Divergences between EU Members on the Sustainability of Road Freight Transport

Abstract

The Europevan Union is highly dependent on energy. This paper analyses energy consumption in the transport sector, representing approximately 30% of total energy consumption. A particular focus will be placed on road freight transport, which accounts for 40% of total transport energy consumption, trying to contribute to its rationalization. The road freight transport volume (TKM: tonne-kilometres) vs. gross domestic product (GDP) ratio fell in 2010–2022 by an average of 17.6% in EU-27, meaning that there was decoupled growth. Germany is the country with the highest decoupling, nearly 36%. On the contrary, in Spain there is a coupling because the evolution of TKM and GDP was almost identical. The paper proposes possible relevant factors in road freight transport activity that may justify the discrepancies in values within EU member countries, showing the positive and negative impacts of the different variables studied. In addition, the SARIMA model is applied to forecast the evolution of relevant indicators of road transport in different countries.

1. Introduction

A significant relationship between transport activity and the economy has existed since ancient times, given that many civilisations based their capacity and economic prosperity on developing transport infrastructures. The first relevant scientific approach to studying this relationship was developed in the 20th century with the design of a transport activity model [1], which led to a direct connection between the demand for transport and the degree of economic activity.

Theses studies were based on previous research on the modelling of pure transport demand on supply-side characteristics [2], as well as on comparative studies of the evolution of demand for road and rail freight transport [3]. Soon the relationship between transport activity and the so-called input–output flows that involve the exchange of goods between the different economic activities that make up GDP was studied [4,5].

The interest in exploring the inverse relationship, linking gross domestic product (GDP) with transport activity, is growing since economic analysts want to obtain what is known as “leading economic indicators”, which will be very useful in business decisions and policies. In this sense, the European Union has promoted the development of a transport policy evaluation model called HIGH-TOOL, which makes it possible to identify and evaluate the impacts of its policies on transport, the environment, and the economy. Its main innovation lies in integrating other models and the fact that it is a free, open-source tool [6]. These tools did not allow us to evaluate and predict general trends in some relevant characteristics in the development of transport activity and have also remained an outdated project, so were not valid for this article.

Recently, different studies have examined the relationship between GDP and the logistics performance index (LPI), a synthetic indicator defined by the World Bank to understand the competitiveness and quality of logistics services. This analysis is carried out on the international trade of nations in each continent and worldwide. Its findings reveal that LPI positively correlates with net exports worldwide [7,8].

The link between the evolution of GDP and freight transport activity is straightforward and responds to specific models and variables [9]. For example, in the USA, there is a strong relationship between economic growth and the increase in air freight and passenger transport activity [10].

On the other hand, in some countries, transport has grown slower than GDP, which the authors define as “decoupling”, indicating weakening of the transport–economy link. In others, transportation has decreased in the context of favourable economic growth rates, manifested as “strong or absolute decoupling”; thus, authors such as Ballinghan et al. [11] go so far as to suggest that the link has been broken. Some researchers also observe more substantial decoupling in “peripheral” EU countries and a stronger correlation in Central European countries [12,13].

In further research, there has been a significant decoupling of freight transport activity and economy in Europe, suggesting that an essential factor in this change may have been the increased efficiency of transport [14,15]. Tight et al. [16] concludes that the evolution of the relationship between transport activity and the economy is moving towards greater sustainability. Other studies in specific domains lead to equivalent conclusions [17,18,19].

Two new concepts are developed: (1) immaterialisation, referring to the reduction in transport intensity and economic activity, and (2) dematerialisation, corresponding to the reduction in emissions and energy needs, closely linked to the decarbonisation process. Studies such as Andreoni and Calmarini [20] do not hesitate to proclaim a strong decoupling of environmental emissions from economic activity, despite the difficult economic situation.

Savy and Burnham [21] reviewed the global realities and figures of transport, pointing towards new, more complex and sophisticated contractual frameworks as opposed to the purely bipolar vision of producers and transporters [21]. Additionally, Liimatainen and Pollänen [22] analyzed how different productive sectors are more or less decoupled from transport activity. Stahel [23] links economic decoupling effects with the emergence of circular economy trends.

From 1995 to 2012, an extensive study of coupling and decoupling in the EU, highlighting the 2008 crisis, has been presented by Botzoris et al. [24]. Meanwhile, Alises and Vassallo [25] updated the study of decoupling of transport activity regarding GDP input–output tables in Spain, which confirms the trend toward decoupling. Loo and Banister [26] deepened the concepts of immaterialisation and dematerialisation to demonstrate, as had already been pointed out in the research field, that the results indicate lower decoupling in peripheral countries than in central countries.

Loo and Banister [26] proposed a new perspective, where the price of oil might have produced different effects on the demand for transport of goods along the supply chains. In geographical areas linked to consumption, such as the EU and, in general, in developed countries, oil prices might constrain transport activity and produce the decoupling of their economies mentioned above. In other countries linked to the production of goods, such as Southeast Asia, decoupling is practically non-existent. In this sense, Kos-Labedowicz and Urbanck [27] concluded, from equivalent partial results, that the explanation can be found in the different technological bases of nations and that the application of new technologies in the transport process leads to decoupling with greater intensity, advising a decisive commitment to them.

Recent research has analysed the coupling and decoupling relationships between transport-related energy consumption and GDP, finding that most countries have passed the coupling threshold after the last crisis and are experiencing a decoupling mode. This situation represents a political and institutional challenge, as the observed decoupling should be abandoned to strengthen and promote green and efficient mobility [28].

In contrast to what is conducted in this article, there were no papers focusing their studies on a sufficiently large number of EU countries or analyzing, in detail, specific parameters of the development of road freight transport activity (TKM, Tonnes, VKM, Empty VKM, Journeys, Empty Journeys, GDP, General Energy Use, and Energy Transport by Road).

Eurostat [29] data will be used for energy needs in the EU from 2010 to 2022. Firstly, energy production in the EU-27 has remained at around 250 M tonnes of oil equivalent (toe) over the last decade, with the lowest output of the entire decade in 2022, probably due to the reduction in activity resulting from the pandemic.

In 2010, the EU-27 consumed 973 Mtoe, falling to 902 Mtoe in 2022, representing a decrease of almost 7% in total energy consumption. This decrease is also observed in the highest consuming countries, such as Germany, France, Italy, and Spain. The EU-27 is highly energy-deficient and, therefore, has to import many energy products.

Analysing total energy consumption in the top-consuming countries in the EU, Germany is in first place, followed by France and Italy, and Spain is in the fourth position.

Analyzing consumption by sector in 2010–2022 (Figure 1), it can be observed that the leading sector is transport, with 29.67% of the total, followed by households with 27.39% and industry with 25.63%.

Smokers et al. [30] found the transport sector must reduce its emissions by 50% compared to its 1990 levels to be sustainable.

This article studies the relationships of economy vs. transport activity and transport activity vs. energy consumption, in order to determine the coupling of these elements and determine whether or not the trend in transport energy consumption is unsustainable by itself and the emissions it generates.

Abid and Sebri [31] focused on specific countries and investigated the causal relationship between energy consumption and economic performance for the industry, transport, and residential sectors. Tvaronavičienė [32] analyzed energy efficiency in the transport sector of three European countries, considering policy implications, resource management and conventional energy resource efficiency. Aza and Escribano [33] indicated that despite the growth experienced in transport services in Spain and Europe, there is a clear downward trend in the intensity of transport services relative to GDP (increasing decoupling).

Bernali and Feki [34] found irrefutable dynamic links between transport, consumption, economy, and gas emissions in a study period of 34 years of development in Tunisia. In this regard, it can be observed that measures adopted in different political and economic environments yield notably different results in the relationships between economy, transport activity, and emissions, as Shafique et al. [35] pointed out. Additionally, Sun et al. [36] also suggested the decisive contribution of technologies in a proportional reduction of emissions in transport activity. Touratier-Muller and Jaussaud [37] provided a further step, recommending that states should encourage shippers to contract their activities with carriers that maintain certain emission characteristics in their vehicle fleets, for example, a certain mix of Euro V and Euro VI vehicles. This hypothesis seems to be supported by the results of the study by Pollänen et al. [38], which show that there has been very little awareness and little action taken by the road freight transport sector over the last 12 years.

On the other hand, concerns about emissions mark much of the research. Achour and Belloumi [39] investigated the causal relationships between transport infrastructure, transport value added, gross capital formation, and transport energy consumption, including CO₂ emissions. Saidi et al. [40] analyzed the same causal relationships. Their results show the long-run unidirectional causality of transport value added, road transport-related energy consumption, CO₂ emissions, and gross capital formation. Mohsin et al. [41] consider the emissions problem, indicating that the transport sector consumes 25% of the world’s energy with 23% of the world’s emissions. A key feature in this relationship will be the adequate utilization or occupation of transport capacities of the means involved [42].

Osorio-Tejada et al. [43] incorporated relevant variables to be taken into account regarding consumption and emissions of the transport fleet, which are those related to orographic conditions and that can enclose variations of up to 145%.

2. Research

This research addresses two distinct lines among the scarce existing specific research that analyses the decoupling between the economy and energy consumption. First, the above-mentioned by Profillidis et al. [28] certifies the existence of this decoupling, without deepening the analysis of the causes or indicating the challenge that needs to be addressed to maintain it. Secondly, more recent works focus on analyzing transport intensity and its effects on the economy, studying the results of social and economic activities and how to model the transport intensity of national economies in European countries.

In addition to the already highlighted connection between economy and transport, we must consider that this connection is particularly marked for sustainability when road freight transport is the case.

Most of the freight transport volumes are carried out by road. Most of the freight transport volumes are carried out by road. The total demand for inland freight transport (i.e., by road, rail, inland waterways and oil pipelines) increased by 22% between 2000 and 2019. In 2019, 2411 billion tonne-kilometre kilometres (TKM, a measure of freight transport activity that we will be explained below in the first section of this point) were transported by these modes. The share of road transport in this demand grew from 68% in 2000 to 73% in 2019 [44].

2.1. Methodology and Structure of the Research

The methodology of this research begins by performing an analysis for different EU territories: (i) energy consumption versus economic evolution, (ii) the evolution of road freight transport activity versus economic evolution, and (iii) a detailed study of the territorial divergence of some of its specific indicators linked to efficiency in this road freight transport activity. We add to the methodology the development of a predictive model for the evolution of these specific indicators.

All the data used were obtained from Eurostat [29], as well as those already used in the energy and economic fields. In this sense, it is now necessary to clarify some specific magnitudes that we will use in our research. These are data related to road freight transport with the following significance:

• Tonne-kilometre (tkm or TKM) is a unit of measure of freight transport that represents the transport of one tonne of goods by a given transport mode (road). It is the principal measure of transport performance, as it combines its two main dimensions, mass transported and transport distance, to signify the energy consumed.
• Tonnes (t or T) is the weight transported and km is the distance travelled in kilometres by transport vehicle displacement and can be a distance travelled loaded or empty.
• Vehicle-kilometre (vkm or VKM) is the distance travelled in kilometres for one trip of the transport vehicle and can be a loaded or empty distance.
• Journeys are the number of trips made by a transport vehicle and can be a loaded or empty trip.

By comparing TKM with the two magnitudes VKM and journeys, we can analyze the efficiency and sustainability improvement of this mode of transport by analyzing distances and journeys made without load of goods in each geographical area and their evolution over time.

2.2. Data and Analysis

Data from Eurostat [29] were used for this study.

2.2.1. Decoupling of the Economy and Energy Consumption

Examining the overall evolution of energy consumption in the EU relative to GDP (Figure 2), the EU-27 achieved significant proportional reductions in its energy consumption relative to its GDP. This trend was general in all EU countries, with only minor mismatches to be highlighted, with larger decoupling in the Netherlands, Belgium, and Germany, medium decoupling in Portugal and France, and smaller decoupling in Spain and Italy.

2.2.2. Decoupling of the Economy and Energy Consumption in Transport

Examining the data from the same perspective of transport energy consumption (Figure 3), there were also differences between countries. Our previous clusters were elongated, showing significant relevant aspects, such as the strong energy decoupling of Dutch transport and the increase in the transport consumption ratio in France.

2.2.3. Decoupling of the Economy and Energy Consumption in Road Transport

Finally, this research was extended to road freight transport, which, as mentioned above, represented a decisive volume of transport activity in Europe. No segregated statistical data were available on which type of consumption belonged to freight and passenger transport in public service and private use, so we had to study its general magnitude. Analysing the evolution of this form of energy consumption (Figure 4), we obtained results practically identical to those of Figure 3.

The economy has been linked to transport activity and vice versa. The evidence suggests that certain decoupling was taking place in this linkage, and (perhaps as a consequence) there was also a decoupling of these magnitudes and energy consumption. However, it was also clear that the exact relationship was not followed in all countries and that there were significant differences between them. The peripheral nature of some countries would offer scientific explanations for this trend, but they might well be due to the influence of other, less apparent variables. Because of this possibility, it was considered necessary to further investigate the knowledge of the magnitudes that might influence this decoupling.

In this sense, Aza and Escribano [33] developed a study focusing on the aggregate transport and storage services sector. They found that the measure of the share of growth in this sector derived from elements and aspects of great relevance, such as information and communication technology use and investment, human capital, technological progress, and total factor productivity. They also mentioned other economic effects that the growth accounting model did not capture but were nevertheless significant.

2.2.4. Decoupling of the Economy and Energy Consumption in Road Freight Transport

This paper will focus on road freight transport activity in the EU and some of its countries, such as Spain. Considering that TKM is a measure of road freight transport activity and GDP is a measure of economic activity developed, for this purpose, the start point will be the ratio between the TKM produced and GDP volume (Figure 5).

It revealed some interesting aspects: (1) Most of the ratios showed a progressive decrease. (2) However, some countries increased the ratio around the middle of the decade. (3) Spain seemed to be the exception, having increased its ratio finally. (4) The EU was on a middle path as a whole and at a level maintained with small fluctuations over the period.

Decoupling or recoupling could not and should not be considered a loss or increase in the relationship but rather a change in how it was established. This change might have been due to multiple factors, including the fact that the economy was becoming more and more linked to services and the management carried out by virtual platforms. Conversely, the economy relied on an increase in the distribution of goods and/or a change in the distribution pattern of goods, for example, transporting more weight over longer distances

The decoupling of transport activity from the economy is a positive aspect for sustainability because it allows us to reduce the environmental impact as well as the economic costs. More specifically, the reason is that it achieves less overloading of infrastructures; reductions in the environmental impact of its emissions; reductions in social damage, especially traffic accidents; and reductions in its energy consumption, since the relevant question here is if there is a desirable decoupling or an undesirable recoupling between the energy consumption of transport and economic growth.

The research by Andrés and Padilla [45] detected an improvement in the energy intensity of road freight transport. They attributed this effect to the technological progress of transport equipment but also to what they saw as a relative improvement in the density of loads; in other words, the capacity of vehicles seemed to be better exploited, simply pointing in this direction without adequate contrast.

2.3. A Selection of Representative EU Countries

The most comprehensive measure of road freight transport activity is the tkm generated, as we have already mentioned. We can observe the magnitude of this measure from 2010 to 2022 in

Table 1. Millions of TKM produced by EU territories from 2010 to 2022.

Country 1	Millions of TKM (2010–2022)	Ranking
EurUnion 27 (from 2020)	22,173,377
Germany	4,048,051	1
Spain	2,929,703	2
France	2,217,939	3
Netherlands	1,729,042	4
Italy	913,865	5
Belgium	441,855	6
Portugal	429,376	7

. Therefore, the first step is to analyze the activity of this transport with the variable that most fully represents its overall activity, which is none other than the TKM carried out.

Table 1 shows that the countries with the greatest contribution to TKM production are Germany, Spain, and France, which together account for 41.5% of the total magnitude in the EU. Furthermore, if we look again at Figure 5, we see that these three countries lead precisely three general trends with respect to the territories studied:

• Germany, like Belgium and the Netherlands, experienced strong growth in economic activity and clear decoupling of its transport activity, which grew slightly.
• France, like Portugal and Italy, experienced average growth in its economy and also maintained decoupling with its transport activity, which decreased slightly.
• Spain (in this case, in stand-alone mode), however, maintained average growth in its economy but did not experience decoupling in the evolution of its transport activity; on the contrary, since it increased more than proportionally to its economic activity, which we can clearly see in this figure if we draw a 45-degree axis in the upper right quadrant.

For this reason, for the purpose of clarification, we will henceforth focus our studies on these countries, together with the general evolution of the EU, in order to analyze specific road freight transport performance magnitudes.

2.3.1. Tonnes Transported by Road vs. TKM for Selected Countries

A partial indicator of the development of road freight transport is the quantity of mass transported. To analyze it, we will examine the evolution of Tonnes transported vs. TKM, based on what was carried out in Section 2.2.4 with GDP vs. TKM, in the evolution of its index, based on a 100 percent increase above the data of the initial year of the series in 2010 (Figure 6).

Although the evolution of the mass transported versus the TKM executed from 2010 to 2022 are related variables, they allow us to analyze differences in the behaviour of each country, between mass and distance transported. Germany experienced a moderate increase in the mass transported, compared to the significant reduction in France and the maintenance in Spain.

2.3.2. Distances Traveled VKM vs. TKM for Selected Countries

Next, we studied the relationship between the distance travelled per trip (vkm) and TKM, considering its evolution, whereby the value 100 is 2010.

In this case, the evolution of the distance travelled versus the TKM executed is comparable to analyzing the evolution of the mass transported. From Figure 7, it can be concluded that Germany kept the measure of its transport distances very constant, France slightly reduced them, and Spain, however, experienced a remarkable tendency to increase them.

2.3.3. Journeys vs. TKM for Selected Countries

The next partial measure of the development of road freight transport is the number of journeys or trips made transporting goods. To study it, once again, we focus on comparing the evolution of its 2010 index 100 against that of TKM.

From Figure 8, it can be concluded that: Germany had a slight tendency to increase its displacements, as opposed to the drastic reduction in France and maintenance in Spain.

2.3.4. Distances Traveled Empty vs. TKM for Selected Countries

One measure of the development of road freight transport is unladen vehicle movements, an obvious indicator of wastage.

From Figure 9, it can be seen that only France has reduced unladen transport and transport volume since 2010, while both Spain and the EU members have increased unladen transport (greater number of vehicles and/or greater distance) and increased transport volume (greater tons transported and/or greater distance). In the case of Germany, the number of unladen transports also increases, while it reduces the volume of transport by almost the same percentage as France.

2.4. No-Load Trips vs. TKM for Selected Countries

Another measure that is also a significant waste indicator for the development of road freight transport is the transportation of unloaded vehicles.

From Figure 10, it can be concluded that only Germany worsened in the trend of this indicator, and while Spain reduced it slightly and France obtained extremely positive results in reducing it.

3. Trends Analysis

Upon the completion of this paper, a set of temporal data have been utilized, encompassing data on kilometres travelled with both empty and loaded loads, as well as TKM.

3.1. Experimental Setup

The years of the dataset vary from 2010 to 2022 (inclusive), encompassing 13 years, with 52 quarterly values for the series of total kilometres. Due to the fact that the other two series are not originally in a quarterly format, they have been weighted annually. Therefore, quarterly weighting has been conducted based on the annual relationship, as demonstrated in the following

Table 2. Example of original data.

	Annual	Q1	Q2	Q3	Q4
Total km	12,000	4000	2000	3500	2500
Full km	9000
Empty km	3000

and

Table 3. Example of relative data.

	Annual	Q1	Q2	Q3	Q4
Total km	12,000	4000	2000	3500	2500
Full km	9000	3000	1500	2625	1875
Empty km	3000	1000	500	875	625

. For the SARIMA model, the data until 2020 (prior to the impact of COVID) are used, and the data from 2021 and 2022 are used to validate the model.

The chosen division allocates 39 quarters for training and 13 quarters for testing, which translates to 75% and 25%, respectively. Therefore, for training, we have data from 2010 to September 2019, and for testing, we have data from December 2019 to December 2022.

3.2. Experimental Design

The present study focuses on comprehensively understanding two characteristics of the dataset. Firstly, it examines the evolution of TKM (tonne-kilometres) in Germany, Spain, and France. Secondly, it analyses the relationship between empty kilometres travelled and loaded kilometres travelled in these same countries, supported by the results of the EU.

To conducting a time series analysis, the well-established Seasonal AutoRegressive Integrated Moving Average (SARIMA) model [46] was employed. In this renowned statistical model, a comprehension of the parameters p, d, and q is imperative, where p denotes the order of the autoregressive component, d signifies the degree of differencing in the model, and q represents the size of the moving average.

To standardize the model, the TKM series from the EU and the series of empty and loaded kilometres within the EU were used as references.

Various tests can be utilized to determine these parameters, such as the augmented Dickey–Fuller test [47] for d. This test establishes a threshold p-value of 0.05 for rejecting the null hypothesis and, depending on the degree of differencing required to attain this value, it is set as the value for d. In the current study, as depicted in the lower graph, a degree of 2 was obtained (Figure 11).

The original series p-value was 0.96382, in order differentiation 1 the p-value was 0.42203, and, finally, in order differentiation 2 the p-value was 0.00053.

Relying on the autocorrelation function (ACF) and partial autocorrelation function (PACF), it has been determined that the values of q and p are both 1.

In this study, the Python packages statsmodels (https://www.statsmodels.org/ accessed on: 12 July 2024) and skforecast (https://skforecast.org/ accessed on: 12 July 2024) were utilized, both of which are widely recognized for their utility. The results they yield, for example, in EU TKM are as show in Figure 12.

The results obtained from the TKM executions are displayed in

Table 4. Results obtained.

	EU TKM		EU VKM Empty		EU VKM Full
	MSE	MAE	MSE	MAE	MSE	MAE
statsmodel	0.0313	0.1204	0.0314	0.1474	0.0202	0.0913
skforecast	0.0434	0.1475	0.0314	0.1474	0.0202	0.0913

. Here, it is evident that the statsmodels library provides superior results in the first instances, whereas in the other two obtain the same results; thus, it will be utilized further for forecasting purposes.

3.3. Forecast

In the forecasting section, we have relied on the previously mentioned configuration. Additionally, the statsmodels library has been employed to further optimize these results.

Figure 13 illustrates the results achieved by SARIMA. In this graph, the ratio of empty kilometres to loaded kilometres for each country has been depicted. Solid lines represent the actual existing series, while dashed lines represent the predicted series.

Despite the pressure caused by rising fuel prices, which is forcing carriers to reduce their level of “dead miles”, it is difficult to keep up with the pull of e-procurement. Therefore, the forecast is increasing empty transportation (Germany and France) or plateau (Spain).

In addition to the waste of gasoline, the vehicle wear and tear and hours of unproductive driving that “dead or empty kilometres” entail also translate into polluting emissions. This clashes squarely with the demand for more sustainable transport and the industry’s own efforts to reduce its carbon footprint. The data by country are also disparate in this case, but, beyond the figures, they point to a more than significant impact on pollution.

Minimizing or even eliminating empty miles is no easy task. It requires balancing routes, organizing services to transport customers there and back, and ensuring that these services are provided with suitable vehicles, equipped, for example, with refrigeration if the goods are perishable. The boosting of digitalization and logistics platforms equipped with artificial intelligence can help to meet this challenge.

In Figure 14, the evolution of the TKM for each of these three countries can be observed. The trend forecast is the convergence of the transport of tons per kilometre for the three countries. This may be due to the pressure caused by rising fuel prices, which is forcing carriers to standardize profitable routes and reduce empty haulage.

4. Conclusions

The research findings, in most cases, revealed a decoupling between macroeconomic variables. A summary of the relevant figures related to road freight transport is presented in

Table 5. Summary of the relevant indicators studied. Period from 2010 to 2022.

	Energy (%)		Road Freight Transport (%)
	General Energy Use vs. GDP	Energy Consumption in Road Transport vs. GDP	TKM vs. GDP	TKM	Tonnes	VKM	Empty VKM	Journeys	Empty Journeys
EU—27	−39.21	−30.44	−17.63	19.31	1.31	15.16	7.67	−5.21	−2.24
Germany	−43.52	−36.84	−35.79	−2.92	13.80	0.15	12.21	11.84	11.30
Spain	−30.74	−24.12	1.16	26.97	1.91	22.76	10.55	−0.83	2.40
France	−34.85	−23.94	−28.06	−4.85	−18.60	−10.63	−36.59	−36.47	−27.31

, highlighting their positive or negative impact on energy consumption in road transport as green (best value) or red (worst value).

The analysis of these data leads to the following conclusions:

• There are obvious cases of energy improvement, such as France, which have achieved significant reductions in energy consumption in road transport supported by good indicators of progress in their road freight transport (reduction of TKM or transport activity and reduction of other ratios related to inefficiencies, such as empty trips and empty kilometers) This is perhaps not the only factor, but it is relevant to the scope of this research. In the future, a study will be proposed to analyze the cabotage activities realized by other countries in France, in order to establish the explanation in order that the improvement of the energy has not been greater.
• Germany is a paradigmatic case. Despite having experienced growth in road freight transport activity and counteracting factors related to inefficiencies, it has nevertheless obtained significant improvements in its energy consumption. This case is undoubtedly supported by other factors external to our study, which can be addressed in future research on dematerialization in the German economy together with possible improvements and reductions in transport processes, could yield relevant results in this regard.
• The European Union as a whole has experienced a reduction in transport energy consumption in the context of an increase in transport activity (TKM), reducing the number of trips, both laden and empty, but increasing the distances traveled empty.
• In contrast to the other EU countries studied, where there was a decoupling between GDP vs. TKM, allowing evidence of an improvement in the sustainability of transport, for Spain, these two variables are coupled, so there is no such improvement. In recent years, in Spain, road freight transport activity (TKM) has grown proportionally more than economic activity (GDP), and it has done so in a particularly intense way with respect to the distances travelled in these trips (Figure 15). Once again, future research on the cabotage activities performed by Spanish vehicles could reveal outstanding results, as well as a further analysis of the efficiency in the transportation activity.

Consequently, given the very detailed data on the performance of this road freight transport activity in Spain, it would be desirable to conduct a specific study of its performance over the years under investigation. As a result, it would probably be possible to reach more concrete conclusions aimed at the desired reduction in energy consumption of this activity, considering its essential impact on general energy consumption.

Although we have commented on the predictions obtained with the SARIMA model applied to transport data in the EU, for future research, we believe that we should study, in depth, the results obtained by relating them to other energy and economic variables.