices in itself is a significant use of the natural environment—the vast majority of these devices are not recyclable or are only recyclable to a small extent (up to 30%) [24].

The situation is similar for the operation of ICT infrastructure. The operation of ICT infrastructure requires significant energy consumption. This picture is complicated in no small measure by the fact that, with the spread of cloud computing, the energy consumption of individual users is decreasing, but huge consumers are emerging: data centers [25].

If we observe success stories [26,27] and analyses [28,29] of the INDUSTRY 4.0 implementation (transformation process), natural sustainability aspects are not included in the maturity criteria (in many cases, the focus is on the benefits of INDUSTRY 4.0 implementation). However, this is also true in reverse, even for cited publications [19,23,24]. Where the negative environmental benefits of INDUSTRY 4.0 solutions are analyzed, the benefits of those solutions are rarely taken into account. There is a lack of research that could address these two aspects, which are very difficult to compare.

To conclude, if an economic actor wants to take into account natural-environmental factors in its digital transformation decisions, it can do so in a very limited way, typically by taking into account production and its direct consequences. Therefore, it can only make a short- to medium-term decision based on partially available data, within the framework of a limited set of criteria.

1.3. ICT Life Cycle and the Environment

The production and the use and end-of life phases of a product’s lifecycle have environmental impacts. The environment impact of a product can be assessed by lifecycle assessment [30]. The impact on the environment should consider all aspects. The production requires natural resources and the increased consumption leads to growing amount of electronic waste [31]. When the lifecycle of a product ends, the product becomes obsolete or is changed by a user to an upgraded version, and as the product is no longer in use, the question of electronic waste treatment arises. The leading collector of electronic waste (e-waste) is Europe, followed by Asia, America, Oceania, and Africa [32]. Collection, sorting and inhomogeneity of waste, low energy density, prevention of further waste, emission and cost-effective recycling are found to be the major challenges of e-waste treatment [32]. The extraction of scarce materials and heavy metals used in telecommunications equipment and the treatment of waste mean a large environmental challenge [33]. A piece of computer equipment is a complicated assembly and has more than a thousand materials, many of which are highly toxic, and there is a health hazard for the workers involved [34].

The extension of the lifespan and the use of such products mean an important and urgent matter [31]. ICT remanufacturing is a growing industry; however, it faces several difficulties, such as the perception that re-used and remanufactured products are of lesser quality [31]. It was found that computer recyclers have high levels of dangerous chemicals in their blood [34], indicating that health and safety should always be a priority. The decision on common chargers for mobile devices in the European Union was a one that not only makes users’ lives more convenient, but is a meaningful move that make compatible use of chargers possible; hence, it can contribute to lower amount of electronic waste in the long run, which is the desired outcome from environmental perspective [35].

1.4. Ways to Reduce the Negative Impact of ICT on the Environment

Companies of these days operate in an environment that is surrounded by ICTs; during the COVID19 pandemic, digital transformation became even the pre-requite of business continuity. Digital transformations transformed the work environment; remote working and home office have become usual. The use of technological innovations and digitalization generate changes in processes [36]. Virtualization of products such as CDs to mp3s, digitalization of information such as catalogues to websites, dematerialization of transport such as flights to teleconferencing, diminishing of warehouses/office spaces and shortening of supply chains are the results of ICTs [34]. ICTs allow access to knowledge and information and communications [37]; however, for this, organizations need to build digital competences besides the technology [38]. In an office environment, the ICTs can clearly positively impact the paper usage from environmental perspective. Digitalization opens up opportunities to move away from and remove paper-intensive processes, even traditionally paper-based and sensitive processes, such as employee contract management, can be managed electronically with the use of digital signatures [36]. The paperless office concept is beneficial, amongst others, from environmental point of view; however, it has its difficulties as well, including initial investment and implementation, training, cultural changes, technical difficulties, adoption of process design, etc. [37]. In addition to the favorable impacts, the risks related to digitalization, such as cybersecurity, are considered as key risk areas companies face these days. The use of technologies should be assessed from security perspective as well [36].

The power consumption—the leading issue with technology development—greatly effects the CO₂ emissions. Moreover, the cooling systems in some data centers also release various chemicals into the atmosphere and contribute to the carbon footprint [39]. The energy that is required to run computers, servers and other electronic devices requires large amounts of natural resources. Standby-mode and Wi-Fi assistive devices are increasingly used; more people are using powerful ICT installed hardware-based devices which ultimately increase energy demand [40].

The Internet of things (IoT) has the potential to be used for environmental protection purposes to monitor, manage and mitigate environmental issues. The applications of the IoT can be used in areas such as air quality monitoring, real-time data collection, water resource management, waste management (including waste sorting and recycling), biodiversity conservation, sustainable agriculture (including smart irrigation systems), etc. [41]. The IoT-based solutions have their challenges such as connectivity; data management, including storage, quality, integration and privacy; sensor calibration and maintenance [41].

Cloud computing is considered as an effective method of saving power and storage and is very power consumptive because servers are always connected to the internet. Cloud computing is a lucrative business. This creates a larger digital CO₂ footprint across the globe [39]. Due to the environmental impact of technologies, movement towards greener digital technologies is needed [42].

The most frequently quoted categorization is called the three-order-effects of ICTs [34]. Most downsides are associated with the first order: direct environmental impacts from ICT infrastructure, e.g., resources consumption, carbon emission during manufacturing and disposal of hardware. Most benefits of ICTs are linked to the second order (indirect) effects created via the ongoing use and application of ICT, e.g., increased efficiency, transparency, speed of transactions, rapid market-clearing, etc. The third order, created by the aggregated effects of large numbers of people using ICT over the medium to long term, involves many uncertainties, effects such as increased consumption due to lower prize online and taking a leisure drive after teleworking [34].

1.5. Related Works, the Research Gap and Hypothesis Design

The literature research proves that ICTs are in support of environmental protection activities in various ways; however, this should be looked together with the potential negative impacts, such as increased energy demand. The environmental impact and footprint of ICTs should be evaluated comprehensively, considering their overall impact throughout their entire lifecycle. ICTs are in operation intensively; their design, usage and post-life should consider environmental impact in the long and short run. ICTs have brought a number of positive environmental advantages; however, at the same time, their negative impacts have to be treated. Research conducted to date in the context of the environmental impact of ICT has focused on a variety of issues. Asongu et al. [43] studied, using the Generalized Method of Moments, how increasing ICT penetration can translate into reducing carbon emissions. Higón et al. [44] studied the non-linear relationship between ICT and carbon emissions globally, according to the EKC framework. The long-term relationship between ICT and carbon emissions in developing countries was analyzed by N’dri et al. [45] using a panel-pooled mean group autoregressive distributive lag. They expressed the ICT variable in the number of mobile and fixed telephone subscriptions per 100 people. Under the same assumption as to how ICT was measured, Opoku-Mensah et al. [46] examined the combined impact of ICT and green institutional quality on the ecological footprint for the Shanghai Cooperation Organization countries using the Driscoll Kraay Standard Estimator Model and the HAC model. Eregha et al. [47] used the augmented mean group estimator to study the impact of ICT on both the ecological footprint and carbon emissions for emerging economies. Alfehaid et al. [48] analyzed the impact of ICT diffusion on carbon emissions and carbon footprint using the dynamic ordinary least squares method. The same relationship was studied by Khan et al. [49], except that they applied the two-step-generalized method of moments. The results of a study by Charfeddine et al. [50] using a panel VAR model indicate that ICT reduces carbon emissions, while it produces mixed results in terms of the environmental footprint.

Tekic and Tekic [51], on the other hand, analyzed using fuzzy-set Qualitative Comparative Analysis how individual ICT elements (Broadband, Cloud, AI, IoT, Technology absorption) affect sustainable development as measured by the SDG Index Score in the areas of social, economic and environment. In their study, Usman et al. [52] proved that ICT has a positive and significant impact on sustainability goals, especially with the support of governance quality and transportation infrastructure; in doing so, they applied the advanced technique of Feasible Generalized Least Square. Chishti et al. [53] showed that green growth depends on the promotion of green ICT, using panel quantile–quantile granger causality testing and method of moments quantile regression. A similar relationship was indicated by Jiang et al. [54] based on the QARDL method.

Despite the multitude and variety of studies on the environmental impact of ICT, they mainly focus on two aspects: carbon emissions and footprint levels, and the level of achievement of the Sustainable Development Goals. There is a lack of research that shows whether the level of digitization is taking into account the environmental impact of ICT services or equipment, and taking measures that affect the paper or energy consumption of ICT equipment. In order to fill this research gap the following research hypotheses were posed:

Hypothesis 1:

There is a positive relationship between the level of measures taken in ICT to protect the environment and the level of digitization of the company.

Hypothesis 2:

An assessment of the level of business involvement in environmental protection in different European Union countries makes it possible to distinguish the degree to which countries think in environmental categories when choosing actions for ICT services and equipment.

2. Materials and Methods

Quantitative analysis will be performed on data describing actions taken by enterprises in the context of the use of ICT equipment for environmental protection and environmental values. The variables and their descriptions are presented in

Table 1. The variables and their descriptions.

Variables	Description of Variables
Enterprise’s involvement in environmental protection	Enterprises with a high digital intensity index, which considered the environmental impact of ICT services or ICT equipment, before selecting them
	Enterprises with a low digital intensity index, which considered the environmental impact of ICT services or ICT equipment, before selecting them
	Enterprises with a very high digital intensity index, which considered the environmental impact of ICT services or ICT equipment, before selecting them
	Enterprises with a very low digital intensity index, which considered the environmental impact of ICT services or ICT equipment, before selecting them
	Enterprises with a high digital intensity index, which apply some measures affecting the paper or energy consumption of the ICT equipment
	Enterprises with a low digital intensity index, which apply some measures affecting the paper or energy consumption of the ICT equipment
	Enterprises with a very high digital intensity index, which apply some measures affecting the paper or energy consumption of the ICT equipment
	Enterprises with a very low digital intensity index, which apply some measures affecting the paper or energy consumption of the ICT equipment
Reducing the amount of paper used	Enterprises applying some measures affecting the amount of paper used for printing and copying
Reducing the amount of energy used	Enterprises applying some measures affecting the energy consumption of the ICT equipment
Taking into account the impact on the natural environment in business activities	The enterprises considered the environmental impact of ICT services or ICT equipment, before selecting them
	When the ICT equipment of the enterprise is no longer used, it is disposed of in electronic waste collection/recycling When the ICT equipment of the enterprise is no longer used, it is kept in the enterprise
	When the ICT equipment of the enterprise is no longer used, it is sold, returned to a leasing enterprise, or donated

.

The data comes from the Eurostat database [55]. Data from 2022 were analyzed for reasons of availability and completeness. All variables were expressed in percentage of enterprises. The study provided data for both the European Union as a whole and the individual 27 countries of the European Union. In addition, the values of the variables observed for the European Union and its individual countries were related to three groups of companies: small, medium and large. The following methods were used to verify the research hypotheses:

• Chi-square test of independence;
• Multivariate data analysis, in particular a linear ordering method;
• A linear correlation coefficient.

The research procedure is shown in Figure 1.

The chi-square test of independence allowed verification of Hypothesis 1. The existence of statistically significant relationships between the level of measures taken in ICT to protect the environment and the level of digitization of the company was checked for each group of companies by size (small, medium and large enterprises). In this way, it was examined whether the size of the company moderates the assumed relationship.

Multidimensional data analysis allowed the estimation of a synthetic measure, indicating business commitment to digitization and environmental protection in different European Union countries. The linear ordering assumes the following nature of variables: (1) stimulants—The enterprises considered the environmental impact of ICT services or ICT equipment, before selecting them; when the ICT equipment of the enterprise is no longer used, it is disposed of in electronic waste collection/recycling; when the ICT equipment of the enterprise is no longer used, it is sold, returned to a leasing enterprise, or donated and (2) destimulants—When the ICT equipment of the enterprise is no longer used, it is kept in the enterprise. In addition, the existence of Pearson’s linear correlation between the synthetic measure of development and the percentage of enterprises with a high and very high digital intensity index was examined. This verified Hypothesis 2 about the variation in the degree to which countries think in environmental categories when choosing ICT services and equipment activities. It was also checked whether this level depends on the level of digitization as measured by the digital intensity index. These analyses were conducted for three groups of companies: small, medium and large.

3. Results

The results of the chi-square tests are shown in

Table 2. The results of a chi-square test (observed test significance level) for the following variables: level of digital intensity index and enterprise’s involvement in environmental protection for small-sized enterprises.

Specification	Enterprises that Considered the Environmental Impact of ICT Services or ICT Equipment, Before Selecting Them	Enterprises that Apply Some Measures Affecting the Paper or Energy Consumption of the ICT Equipment
Countries with a statistically significant chi-square test result	-	Sweden (0.058)
Countries with a statistically insignificant chi-square test result	European Union (0.631), Belgium (0.632), Bulgaria (0.926), Czechia (0.512), Denmark (0.316), Germany (0.463), Estonia (0.779), Ireland (0.409), Greece (0.998), Spain (0.602), France (0.437), Croatia (0.974), Italy (0.336), Cyprus (0.558), Latvia (0.891), Lithuania (0.792), Luxembourg (0.800), Hungary (0.892), Malta (0.462), the Netherlands (0.441), Austria (0.739), Poland (0.774), Portugal (0.615), Romania (0.643), Slovenia (0.578), Slovakia (0.699), Finland (0.318), Sweden (0.107)	European Union (0.569), Belgium (0.479), Bulgaria (0.515), Czechia (0.420), Denmark (0.144), Germany (0.344), Estonia (0.707), Ireland (0.470), Greece (0.158), Spain (0.451), France (0.448), Croatia (0.707), Italy (0.327), Cyprus (0.576), Latvia (0.772), Lithuania (0.807), Luxembourg (0.695), Hungary (0.989), Malta (0.531), the Netherlands (0.288), Austria (0.646), Poland (0.845), Portugal (0.574), Romania (0.723), Slovenia (0.526), Slovakia (0.728), Finland (0.111)

,

Table 3. The results of a chi-square test (observed test significance level) for the following variables: level of digital intensity index and enterprise’s involvement in environmental protection for medium-sized enterprises.

Specification 1	Enterprises that Considered the Environmental Impact of ICT Services or ICT Equipment, Before Selecting Them	Enterprises that Apply Some Measures Affecting the Paper or Energy Consumption of the ICT Equipment
Countries with a statistically significant chi-square test result	Ireland (0.023), Italy (0.058), Cyprus (0.018), Malta (0.051), the Netherlands (0.032), Finland (0.079), Sweden (0.099)	European Union (0.072), Belgium (0.071), Czechia (0.056), Denmark (0.061), Germany (0.046), Ireland (0.020), Spain (0.080), Italy (0.070), Cyprus (0.012), Luxembourg (0.096), Malta (0.073), the Netherlands (0.007), Austria (0.031), Slovenia (0.040), Finland (0.003), Sweden (0.021)
Countries with a statistically insignificant chi-square test result	European Union (0.138), Belgium (0.168), Bulgaria (0.579), Czechia (0.121), Denmark (0.212), Germany (0.148), Estonia (0.110), Greece (0.193), Spain (0.135), France (0.330), Croatia (0.674), Latvia (0.387), Lithuania (0.285), Luxembourg (0.165), Hungary (0.194), Austria (0.104), Poland (0.123), Romania (0.709), Slovenia (0.133), Slovakia (0.404)	Bulgaria (0.674), Estonia (0.114), Greece (0.219), France (0.227), Croatia (0.314), Latvia (0.208), Lithuania (0.165), Hungary (0.375), Poland (0.104), Romania (0.762), Slovakia (0.275)

¹ Portugal was excluded from the analysis due to lack of data.

and

Table 4. The results of a chi-square test (observed test significance level) for the following variables: level of digital intensity index and enterprise’s involvement in environmental protection for large-sized enterprises.

Specification 1	Enterprises that Considered the Environmental Impact of ICT Services or ICT Equipment, Before Selecting Them	Enterprises that Apply Some Measures Affecting the Paper or Energy Consumption of the ICT Equipment
Countries with a statistically significant chi-square test result	European Union (0.078), Bulgaria (0.067), Denmark (0.091), Greece (0.003), Spain (0.021), France (0.066), Cyprus (0.003), Latvia (0.001), Lithuania (0.039), Luxembourg (0.030), Hungary (0.054), Malta (0.065), the Netherlands (0.097), Poland (0.066), Portugal (0.020), Romania (0.009), Slovakia (0.025)	European Union (0.042), Bulgaria (0.051), Czechia (0.099), Denmark (0.054), Germany (0.035), Greece (0.067), Spain (0.011), France (0.029), Cyprus (0.003), Latvia (0.000), Lithuania (0.021), Luxembourg (0.010), Croatia (0.078), Malta (0.045), the Netherlands (0.025), Austria (0.059), Poland (0.044), Portugal (0.024), Romania (0.013), Slovakia (0.008)
Countries with a statistically insignificant chi-square test result	Belgium (0.363), Czechia (0.257), Germany (0.106), Estonia (0.264), Croatia (0.473), Italy (0.106), Austria (0.138), Finland (0.645), Sweden (0.288)	Belgium (0.294), Estonia (0.296), Italy (0.111), Hungary (0.190), Finland (0.451), Sweden (0.248)

¹ Ireland and Slovenia were excluded from the analysis due to lack of data.

.

The chi-square test showed a statistically significant relationship between the level of measures taken in ICT to protect the environment as expressed by the variable Enterprises that considered the environmental impact of ICT services or ICT equipment, before selecting them and the level of digitization of the company:

• In the case of small enterprises, this does not exist in any European Union country;
• For medium-sized companies, it exists in Ireland, Italy, Cyprus, Malta, the Netherlands, Finland, and Sweden;
• For large companies exists in Bulgaria, Denmark, Greece, Spain, France, Cyprus, Latvia, Lithuania, Luxembourg, Hungary, Malta, Poland, Portugal, Romania, and Slovakia.

In addition, a statistically significant relationship between the level of measures taken in ICT to protect the environment expressed by the variable Enterprises that apply some measures affecting the paper or energy consumption of the ICT equipment and the level of digitization of the company:

• For small enterprises, it exists only in Sweden;
• For medium-sized enterprises, exists in Belgium, Czechia, Denmark, Germany, Ireland, Spain, Italy, Cyprus, Luxembourg, Malta, the Netherlands, Austria, Slovenia, Finland, and Sweden,
• For large enterprises, it exists in Bulgaria, Czechia, Denmark, Germany, Greece, Spain, France, Croatia, Cyprus, Latvia, Lithuania, Luxembourg, Malta, the Netherlands, Austria, Poland, Portugal, Romania, and Slovakia.

For companies operating in the European Union, a statistically significant relationship was observed between the level of measures taken in ICT to protect the environment and the level of digitization of the company only for medium and large companies, whereby:

• For large enterprises, the relationship occurred between the level of measures taken in ICT to protect the environment as expressed by both the variable Enterprises that considered the environmental impact of ICT services or ICT equipment, before selecting them as well as the variable Enterprises that apply some measures affecting the paper or energy consumption of the ICT equipment and the level of digitization of the company,
• For medium-sized enterprises, the relationship occurred only between the level of measures taken in ICT to protect the environment expressed by the variable Enterprises that apply some measures affecting the paper or energy consumption of the ICT equipment and the level of digitization of the company.

Table 5. The results of a linear ordering method; a measure of development of the European Union countries according to the level of business commitment to environmental protection in different European Union countries makes it possible to differentiate the degree to which countries are thinking in environmental categories when choosing actions for ICT services and equipment in the year 2022.

Countries	Measure of Development (Rank) for
	Small-Sized Enterprises	Medium-Sized Enterprises	Large-Sized Enterprises
Austria	0.594 (5)	0.638 (3)	0.692 (4)
Belgium	0.523 (11)	0.514 (9)	0.625 (8)
Bulgaria	0.189 (27)	0.182 (26)	0.309 (25)
Croatia	0.431 (21)	0.367 (22)	0.265 (26)
Cyprus	0.522 (12)	0.437 (17)	0.576 (11)
Czechia	0.722 (1)	0.659 (2)	0.653 (5)
Denmark	0.522 (13)	0.587 (6)	0.778 (1)
Estonia	0.547 (9)	0.511 (12)	0.505 (16)
Finland	0.699 (2)	0.663 (1)	0.740 (2)
France	0.437 (20)	0.405 (19)	0.473 (19)
Germany	0.382 (22)	0.383 (21)	0.536 (15)
Greece	0.300 (26)	0.251 (25)	0.174 (27)
Hungary	0.308 (25)	0.263 (24)	0.362 (24)
Ireland	0.309 (24)	0.420 (5)	0.415 (22)
Italy	0.559 (7)	0.571 (18)	0.626 (7)
Latvia	0.349 (23)	0.263 (17)	0.416 (21)
Lithuania	0.491 (16)	0.479 (15)	0.474 (18)
Luxembourg	0.586 (6)	0.514 (10)	0.564 (12)
Malta	0.501 (15)	0.501 (13)	0.494 (17)
Netherlands	0.538 (10)	0.543 (8)	0.546 (14)
Poland	0.462 (17)	0.499 (14)	0.555 (13)
Portugal	0.557 (8)	-	0.443 (20)
Romania	0.505 (14)	0.404 (20)	0.410 (23)
Slovakia	0.620 (4)	0.596 (4)	0.630 (6)
Slovenia	0.650 (3)	0.593 (5)	0.577 (10)
Spain	0.457 (19)	0.451 (16)	0.602 (9)
Sweden	0.460 (18)	0.513 (11)	0.697 (3)

shows the results of linear ordering.

The highest level of business involvement in environmental protection manifested in taking environmentally friendly measures in relation to ICT services and equipment is observed for the following countries and sizes:

• Czechia, Finland, Slovenia, Slovakia, Austria and Luxembourg in the case of small-sized enterprises;
• Finland, Czechia, Austria, Slovakia, Slovenia and Denmark in the case of medium-sized enterprises;
• Denmark, Finland, Sweden, Austria, Czechia and Slovakia in the case of large-sized enterprises.

Thus, it is possible to identify countries that should be the pattern due to measures aimed at pro-environmental behavior in relation to ICT services and equipment; these include the Czech Republic, Finland, Slovakia and Austria, which are at the top of the rankings made for each group of companies. The lowest places in the rankings for each size of enterprise are occupied by countries such as Hungary, Greece and Bulgaria. Germany’s low ranking for small (22nd place) and medium-sized enterprises (21st place) may come as a surprise.

Attention should also be paid to the values of the development measures in each ranking. The largest differences in development measure values are observed for large and small enterprises; thus, these groups of enterprises have the greatest differences in the level of business commitment to environmental protection in the context of thinking in environmental categories when choosing activities for ICT services and equipment. The highest measures of development in these two groups of enterprises take values of 0.722 for small and 0.778 for large enterprises, while the lowest are 0.189 for small and 0.174 for large. Such large differences in the largest and smallest values of the development measure testify to large differences in environmental category thinking with regard to activities undertaken in the context of ICT services and equipment. Although for medium-sized companies the differences between the largest and smallest values of the development measure are smaller; also, for this group of companies, it can be argued that there is a large variation among European Union countries with regard to pro-environmental measures taken in the context of ICT services and equipment.

The linear correlation coefficients for the synthetic measure of development and the digital intensity index were as follows:

• For small enterprises: −0.310;
• For medium enterprises: −0.346;
• For large enterprises: −0.033.

Unfortunately, they turned out to be statistically insignificant at the significance level of 0.05. Thus, there is no statistically significant linear relationship between the synthetic measure of development and the value of the digital intensity index. A high level of digitization of enterprises therefore does not influence environmental category thinking in the context of actions taken with regard to ICT services and equipment.

4. Discussion

The analyses conducted were aimed at verifying two research hypotheses. Hypothesis 1, there is a positive relationship between the level of measures taken in ICT to protect the environment and the level of digitization of the company, was verified using a chi-square test, which showed that statistically significant relationships exist only for medium and large companies. The relationship relates to the implementation of measures to reduce paper and energy consumption of ICT equipment. These measures are part of the general trend of reducing energy consumption by ICT equipment, which is currently estimated at about 10% of global electricity consumption [56]. Given the environmental focus of companies, they are striving to significantly reduce and temporarily mitigate the growth of electricity consumption by ICT equipment. The importance of this is to manage ICT systems in such a way as to meet the high-speed and short-latency needs of enterprises while reducing electricity consumption [57]. The occurrence of a relationship between the level of measures taken in ICT to protect the environment and the level of digitization of the company in medium-sized and large enterprises is due to the fact that these enterprises have a much higher level of digitization [58]. In addition, large companies are more likely to implement ESG reporting [59], especially as the strategic importance of ESG initiatives to ensure long-term corporate success is pointed out [60]. In addition, the need to implement solutions for the consumption of resources related to ICT functionalities in enterprises is justified due to the negative impact of the digitalization of enterprises on corporate carbon emissions [61].

Hypothesis 2, an assessment of the level of business involvement in environmental protection in different European Union countries makes it possible to distinguish the degree to which countries think in environmental categories when choosing actions for ICT services and equipment, was verified using the linear ordering method. Countries such as Czechia, Finland, Slovakia and Austria have been identified as pattern due to their pro-prosperity-oriented efforts with regard to ICT services and equipment. Research by [62] shows that Slovenia and Austria are the leaders in enterprise ICT applications, while Germany, the Czech Republic and Slovakia are characterized by relatively extensive use of ICT for e-business and e-commerce.

In addition, it was shown that there was no statistically significant relationship between the high level of digitization of enterprises and thinking in environmental categories in the context of actions taken with regard to ICT services and equipment, these results contradict the research of [63], which showed that digitization is associated with a higher propensity of enterprises to implement ICT-related sustainability practices. These discrepancies may be due to the context of the studies conducted: most of the research focuses on the use of ICT to achieve environmental goals [27,64,65,66], rather than on the environmental impact of using ICT equipment. Hence, the literature points to the continuing need for in-depth research on the environmental impact of ICT [67]. At least three reasons are cited for this: (1) digital tools and applications facilitate the transition to sustainable production and consumption patterns; (2) digital applications may conflict with sustainability goals due to energy and resource intensity; and (3) digitization brings challenges regarding, for example, privacy, information asymmetry, and data security [68]. Therefore, the direction of future research will be an in-depth analysis of the impact of pro-environmental measures for ICT services and equipment on the efficiency of business operations. This is because pro-enterprises will be inclined to take such measures when they see in them a significant benefit to efficiency and productivity.

5. Conclusions

In the first stage, the relationship between the level of measures taken in ICT to protect the environment and the level of digitization of the company was determined. It was found that statistically significant relationships were observed only for medium-sized and large enterprises, with the relationship for medium-sized enterprises occurring only between the variable Enterprises that apply some measures affecting the paper or energy consumption of the ICT equipment and the level of digital intensity index. In contrast, for large enterprises, the relationship occurred both between the variable Enterprises that considered the environmental impact of ICT services or ICT equipment, before selecting them and the level of the digital intensity index, and the variable Enterprises that apply some measures affecting the paper or energy consumption of the ICT equipment and the level of the digital intensity index.

In the next step of the research, it was shown that the countries of the European Union are characterized by a wide variation in terms of pro-environmental measures taken by companies with regard to ICT services and equipment. These measures include the following: applying measures affecting the amount of paper used for printing and copying; applying measures affecting the energy consumption of the ICT equipment; considering the environmental impact of ICT services or ICT equipment, before selecting them; disposal of electronic waste collection/recycling of ICT equipment when it is no longer in use; leaving ICT equipment at the enterprise when it is no longer in use; and selling, returning to a leasing company or donating ICT equipment when it is no longer in use. Based on linear ordering, the following model countries were identified in terms of measures taken to target pro-environmental behavior with regard to ICT services and equipment: Czechia, Finland, Slovakia and Austria. However, it was noted that there was no statistically significant linear relationship between the high level of digitization of businesses and environmental category thinking in terms of actions taken with regard to ICT services and equipment.

Several limitations can be pointed out in connection with the study. First of all, only one year, 2022, was analyzed, due to the availability of data. In addition, the analysis of dependencies was carried out only using the chi-square test, omitting the measurement of the strength of dependencies and the identification of cause-and-effect relationships that could be indicated on the basis of an econometric model. Limitations can also be pointed out with regard to linear ordering: the measure of development was estimated under the assumption that all variables have the same weights. It is also a limitation to carry out analyses on the basis of the available data in the Eurostat database.

The direction of future research should therefore be to study the dynamics of changes in the environmental impact of ICT in enterprises on the basis of dynamic panel models. There is also a need for a study in which the statistical unit will be enterprises, since it is enterprises that should build their strategies taking into account the need to reduce the negative impact of ICT on the environment. A particularly important issue that sets the direction of future research is the learning effect between enterprises, which is related to the exchange of knowledge and good practices in measures to reduce the negative impact of ICT on the environment. Research would also be worthwhile to target the value of data, which would help conduct analyses related to the environmental effectiveness of ICT use in enterprises.