Research on the Relationship between CO₂ Emissions, Road Transport, Economic Growth and Energy Consumption on the Example of the Visegrad Group Countries

Abstract

This study examines the cointegration and short- and long-run causal relationships between economic growth and CO₂ emissions, fuel consumption and the amount of freight transported by road transport in the Visegrad Group countries between 1990 and 2019. The autoregressive distributed lag (ARDL) bound testing methodology was used to develop the results for each tested dependent variable. The results confirmed a long-run cointegration between economic growth and fuel consumption by road transport in the Czech Republic and Slovakia. In addition, a long-run relationship between economic growth and freight volume and CO₂ emissions by transport was demonstrated for Slovakia. For Poland, there was only a long-run relationship between energy consumption by road freight transport and CO₂ emissions. In the short run, relationships were shown between the volume of freight transported by the road freight transport sector and CO₂ emissions for Hungary and Poland. Short-run relationships between economic growth and the volume of freight transported in Poland, the Czech Republic and Hungary are also interesting observations. The obtained results expand the information resources needed to make decisions on the direction of change in road transport. Additionally, the results can be used to prepare a proper environmental policy for road freight transport within the framework of the European Green Deal.

1. Introduction

The Visegrád Group (V4) is a cultural, economic and political alliance of four central European countries: the Czech Republic, Hungary, Poland and Slovakia. It was established on 15 February 1991, and now the countries in the group belong to the most dynamically developing countries in the region of Central and Eastern Europe (CEE).

Road freight transport in the Visegrad countries has been an important determinant of their economic growth rates for nearly two decades. Road transport services in Poland, the Czech Republic, Slovakia and Hungary (domestic and international) are a crucial element of the production and supply chain for many industrial and commercial sectors. After joining the European Union (in 2004), the countries mentioned above began to play an important role in the EU road freight market, owing to the competitiveness of their transport companies [1] and have contributed to the development of trade in the community (especially since 2008, following the onset of the global crisis). In the case of Poland, the leader of European international transport in all categories (bilateral transport, cross-trade and cabotage), experts estimate that road freight transport contributes 50% of the country’s GDP [2]. Road freight transport in the Visegrad Group countries is far more important to their economies than the direct contribution of this sector to corporate revenues. It manifests itself in foreign trade turnover, the impact on the state of public finances (tax revenues), and the offering of attractive jobs in terms of wages (in international transport).

At the same time, it should be noted that the intensively developing road transport and the growing number of passenger cars and trucks in highly developed countries have constantly contributed to the increase in emissions of harmful substances into the atmosphere [3]. The same phenomenon occurred in the Visegrad Group countries, although later—only after systemic changes in the 1990s. They have contributed to the negative consequences of the development of road transport, including increased emissions of greenhouse gases and other substances harmful to the environment.

The issue of road transport emissions (including freight) is not only an environmental problem for the V4 countries, but also an economic problem. It is indicated by studies on the relationship between the value of the gross domestic product and the emissions of harmful substances into the atmosphere conducted in various countries. Oreshnyak established a correlation between the rate of economic growth per capita at purchasing power parity (GDP per capita PPP) and changes in the volume of CO₂ emissions, as well as the changes in CO₂ emissions and the share of individual sectors in the economy [4]. Shpak et al. [5] analysed in the long term (1970–2020) the relationship between macroeconomic indicators (GDP, exports, imports, inflation rate and unemployment) and the volume of CO₂ emissions in the economies of EU regions. Their research shows that the emissions of this greenhouse gas depend on macroeconomic indicators in these regions. The correlation between CO₂ emissions and exports is high and negative, as is the correlation between CO₂ emissions and GDP and between CO₂ emissions and imports. A strong and positive relationship was noted in the correlation between the volume of CO₂ emissions and the inflation rate. On the other hand, the relationship between CO₂ emissions and unemployment rates in EU regions has shown a low but positive correlation.

Zubedi and his team [6] examined the relationship between CO₂ emissions and factors such as energy consumption, economic growth and foreign direct investment in the long term (1979–2019) for a specific area of the Belt and Road Initiative and Pakistan using cointegration. The findings indicate the presence of a long-run relationship between CO₂ emissions and economic growth, trade development and foreign direct investment. They will also reveal the short-run relationship between energy consumption and CO₂ emissions.

The literature review shows that there are no studies devoted to road freight transport in the Visegrad Group countries, which would indicate short-run and long-run relationships between economic growth, energy (fuel) consumption and CO₂ emissions in the countries of this group. The authors identified a research gap and conducted a study in this area. Their goal is to study the cointegration and short- and long-run causal relationships between economic growth and CO₂ emissions, fuel consumption and the amount of freight transported by road transport in the Visegrad Group countries.

The ARDL Bound Testing methodology was used for each of the dependent variables tested to verify these hypotheses. The analyses were performed for the years 1990–2019, with n = 30 for Poland and Hungary and n = 27 for the Czech Republic and Slovakia. The length of the study periods was determined by the systemic transformation periods in the 1990s in the V4 countries. The article is further structured as follows: Section 2 provides an overview of the literature. Section 3 presents the research methodology and the data sources used. Section 4 contains the results obtained from the unit root test and ARDL bound test, as well as the estimates of the short- and long-run models. In Section 5, the authors discussed the research results, allowing the final Section 6 of the paper to present conclusions and recommendations for transportation policy in the Visegrad Group countries.

2. Literature Review

In the 1990s, Poland, the Czech Republic, Slovakia and Hungary, together with other post-communist countries, began the transformation of their political and socio-economic systems. In February 1991, they set up an informal regional cooperation forum called the Visegrad Group (V4) aimed at European integration. One of the main objectives of each member state of the group was to redirect the foreign trade of these countries to Western Europe. During the previous political and economic system, foreign exchange between these countries took place primarily within the framework of the Council for Mutual Economic Assistance (Comecon), which brought together the countries of the so-called people’s democracies of Central and Eastern Europe [7,8,9,10,11]. In the following decades, the Visegrad Group countries moved from a command-and-quota economy to a market economy. They were supported by activities such as signing the CEFTA Agreement in December 1992. The agreement has enabled the creation of a free trade area between the countries of Central Europe. Another significant milestone in the development of the member states of this grouping and the entire Visegrad Group was the integration with the European Union in 2004 [12]. As indicated by Łącka et al. [13] the processes of transformation, geopolitical changes, the initiation of regional cooperation and then entry into the European Union enabled the V4 countries to expand into new markets and benefit from international exchange. In the years 1991–2019, the GDP of the entire Visegrad Group increased by 155%. The experts of the Polish Economic Institute found that between 1995 and 2019, the value of exports of goods in the V4 countries expanded by more than 19 times, imports by 16 times and investment in fixed assets increased three times faster than in the EU-15 [14]. It points to a high economic growth rate, although periodic fluctuations in this indicator were observed during this period [15]. The region’s high investment attractiveness enabled the V4 economies to integrate into global value chains [16]. However, this success has been achieved by all Visegrad member states with increasing energy consumption and rising CO₂ emission rates [17]. Data shows that the emissivity of the economies of this grouping was 35% higher in the years 1991–2021 than in the western countries of the European Union. [14].

Road freight transport in the member countries of the Visegrad Group considerably impacted achieving high economic growth. Its intensive development over the past thirty years and its widespread use for freight transport in V4 resulted from the availability of transport infrastructure and lower costs compared to other modes of transportation.

The road transport sector (transport of people and freight) is important for the socio-economic development of the countries in the grouping and their economic growth. It also affects the environment, produces emissions, noise and vibrations, and poses risks to human health and safety. Kučera and Viteková studied this problem in the case of road freight transport in their comparison of the effectiveness of road freight tolls in the Visegrad countries in the context of sustainable regional development [18].

The studies by Włodarczyk and Mesjasz-Lech [19] determined the Granger causal relationship between the energy use of road transport, road length, the number of cars and emissions of nitrogen oxides (NOx) emitted from road transport activities in the Visegrad Group countries between 1991 and 2015. Unfortunately, these studies referred to all road transport, so it is difficult to determine the impact of freight transport on nitrogen oxide emissions. The authors did not analyse the impact of this form of transport on CO₂ emissions.

Kisielinska et al. conducted a study assessing and comparing European countries. In that study were check the terms of using renewable power sources in road transport [20]. The result of obtaining research was the assessment of energy consumption of different energy sources in road transport in EU countries between 2008–2019. The results and conclusions of the research include information on the use of renewable energy sources in the form of blended biodiesel and blended biogasoline by road transport in the Visegrad Group countries (Poland, the Czech Republic, Slovakia and Hungary). However, these analyses concerned all road transport. Based on them, it is difficult to assess shifts in the use of renewable energy resources by road freight transport (trucks) in the V4 countries.

Wawrzyniak’s research [21] contributed to the study of factors influencing the changes in CO₂ emissions in the V4 countries in the years 1993–2016. The author used the LMDI (Logarithmic Mean Divisia Index) decomposition method and Kay’s identity for the analysis. It allowed her to identify the factors that contributed the most to the changes in CO₂ emissions in the studied countries. These included energy intensity and economic growth expressed in GDP per capita. The reduction in emissivity of the Visegrad economies was primarily the result of improvements in energy efficiency and, to a lesser extent, changes in the energy mix (growth of renewable energy sources).

The literature review reveals a lack of studies devoted to road freight transport in the Visegrad countries, which would show the relationship between economic growth and energy consumption, CO₂ emissions and the volume of road freight transport. Such research would allow, on the one hand, an understanding of the cause-and-effect relationships between these phenomena and, on the other hand, would make it possible to present the impact of this form of transportation on the emission of the most important greenhouse gas, which is a major contributor to global warming.

Wanag et al. stated that, “Carbon dioxide is one of the most important of Earth’s greenhouse gases, and its emission is the primary driver of global climate change. It is widely known that in order to avoid the worst impacts of climate change, the world needs to reduce CO₂ emissions or find a new and efficient method for CO₂ capture or transformation into useful chemical products” [22]. Road transport is responsible for relatively large emissions of greenhouse gases. Global freight and passenger transport accounts for 20% of global greenhouse gas emissions, including carbon dioxide (CO₂). According to Dong and others [23], “in 2019, CO₂ emissions generated by the global transport sector accounted for approximately 1/4 (24.2%) of total global CO₂ emissions, with emissions of approximately 8 million tonnes, of which emissions in the road transportation sector were approximately 6.5 billion tonnes, accounting for 81% of the whole transportation sector.”

However, according to data from the European Environment Agency (EEA), in 2020, approximately 25% of the total CO₂ emissions in the European Union came from the transport sector. Among the different modes of transport, road transport accounts for the largest share of CO₂ emissions. It accounts for as much as 71.7% of the EU’s total greenhouse gas emissions. The same data show that passenger cars account for the largest share of CO₂ emissions in the community. It accounts for 60.6% of total road transport carbon dioxide emissions. The percentage of trucks engaged in road freight transport in the total CO₂ emissions of this form of transport is 27.1%. It results mainly from the use of fossil fuels (diesel and petrol) by vehicles [24].

The lower share of heavy goods vehicles in CO₂ emissions compared to passenger cars is due to the fact that regulatory changes in many developed countries, which have been progressing for several years, the growing competitiveness of green propulsion systems (which enable zero emissions), and the growing awareness of the consequences of climate change are fostering, in Europe, the US and China, an increase in the number of zero-emission (electric) trucks in carriers’ fleets. One such regulation is the provision resulting from the assumptions of the European Green Deal, announced in December 2019. According to the newly adopted European Union, Regulation (EU) 2019/1242, CO₂ emission standards for trucks will be very restrictive in 2030. The European Commission assumed that all new vehicles launched in 2025 would have to have 15% fewer CO₂ emissions than the current standards. However, in 2030, emissions from new trucks should fall by as much as 30%.

In the case of China, such regulations are the Action Plan for Carbon Dioxide Peaking Before 2030, the NEV Industrial Development Plan 2021–2035 and the Technology Roadmap for Energy-Saving and New Energy Vehicles 2.0—the country’s road transport emissions could peak before 2030 and petroleum consumption before 2027 [25]. They were forced by increasing CO₂ emissions from road freight transport and the negative consequences of this phenomenon caused by the intense pace of economic growth [26].

In 2022, in the United States, for the first time in two decades, President J. Biden’s administration proposed detailed new limits on pollution generated by buses, vans, tractor-trailers and other heavy trucks to reduce nitrogen dioxide emissions by 90% by 2031. Its emissions contribute to lung cancer, heart disease and premature deaths. Together with these provisions, the Environmental Protection Agency announced plans for a slight tightening of allowable truck CO₂ emissions. The new rules on nitrogen oxide pollution would apply to heavy goods vehicles from 2027, while the rules on carbon dioxide pollution would apply to heavy goods vehicles from 2024 [27].

In light of the regulations being undertaken by states to reduce greenhouse gas emissions in the coming years, truck manufacturers are trying to prepare in advance for the challenges of zero-emission transportation, and trucking companies are investing in new, greener vehicles. The Transport and Environment report shows that heavy goods vehicle manufacturers, thanks to their technologies (changes in aerodynamics and fuel efficiency and the use of information and communication technologies), will be able to meet CO₂ reduction targets starting sooner than indicated in EU documents [28].

In the European Union, carbon dioxide emissions from road transport decreased by 5.5% between 2010 and 2020. It resulted in an increase in biofuel consumption. Eleven EU member states managed to reduce CO₂ emissions intensity by more than 6%. Among them were also the Visegrad Group countries [20]. These actions, while positive, show that the decarbonisation of the road transport sector in EU countries is taking place too slowly in relation to the objectives of the European Green Deal and the goals for member countries to achieve zero emissions in 2035 and climate neutrality in 2050.

Studies using ARDL models for the relationship between CO₂ emissions, energy consumption and economic growth and other variables such as urbanisation and foreign exchange for various countries, including China, Pakistan, Turkey, Singapore, the Danube Region, SAARC and ASEAN, among others, have also been conducted to date, revealing both short- and long-run relationships between the variables [29,30,31,32,33,34,35,36,37]. Some researchers have also researched the use of ARDL models in the search for links between transport in the broad sense and CO₂ emissions and energy consumption [38,39,40,41,42].

The literature also includes studies dedicated to finding factors influencing CO₂ emissions by road transport in China [43] and in the region of North America [3]. An interesting study on the possibilities of achieving climate neutrality for China [23] by reducing CO₂ emissions from the road transport sector was investigated by Dong et al. The authors introduced several scenarios for developing low-carbon road transport using specific policies to ensure China achieves its carbon neutrality goal.

3. Materials and Methods

3.1. Data

The article analyses the relationship between economic growth (GDP per capita), energy consumption by the road transport sector (EC), CO₂ emissions by the road freight transport sector (CO₂) and the volume of freight transported via road transport (FRT).

The time series differs from country to country. In the case of Poland and Hungary, data were obtained for the period 1990–2019, while for the Czech Republic and Slovakia, they were obtained for the years 1993–2019. It is due to the political and economic transition and the breakup of Czechoslovakia in the early 1990s. A more comprehensive range of data for Poland and Hungary could not be used either due to the centrally planned economic systems in these countries until 1990 and the lack of reliable statistical data. In addition, using data from the communist system could have destroyed the results obtained since the economic conditions were completely different from those existing in the market system.

Thus, the study used the broadest possible and available range of statistical information. The data were obtained from the databases of the Organisation for Economic Cooperation and Development (OECD) and the United Nations Framework Convention on Climate Change (UNFCCC). The information used is publicly available, and the target set of variables with sources is shown in

Table 1. Type, measurement units and sources of data used in the research.

Variable	Names of Variables	Unit	Source
EC	Oil products’ final consumption in transport	(TJ—Terajoule)	OECD
CO2	CO2 Emission from road transport	(kt—Kilotonnes)	UNFCCC
GDP	Gross domestic product per capita	(US$ constant 2015)	OECD
FRT	Volume of the freight road transport	(Mtkm—Million tonne-kilometres)	OECD

and dataset is in Table S1.

3.2. Methodology

The proposed study draws its methodological basis from previously conducted analyses on the relationships between CO₂, energy consumption and GDP [44,45]. Three hypotheses were taken into consideration in the design of the research:

Hypothesis 1 (H1).

In the Visegrad Group countries, there is a long-run relationship between economic growth (GDP) and CO₂ emissions from the road freight transport sector.

Hypothesis 2 (H2).

In the Visegrad Group countries, there is a long-run relationship between energy consumption by transport and economic growth (GDP).

Hypothesis 3 (H3).

In the Visegrad Group countries, there is a short-run relationship between economic growth (GDP) and the number and weight of loads transported by road transport.

To achieve the objectives of the study and verify the assumed hypotheses, the authors decided to modify the scope of the analysed variables to estimate the links between the economy, road freight transport and the effects of the indicated activity in the form of oil consumption and environmental pollution. At the same time, the methodology refers to earlier studies carried out in other countries for similar types of variables [46,47]. In order to stabilise the time series, the variables were used in the form of natural logarithms [48]. The long-run form of the models studied, following the theoretical basis and conclusions of other studies, is presented below [38,49,50]:

$$\begin{gathered} LGD{P}_t={\beta }_1+{\beta }_{2}LE{C}_t+{\beta }_{3}LCO2{ }_t+{\beta }_{4}LFR{T}_t+u_t \\ LE{C}_t={\beta }_1+{\beta }_{2}GD{P}_t+{\beta }_{3}LCO2{ }_t+{\beta }_{4}LFR{T}_t+u_t \\ LCO{2}_t={\beta }_1+{\beta }_{2}LE{C}_t+{\beta }_{3}LGDP{ }_t+{\beta }_{4}LFR{T}_t+u_t \\ LFR{T}_t={\beta }_1+{\beta }_{2}LE{C}_t+{\beta }_{3}LCO2{ }_t+{\beta }_{4}GD{P}_t+u_t \end{gathered}$$

(1)

where β are coefficients and u_t is the error value.

The study used the methodology proposed by Pesaran [51,52] and developed by Kripfganz and Schneider [53]. The choice of the ARD-bound test procedure resulted from the fact that it allowed us to find a relationship regardless of whether the variables were integrated at the I(0) or I(1) level. Since the time series were relatively short, ARDL models were used. It was the best choice for the variables analysed to determine short-run and long-run relationships.

The design of the research procedure followed the specified methodology and the algorithm provided by the manufacturer of Eviews software. The research process consisted of the following steps, based on the literature [53,54,55,56,57,58,59,60]:

–

Conducting the augmented Dickey and Fuller unit root test (ADF) and the Kwiatkowski–Phillips–Schmidt–Shin test (KPSS) to determine the degree of integration of variables;

–

determining and specifying in terms of the number of lags in the ARDL model using the Akaike information criterion (AIC) and Schwarz (SC) criteria to select the optimal model;

–

ARDL model estimation;

–

performing diagnostics on the estimated model;

–

performing ARDL bound testing;

–

determining the nature of the relationship and a possible estimation of the error correction model (ECM)—estimate speed of adjustment;

–

conducting tests for diagnostics and the stability of estimated models (including the analysis of the CUSUM and CUSUM² graphs).

The study used the ARDL model to find cointegration relationships between the variables under study, such as economic growth, volume and amount of freight transported, CO₂ emissions and energy consumption of road transport. The ARDL model used (p, q1, q2 and q3) is expressed by the following formula [37,56]:

$$yt=c_0+{{\displaystyle \sum }}_{i=1}^p{\phi }_i{y}_{t-1}+{{\displaystyle \sum }}_{i=0}^{q1}{\beta }_{1,i}{x}_{1, t-i}+{{\displaystyle \sum }}_{i=0}^{q2}{\beta }_{2,i}{x}_{2, t-i}+{{\displaystyle \sum }}_{i=0}^{q3}{\beta }_{3i}{x}_{3, t-i}+ut,$$

(2)

where p ≥ 1, q1 ≥ 0, q2 ≥ 0 and q3 ≥ 0 are the optimal lags for the dependent variable y and independent variables x₁, x₂ and x₃; φ and β are coefficients, c₀ is a constant and ut is the error term (ECT).

The following model forms were derived for the variables tested using the specific form of the ARDL models:

$$\begin{gathered} LGDP=c_0+{\displaystyle \sum }_{i=1}^p{\phi }_{i}LGD{P}_{t-1}+{\displaystyle \sum }_{i=0}^{q1}{\beta }_{1,i}LE{C}_{ t-i}+{\displaystyle \sum }_{i=0}^{q2}{\beta }_{2,i}LCO{2}_{ t-i}+{\displaystyle \sum }_{i=0}^{q3}{\beta }_{3i}LFR{T}_{ t-i}+ut \\ LE{C}_t=c_0+{\displaystyle \sum }_{i=1}^p{\phi }_{i}LE{C}_{t-1}+{\displaystyle \sum }_{i=0}^{q1}{\beta }_{1,i}LCO{2}_{ t-i}+{\displaystyle \sum }_{i=0}^{q2}{\beta }_{2,i}LGD{P}_{ t-i}+{\displaystyle \sum }_{i=0}^{q3}{\beta }_{3i}LFR{T}_{ t-i}+ut \\ LCO{2}_t=c_0+{\displaystyle \sum }_{i=1}^p{\phi }_{i}LCO{2}_{t-1}+{\displaystyle \sum }_{i=0}^{q1}{\beta }_{1,i}LE{C}_{ t-i}+{\displaystyle \sum }_{i=0}^{q2}{\beta }_{2,i}LGD{P}_{ t-i}+{\displaystyle \sum }_{i=0}^{q3}{\beta }_{3i}LFR{T}_{ t-i}+ut \\ LFRT=c_0+{{\displaystyle \sum }}_{i=1}^p{\phi }_{i}FR{T}_{t-1}+{{\displaystyle \sum }}_{i=0}^{q1}{\beta }_{1,i}LE{C}_{ t-i}+{{\displaystyle \sum }}_{i=0}^{q2}{\beta }_{2,i}LCO{2}_{ t-i}+{{\displaystyle \sum }}_{i=0}^{q3}{\beta }_{3i}LGD{P}_{ t-i}+ut \end{gathered}$$

(3)

Using the ARDL bound test, an F statistic was calculated to test the hypothesis that cointegration exists. Due to the smaller data sample, a T-statistic test was performed for the Czech Republic and Slovakia. In the case of confirmation of the cointegration, according to the methodology, an error correction model (ECM) was estimated and is expressed as follows [61,62]:

$$\Delta Y_t=c0-\lambda \left(y_{t-1}-{\theta }_{x1,t}-{\theta }_{x2,t}-{\theta }_{x3,t}\right)+{{\displaystyle \sum }}_{i=1}^{p-1}{\psi }_i\Delta y_{t-i}+{{\displaystyle \sum }}_{i=0}^{q_1-1}{\gamma }_{1,i}\Delta x_{1, t-i}+{{\displaystyle \sum }}_{i=0}^{q2-1}{\gamma }_{2,i}\Delta x_{2 t-i}+{{\displaystyle \sum }}_{i=0}^{q3-1}{\gamma }_{3,i}\Delta x_{3, t-i}+ut$$

(4)

where

• θ is the long-run parameter,
• ψ,γ are short-run parameters,
• λ is the speed adjustment parameter (between 0 and −1 and statistically significant).

In order to optimally select the number of lags and model specifications, the study used both the AIC and SC selection criteria. All models have been diagnosed for homoskedasticity, serial correlation and residue normality [49]. In addition, stability was further investigated using the Ramsey RESET test [63]. All tests were performed at a statistical significance level of 5%.

4. Results

4.1. Descriptive Statistics

Descriptive statistics for the time series under study are presented in

Table 2. Descriptive statistics.

Variable	Country	Mean	Median	Max	Min	Std. Dev.	n
EC	Czech Republic	202,677.40	221,354.00	256,788.00	102,102.00	49,044.76	27
	Hungary	141,816.60	143,445.00	195,013.00	97,404.00	29,565.94	30
	Poland	531,418.20	529,676.00	881,205.00	288,267.00	174,494.40	30
	Slovakia	71,760.07	71,058.00	102,307.00	40,772.00	17,052.10	27
GDP	Czech Republic	28,402.17	30,241.13	38,419.16	19,456.20	5634.99	27
	Hungary	22,588.19	23,776.73	31,696.39	15,538.70	4572.25	30
	Poland	20,030.20	19,439.54	31,717.59	10,796.50	6055.08	30
	Slovakia	22,458.49	22,653.35	33,317.17	12,387.68	6641.59	27
CO2	Czech Republic	13.94	15.13	20.80	7.52	4.01	27
	Hungary	21.25	23.04	32.62	12.01	6.36	30
	Poland	12.53	12.16	21.78	5.89	4.47	30
	Slovakia	18.10	17.96	28.94	9.14	5.91	27
FRT	Czech Republic	42,982.96	44,954.00	58,715.00	13,014.00	10,443.33	27
	Hungary	26,170.41	30,495.00	40,002.00	11,744.00	11,038.26	30
	Poland	161,055.00	136,490.00	348,952.00	40,744.00	99,482.96	30
	Slovakia	23,316.70	22,550.00	36,139.00	6833.00	8543.79	27

Source: OECD and UNFCCC database.

, where the mean, median, minimum and maximum value and standard deviation were calculated for each country, and the number of observations is given. Eviews 12 software was used to process and analyse data and develop models.

The characteristics of the time series of energy consumption by the road transport sector are shown in Figure 1. During the period considered, road transport in Poland (881,205 Tj) and the Czech Republic (256,788 Tj) used the most energy, while in the Slovak Republic, it consumed the least (101,505 Tj). In all countries, during the period under review, there was an increase in energy consumption by the road freight transport sector. The analysis showed that in the years 1990–2019, the most significant increase in energy consumption occurred in Poland (224.4%), and the smallest was observed in Hungary (64.6%).

Among the countries surveyed, the most considerable increase in gross domestic product per capita during the period under review was recorded in Poland (up 193.3%) and Slovakia (up 211.92%). The GDP per capita of Hungary and the Czech Republic increased by 97.2% and 73.8%, respectively, during the period in question. The selected countries were characterised by stable gross domestic product growth per capita. The most sustainable GDP growth has been recorded in Poland and Slovakia. In the case of Poland, the average annual growth rate was 4.14% per year, while in Slovakia, the rate reached 3.82%. The lowest GDP growth occurred in the Czech Republic, where the average annual pace was 1.99% (Figure 2).

Another variable analysed was CO₂ emissions from the road freight transport sector (in kilotonnes), the time series of which is illustrated in Figure 3. Another variable analysed was CO₂ emissions from the road freight transport sector (in kilotonnes), the time series of which is illustrated in Figure 3. In the group of surveyed countries, the largest emitters of CO₂ by road transport (2019) were Poland (64,096.02 kt) and the Czech Republic (18,485.37 kt). The lowest CO₂ emissions in the road transport sector among the V4 countries were recorded in Slovakia (7549.99 kt). At the same time, studies have shown a dynamic increase in CO₂ emissions by the road transport sector over the period considered in all countries concerned. Per capita, CO₂ emissions increased to the greatest extent in Poland (an increase of 247.6%), while they increased the least in Slovakia (a rise of 67.66%).

In terms of the volume of freight transported, road carriers from Poland transported the most (2020 or, rather, 2019) (348,952 Mtkm), while those from Slovakia transported the least (33,941.00 Mtkm). Compared to the initial period of the study, the amount of freight transported increased the most in Poland (766.04% increase) and the least in Hungary (143.76% increase)—Figure 4.

4.2. Results for Unit Root Test and ARDL Bound Testing for Cointegration

The essential requirement for using the proposed research methodology was to conduct a unit root test to determine the level at which the studied series were stationary. The ADF (Dickey–Fuller) test results for each level and the first differences are shown in

Table 3. Unit root test results (the augmented Dickey–Fuller test (ADF)).

Country	Variable	Original Variables		First Differences
		Intercept	Intercept with Trend	Intercept	Intercept with Trend
Poland	EC	−0.63	−4.21 **	−4.25 ***	−4.16 **
	GDP	−0.27	−2.22	−3.22 **	−4.91 ***
	CO2	1.04	−3.38 *	−3.26 **	−3.44 *
	FRT	−0.05	−1.97	−5.27 ***	−5.17 ***
Hungary	EC	−0.76	−3.47 *	−7.54 ***	−7.4 ***
	GDP	0.47	−3.66 **	−3.95 ***	−3.99 **
	CO2	0.60	−2.84	−5.67 ***	−5.69 ***
	FRT	−2.85	−3.87 **	−10.36 ***	−9.82 ***
Czech Republic	EC	−1.64	−1.42	−7.75 ***	−8.94 ***
	GDP	−0.81	−2.38	−5.66 ***	−5.28 ***
	CO2	0.26	−2.40	−4.87 ***	−4.81 ***
	FRT	−3.08	−1.36	−4.92 ***	−6.14 ***
Slovakia	EC	−0.21	−5.67 ***	−4.96 ***	−4.85 ***
	GDP	−0.16	−1.99	−4.49 ***	−4.42 ***
	CO2	1.73	−3.49 *	−7.81 ***	−3.87 **
	FRT	−1.40	−5.65 ***	−3.75 **	−4.18 **

An asterisk indicates the significance of the coefficients in the tables, where *, ** and *** denote 5%, 1% and 0.1% significance levels, respectively.

. All tested series for the tested variables are stationary at the level. The variable EC is fixed at the level for Poland, Hungary and Slovakia. In contrast, the CO₂ variable is stationary for Poland and Slovakia. Additionally, the FRT variable is stationary at the same level for Hungary and Slovakia. The variables reject the null hypothesis that the series has a unit root on the first difference [54]. Therefore, it is possible to conclude they are all stationary at either I (0) or I (1), which is consistent with the premise of the ARDL bound test.

In addition, to check whether the variables meet the ARDL assumptions, it was checked whether they were not integrated at the second difference. The analysis showed that no variables were integrated at level I(2).

In addition, given that the time series under study were relatively short, an additional test was conducted using the KPSS test [56]. If the result of the two tests is the same, it confirms the stationary or nonstationary nature of the time series. At the significance level of 5%, the results obtained in the ADF test were confirmed with the KPSS test. At the significance level of 5%, the results obtained in the ADF test were confirmed with the KPSS test (

Table 4. Unit root test results (KPSS).

Country	Variable	Original Variables		First Differences
		Intercept	Intercept with Trend	Intercept	Intercept with Trend
Poland	EC	0.70 **	0.06	0.05	0.04
	GDP	0.70 **	0.11	0.12	0.10
	CO2	0.71 **	0.10	0.20	0.07
	FRT	0.70 **	0.10	0.13	0.13 *
Hungary	EC	0.60 **	0.09	0.12	0.09
	GDP	0.68 **	0.09	0.09	0.09
	CO2	0.69 **	0.08	0.16	0.07
	FRT	0.67 **	0.10	0.23	0.08
Czech Republic	EC	0.62 **	0.17 **	0.30	0.12
	GDP	0.68 **	0.08	0.26	0.16 **
	CO2	0.70 **	0.07	0.16	0.13 *
	FRT	0.52 **	0.17 **	0.44 *	0.11
Slovakia	EC	0.67 **	0.06	0.15	0.08
	GDP	0.69 **	0.11	0.13	0.13 *
	CO2	0.70 **	0.16 **	0.23	0.08
	FRT	0.65 **	0.10	0.19	0.08

An asterisk indicates the significance of the coefficients in the tables, where *, ** and *** denote 5%, 1% and 0.1% significance levels, respectively.

). In view of the above information, the possibility of using the ARDL bound testing method to demonstrate cointegration was confirmed.

Table 5. The ARDL bound testing for integration with critical values.

Country	Dep. Variable	F-Statistics	I (0)	I (1)	Cointegration	Model
Poland	EC	8.76	3.23	4.35	YES	ECM
	GDP	1.60	3.23	4.35	NO	short-run model
	CO2	2.98	2.79	3.67	YES	short-run model
	FRT	3.52	3.23	4.35	YES	short-run model
Hungary	EC	3.39	3.23	4.35	NO	short-run model
	GDP	2.27	3.23	4.35	NO	short-run model
	CO2	2.02	2.79	3.67	NO	short-run model
	FRT	1.36	2.79	3.67	NO	short-run model
Czech Republic	EC	29.32	3.23	4.35	NO	ECM
	GDP	2.34	2.79	3.67	NO	short-run model
	CO2	3.96	3.23	4.35	NO	short-run model
	FRT	2.66	3.23	4.35	NO	short-run model
Slovakia	EC	6.02	3.23	4.35	YES	ECM
	GDP	2.068	3.23	4.35	NO	short-run model
	CO2	1.88	3.23	4.35	YES	short-run model
	FRT	2.67	3.23	4.35	YES	short-run model

presents ARDL bound testing results for F statistics at the significance level of 5%. If the F value was estimated to exceed the critical value of I(1), it was reasonable to conclude that integration was present [51,53]. If the value of the F statistic was less than I(0), it had to be assumed that cointegration between the variables did not occur, and the relationship could only be short-run.

For the EC variable in countries such as Poland, the Czech Republic and Slovakia, it has been shown that there is a cointegration between energy consumption through transport and economic growth, CO₂ emissions and the volume of freight carried. Therefore, it became necessary to estimate an error correction model (ECM) for the indicated cases to determine the speed of adjustment. In the remaining cases, the occurrence of cointegration was not proven, in view of which the decision was made to estimate only the ARDL short-run model.

4.3. Results of Evaluations of Short- and Long-Term Models

The results of the estimated long- and short-run models are shown in

Table 6. Long-run statistics.

Country	Dep. Variable	GDP	CO2	EC	FRT	ECT
Poland	EC	−0.04	0.05		0.41 *	−0.41 ***
Czech Republic	EC	0.55	0.06 **		0.47 ***	−0.64 ***
Slovak Republic	EC	0.79 *	0.01		0.06	−0.94 ***
	FRT	0.93 *	0.01	0.44		−0.45 *

An asterisk indicates the significance of the coefficients in the tables, where *, ** and *** denote 5%, 1% and 0.1% significance levels, respectively.

and

Table 7. Short-run statistics.

Dep. Variable	EC					GDP					CO2					FRT
	t	t-1	t-2	t-3	t-4	t	t-1	t-2	t-3	t-4	t	t-1	t-2	t-3	t-4	t	t-1	t-2	t-3	t-4
PL_EC
PL_GDP	0.16 **	0.18 *	−0.03	0.11 **			0.62 ***	−0.73 ***	0.62 ***	−0.25 ***	−0.01	−0.01				0.01	−0.08 *	0.05	−0.05
PL_CO2	13.66 ***	15.00 *	9.05	11.44	10.29 *	−30.34 *	1.34	−16.54				−0.9 *	−0.52	−0.41	−0.73 *	−3.09	−7.2 *	−3.31	−3.72
PL_FRT	0.92 *					2.07	−2.17	3.23	−3.23 *	1.13 *	−0.07 **					0.4	−0.3	0.47 *
HU_EC		0.19				3.32 **	−1.58	−0.21	0.08	1.18	−0.03	0.03	−0.04	−0.01	−0.03	0.19	0.22	−0.60 *	0.57 *
HU_GDP	0.01 *	0.01	0.05	−0.01	0.05		1.13 *	−0.62	1.14 *	−0.24	−0.02 *	0.01	−0.01 *	−0.01	−0.01	−0.07 *	0.05	−0.06
HU_CO2	−2.33	1.86	2.73 *	−1.3		0.01	0.01 **	−30.12 **	48.59 ***		0.27	−0.76 ***	−0.3 *	−0.37 **		−1.76	0.01 **	−5.97 **	3.99
HU_FRT	0.74	−0.7	0.42			−4.06	8.18 **	−4.15	4.24 **	−2.46	−0.06						0.26	0.13	−0.6	0.52
CZ_EC
CZ_GDP	0.63 *	0.05	0.1	−0.33 *			0.6 *	−0.17	−0.01	0.39	−0.02 *					0.12	−0.2 *
CZ_CO2	10.86 **	−6.8	4.06	−3.82		−5.1	17.86 **	1.95	5.45			0.22	−0.53 *	−0.08	−0.49 *	0.99	−1.81	−0.8
CZ_FRT	−0.9	0.71	−0.76 *	0.69 *		1.58	−1.91 *				0.04						0.31 *
SV_EC
SV_GDP	0.28	0.39 *	0.12			0.45	−0.67	0.46			0.03 *	0.02		−0.02	−0.02 *	−0.11	−0.05	0.13	0.12	−0.09
SV_CO2	−4.08	−6.79 *				11.22	−12.16	16.01 *	−20.81 **	6.11		−0.33	−0.21	0.23	0.43	7.11 **
SV_FRT

An asterisk indicates the significance of the coefficients in the tables, where *, ** and *** denote 5%, 1% and 0.1% significance levels, respectively. PL—Poland, HU—Hungary, CZ-Czech Republic and SV—Slovak Republic.

. The error correction term (ECT) represents the speed at which the variables in the model return to equilibrium after a shock or disturbance. It is a measure of the long-run relationship between the variables in the model. The correct ECT can take on any value between −1 and 1. All ECT parameters were right and statistically significant at the 1% level. Additionally, ECT had the correct sign, implying that the series is non-explosive and that long-run equilibrium is attainable.

If, after ARDL bound testing, the value of the F statistic was less than I(0), short-run models (ARDL) were developed. For values above the I(1) interval, ECM models were prepared. All models have been diagnosed, and the results did not indicate the presence of errors in the performed estimation nor stability problems.

The ECM regression results seem to fit quite well and pass the diagnostic tests against serial correlation, heteroscedasticity, functional form misspecification (RESET) and non-normal errors. The stability of the coefficients of the estimated model was also checked by using the cumulative sum (CUSUM) and cumulative sum of squares (CUSUMS²) stability tests that employ recursive residuals. The plots of CUSUM statistics for all models, presented in Appendix A, are within the 95% critical bounds, indicating that all coefficients in the estimated ECM model are stable over the sample period.

All ECT coefficients were statistically significant in the ECM models for the EC variable at the 0.1% level. For the FRT variable for Slovakia, the ECT coefficient was significant at the 5% level. According to the results obtained for the ECM models:

–

The error correction term (ECT) coefficient for Poland (EC variable) is −0.41, so it means s the speed of the adjustment is 41%. It suggests that the energy consumption in transport (EC) converges to the long-run equilibrium by 41% in one period with the speed adjustment via the amount of road transport (FRT).

–

The ECT coefficient of a stable long-run relationship for the Czech Republic is −0.64. Therefore, this implies that the energy consumption in transport (EC) converges to the long-run equilibrium by 64% in one period, with the speed adjustment via the channel of the amount of road transport (FRT) and CO₂ emissions (CO₂).

–

The (ECT) coefficients for Slovakia are −0.91 for the EC variable and −0.45 for the FRT variable. Therefore, this implies that the energy consumption in transport (EC) converges to the long-run equilibrium by 91% in one period with the speed adjustment via economic growth. Additionally, the speed adjustment for the amount of road transport is 45% in one period (via economic growth).

The diagnostic tests for the ARDL model suggest the absence of any serial correlation between the residuals and heteroscedasticity. The Ramsey RESET test suggests that the estimated model is stable. Combined with the findings of the diagnostic tests, this suggests the stability of the model in this study. The diagnostic tests further confirmed the reliability and validity of our estimations. All diagnostic results are presented in

Table 8. Diagnostic tests of estimated models (p-values).

Country	Dep. Variable	Homoskedasticity a	Ser. Corr b	Normality c	Stable d
Poland	EC	0.81	0.65	0.12	Stable
	GDP	0.43	0.06	0.40	Stable
	CO2	0.57	0.11	0.80	Stable
	FRT	0.63	0.22	0.88	Stable
Hungary	EC	0.90	0.50	0.78	Stable
	GDP	0.74	0.20	0.43	Stable
	CO2	0.36	0.27	0.73	Stable
	FRT	0.39	0.14	0.71	Stable
Czech Republic	EC	0.62	0.54	0.95	Stable
	GDP	0.76	0.18	0.76	Stable
	CO2	0.58	0.28	0.22	Stable
	FRT	0.75	0.73	0.81	Stable
Slovak Republic	EC	0.86	0.84	0.17	Stable
	GDP	0.43	0.13	0.51	Stable
	CO2	0.75	0.33	0.83	Stable
	FRT	0.16	0.21	0.65	Stable

^a Breusch–Pagan–Godfrey Test, ^b Breusch–Godfrey LM Test, ^c Jarque–Bera Test, ^d Ramsey RESET Test.

.

5. Discussion

The results illustrated in Figure 5 indicate the presence of both short- and long-run unidirectional and bidirectional relationships. Long-run bidirectional causality was not present between any of the variables studied. Thus, it cannot be concluded that there was any feedback between the factors studied. However, unidirectional long-run causality between the GDP and EC variables was present in the Czech Republic and Slovakia. It means that economic growth can increase energy consumption in the form of fuel for road transport. A unidirectional relationship in the long run also occurred between the EC and CO₂ variables for Poland, indicating that an increase in CO₂ emissions by road transport also entails an increase in energy consumption in the form of fuels. For Slovakia, a long-run relationship was also observed between CO₂ emissions from transportation and economic growth. In addition, economic growth also affected the amount and volume of freight transported in the long term.

The results, thus, show that the assumed hypothesis 1 (H1) is only valid for Slovakia. Hypothesis 2 (H2) was confirmed for the Czech Republic and Slovakia. In contrast, hypothesis 3 (H3) is valid for the Czech Republic, Poland and Hungary.

The results obtained at the same time showed similar conclusions to earlier studies for Pakistan [38], China [64] and Tunisia [47]. Although a different methodology was used in the indicated studies, the authors pointed out the existence of long- and short-run links between road transport and environmental pollution. On the other hand, it was also confirmed that economic growth entails increasing the number and volume of road transport.

In the short term, relatively varied results were observed. Unidirectional causality was found between economic growth and energy consumption by road transport in the Czech Republic, Slovakia and Poland. In the case of Hungary, there was a bidirectional relationship between the variables indicated. A bidirectional relationship, in the short term, was also observed between economic growth and CO₂ emissions in the Czech Republic and Hungary. There was also unidirectional causality between transportation energy consumption and CO₂ emissions in the Czech Republic, Poland and Hungary. Thus, the obtained causal relationships in the short term are similar to the results of similar studies in other regions of the world [39,41].

6. Conclusions

In the current situation, climate change caused by energy consumption and CO₂ emissions is a global problem. It also affects all countries in the European Union (within which there is an informal alliance of regional cooperation V4). The European Union has been the fastest and most responsive to the challenges of this problem by preparing the European Green Deal strategy and setting the principles of the so-called Fit for 55 plan, as well as indicating guidelines for manufacturers of new road rolling stock in terms of CO₂ emissions in the coming years (environmentally friendly transportation is referred to as zero-emission). According to the indicated vision, it is assumed that CO₂ emissions, for which road freight transport is to some extent responsible, will be reduced and that it also aims to reduce the dependence of transport on fossil fuels (by switching to electric propulsion). Oil, for geopolitical reasons, is a commodity susceptible to price fluctuations. In 2022, due to the war in Ukraine, many European countries observed how the geopolitical situation affected transportation based on conventional fuels.

The study focuses on the Visegrad countries, which face similar challenges in terms of energy and transportation policy changes because of their economic ties, history and political-economic transition processes. Proper transition in these areas can be an opportunity for them to achieve more dynamic, secure growth and economic development, on the one hand. On the other hand, it can become a threat to the economy. The study results indicate that despite similar economic circumstances in various countries in the Visegrad Group, there were different relationships between the variables studied during the period under review. It indicates the need to consider each country separately. Nevertheless, a common feature of all the countries studied is the long- and short-run causal relationships between economic growth and energy consumption and the transportation volume performed by road transport companies.

In view of the results, the authors provided several recommendations for the V4 countries’ economic policies in the field of transport policy. Policymakers should focus on creating such conditions so that road freight and public transport companies replace their transport fleets with vehicles using renewable electricity as soon as possible (regarding EU regulations). Eurostat data shows that in 2020 already, 37%, or almost 2/5, of the electricity consumed in the European Union was generated from renewable sources [65]. However, not all EU member states show the same commitment to developing “green” energy. The worst results in this respect were noted for the Visegrad countries. In ranking the share of renewable energy in total energy consumption in EU countries, Slovakia ranked 20th, Poland 22nd, the Czech Republic 23rd and Hungary 26th. The share of renewable energy in the total energy used in these countries ranged from approximately 10% for Hungary to approximately 30% for Slovakia. The above data reveal a significant disproportion in the energy transition of the Visegrad Group countries compared to other EU countries [66].

Activities related to replacing road transport vehicles with zero-emission vehicles should be one of the elements of the energy transition in these countries. However, at the same time, they depend on this transition and the development of “green” energy technology and its infrastructure. Energy transformation is cited as essential to maintaining the Visegrad countries’ ability to grow and develop their economies in the future [67]. With regard to road transport, it will reduce its dependence on fossil fuels and its vulnerability to fluctuations in oil prices. It will also enable the long-term and sustainable development of the entire economy. Such solutions will also contribute to reducing the negative effects associated with road transportation in the form of CO₂ emissions and other harmful substances. A sufficiently rapid and smooth transition to modified forms of propulsion will allow these countries to continue to benefit from the large share of road freight transport in GDP, its high importance for exports and imports, and employment in the road freight transport sector in these countries. Simultaneously, a substantial boost in the demand for renewable electricity for road vehicle propulsion will accelerate technological development in renewable energy and electric propulsion and promote economies of scale during energy transition projects in the transportation sector.

In conclusion, it should also be pointed out that the study conducted has some limitations, mainly due to the short time series analysed. It is mainly due to the consequences of historical events, as a result of which the V4 countries functioned as communist countries until 1990 under centrally planned political systems. The rule at the time was that public statistics were not collected in the areas studied, or the data were falsified. In the future, when more data is collected, the study can be repeated to include new events and additional factors and their impacts. Given the gap in the research area, it would also be reasonable to conduct a study covering all the European Union countries using the NARDL models.

Based on our study’s findings, the V4 group should invest in green transport. Specifically, the Visegrad Group countries’ governments need to stimulate CO₂-reducing activities in road transport. It is possible through increasing alternative energy resources in transportation, such as biodiesel fuel and electromobility. Additionally, other types of freight transport, such as railways and inland sailing, require development. Countries should also promote intermodal and combined transport as well.

Reducing fuel consumption and CO₂ emissions in road transport can also be achieved through tax preferences for companies investing in modern vehicles that meet the EURO 6 emission standards. The results of this research can also be used to create a proper future environmental policy for road freight transport in the V4 counties group.