Exploring External Costs on the Example of Sea–Land Transport Chains of Refrigerated Cargo Between Spain and Poland

Abstract

The aim of this study is to identify sea–land transport chains of refrigerated cargo against the background of direct road transport on the Valencia–Warsaw route based on the analysis of external costs with particular consideration of the external costs of energy production in road and sea transport. Fruits and vegetables were assumed to be a group of refrigerated cargo in the study. The considerations so far have shown that many refrigerated cargo operators transported between Spain and Poland are only sometimes guided by full and up-to-date knowledge of the amount of external costs in their transport decisions. It seems reasonable to undertake information activities at the national level addressed to cargo operators and organizations associating with them, aimed at presenting the total costs of both direct and external transport.

1. Introduction

Along with the progressive growth in freight transport, there has been a parallel increase in the Earth’s average temperature, causing climate change. Since the second half of the 19th century, the Earth’s average temperature has increased by about 1 °C [1]. This phenomenon is the result of excessive emission of greenhouse gases into the atmosphere, mainly from the combustion of fossil fuels. The intensity of the effects of climate change depends primarily on the implementation or omission of actions aimed at limiting a further increase in the average temperature on Earth [2,3]. It is essential from the point of view of environmental protection to consider aspects related to the impact of these transport solutions on the natural environment and human health and life, in parallel with planning new sustainable transport chains.

One of the fundamental challenges in building a sustainable transport policy is to seek the optimum functioning of all modes of transport [4]. One solution that allows for maintaining the important advantages of land transport while contributing to reducing external costs related to the impact on the environment is sea–land transport chains.

Maritime transport plays a significant role in world trade in terms of volume and value [5,6]. However, only a small amount of cargo reaches recipients in port cities. Most of it requires further transport inland, which requires cooperation with land modes of transport [7]. In many cases, integrating sea–land modes into a single transport chain is the most optimal solution for cargo transport [8].

The sea–land transport chain is created to meet every transport need requiring the involvement in the movement process between the sender and the recipient of at least two modes of transport, one of which is always sea transport. J. Neider and D. Marciniak-Neider defined [9] the transport chain as “a rational sequence of transport, reloading and storage processes coordinated from the technical, technological, organizational and commercial point of view—aimed at moving goods necessary for the functioning of the economy”.

Sea–land transport chains covering all technological processes of cargo transport from the sender to the recipient may, depending on the model, occur in a simple or complex form and may also involve means of transport of two or more modes of transport. There may be a different number of links in the chain. The links of the chain interconnected at the starting and ending points involve means of transport, loading equipment, and auxiliary means necessary to create the entire transport chain [10].

Sea–land transport chains are complex and consist of combinations of sea transport links with land transport branches. They can be an alternative to direct road, rail, or rail-road chains [11,12].

Both Poland [13] and Spain [14], in the assumptions of their national transport development strategies, indicate interest in the development of short-range shipping and cooperation of sea transport with land modes of transport within sea–land transport chains. This results primarily from the need for greater diversification of transport modes, fearing the inefficiency of road transport infrastructure, and the growing pressure to reduce the adverse impact of this mode of transport on the natural environment and human health and life.

Lower unit costs over short distances characterize road transport. With increasing distance, the unit cost for road transport increases faster than rail and sea transport (The presented picture is simplified. In the case of sea transport, the unit cost of transport tends to decrease with increasing distance.). Rail transport becomes more cost-effective for distance A, while sea transport becomes more cost-competitive for distance B. However, the above relation is a simplification resulting from, among other things, the assumption that all modes of transport are interchangeable (in reality, for many locations, rail or sea transport is not available) [15] (pp.106–107). The topic of dependencies of unit transport costs in sea transport was discussed in more detail by D. Bernacki and Ch. Lis [16]. A natural barrier distorting the above relations may be the geographical issue of the lack of possibility of functioning by sea transport in the land environment and by land transport in the sea environment. In addition, it is necessary to remember the difference in the size of a single batch of cargo that can be transported by means of different transport modes.

It should also be noted that each transport branch is susceptible to the general economic situation or social development. This means that the position of individual transport branches may differ in different periods. This is influenced by factors that can be divided into two groups, where some have a direct and others an indirect impact [17].

The research undertaken in this article refers to the example of transport service of trade between Spain and Poland in the segment of refrigerated cargo. According to data from the National Revenue Administration in Poland (Polish: Krajowa Administracja Skarbowa, KAS), the turnover of goods in Poland and Spain in 2019 amounted to 4,028,657 tons, of which over 20% were refrigerated cargo. As data on trade exchange indicated, as much as 98% of refrigerated cargo transport between Spain and Poland was handled as part of direct road transport. Such significant transport carried out on average over 2.5 thousand km raises the question of whether the best solution from the point of view of sustainable development is the absolute dominance of road transport in this service.

The article presents a research hypothesis: Adopting social costs as a criterion for assessing the competitiveness of various transport solutions contributes to improving the competitiveness of sea–land transport chains of refrigerated cargo between Spain and Poland in relation to direct road transport.

The analysis was conducted based on an evaluation of competitiveness, which used traditional criteria focused on internal (own) costs. This study utilized the research methodology outlined by L. Filina-Dawidowicz, M. Kaup, and A. Wiktorowska-Jasik [18] to analyze the transport route between Valencia and Warsaw. Additionally, a comparison of external costs was made for two transport scenarios: direct road transport and a sea–land transport chain, based on the methodology presented in the Handbook on External Transport Costs [19].

The Handbook [19] provides standardized, widely accepted methodologies tailored to European conditions, ensuring consistency and comparability in the evaluation of external costs. This makes the findings more applicable and actionable for policymakers and industry stakeholders operating within the region.

Most research on transport competitiveness overlooks the broader impacts of social and environmental factors, focusing instead on narrower economic aspects. This study takes a different approach by emphasizing the role of social costs in evaluating competitiveness. By exploring the trade-offs between sea–land and direct road transport, it offers a more comprehensive perspective. This approach provides valuable insights for policymakers and industry leaders aiming to refine transportation strategies and reduce environmental impacts.

2. Methodology

In order to achieve the stated goal of the study and to obtain answers to questions related to the basic scientific problems of the work, the following research methods were used:

• Critical analysis of the literature on the subject—Review of the theoretical foundations of the topic and criticism of the research problem in the light of current scientific achievements and directions of development of transport policy. This method also proved helpful in the development of an interview questionnaire used in empirical research.
• Method of document research (statistical and market reports, internal materials of companies)—Analysis of available domestic and foreign statistical data on trade, in particular on refrigerated cargo and information on the current state of refrigerated cargo handling within the sea–land transport chains between Spain and Poland.
• Interview method—Research based on an interview questionnaire addressed to administrators of refrigerated cargo imported from Spain to Poland. In the analysis of the information obtained during the interviews, the Pearson correlation coefficient was used, m.in. Two approaches to the analysis and assessment of various transport scenarios are also presented, the traditional one based on the amount of internal costs and the one extended by the amount of external costs.
• Inductive and deductive reasoning methods—The application of which made it possible to synthesize, generalize, and interpret the obtained and analyzed results of theoretical and empirical research. Inductive reasoning was used in empirical research (model construction), while deductive reasoning was used in the verification of hypotheses and theoretical assumptions made in the work.

In the research of charge administrators, one of the methods of non-random sampling was used—purposive selection, commonly used in qualitative research, which is not based on the theory of probability (hence, it is not the basis for statistical generalizations) but is made according to the subjective beliefs of the researcher and his knowledge about the studied subjects, taking into account the objectives of the research. In the case of the conducted research, the choice of the method of purposive selection meant that the study covered those entities whose main load in the entire turnover was vegetables and/or fruit imported from Spain to Poland. These cargo groups dominated the total trade in refrigerated cargo between Spain and Poland. Thanks to the use of deliberate sampling, the study involved people who provided knowledge useful in the context of the objectives of the dissertation.

3. Assumptions and Input Data for the Case Study

The direct road transport route between Valencia and Warsaw and the sea–land chain between Valencia–Gdańsk–Warsaw is shown in Figure 1.

For the analysis, the size of the transport batch was assumed to be 1.99 thousand tons. The size results from the technical conditions of the model ship of 1100 TEU used for the analysis, in which 20% is the number of places intended to transport refrigerated containers. In order not to assume the maximum load, the load was taken at the level of 18%, which gives us 207 TEU or 1.987 thousand tons (considering the weight of 1 TEU = 9.6 t [20]). The adopted assumptions make it necessary to use 104 trucks carrying a 40’ container for the land sections.

For the purposes of the analysis, the same volume of transport was assumed: 156 thousand tons, which constituted 25.7% of the annual volume (total import of cargo groups 07 (vegetables)—239,860 tons and 08 (fruit)—368,135 tons from Spain in 2019 amounted to 608 thousand tons. Data from the National Revenue Administration (Polish: KAS)) of fruit and vegetable imports from Spain to Poland in 2019, and 24.2% of the total fruit and vegetable trade between Spain and Poland (the total trade in fruit and vegetables with Spain in 2019 amounted to 645 thousand tons. Data from the National Revenue Administration). External costs were estimated using the methodology proposed by the European Commission in 2019 [19].

As mentioned earlier, external costs were calculated for the connection between Valencia and Warsaw for two alternative transport solutions based on the data on transport distance and transport work presented in

Table 1. Average distance and transport work for refrigerated cargo transported between Valencia and Warsaw.

Transport Mode	Transport (Thousand tons)	Average Transport Distance (km)		Transport Performance (Thousand tkm)
road	156	2750		429,000
sea–land *	156	4540 (sea)	4 890	708,240 (sea)	762,840
		350 (land)		54,600 (land)

* Sea–land transport was divided into two sections: sea: Valencia–Gdańsk 4540 km (i.e., 2453 NM) and land: Gdańsk–Warsaw 350 km. Source: own study.

. Distance information was developed based on data from Searoutes and Openmaps.

4. Characteristics of the Refrigerated Cargo Group

Refrigerated cargo requires temperature control regardless of weather conditions. Transporting this type of cargo requires appropriate installations, most often refrigerators, freezers, isotherms, delivery vehicles equipped with an insulated body, or refrigerated containers with a cooling unit. Refrigerated transport is very demanding and is often subject to strict sanitary conditions [21].

Trade in sensitive cargo, which requires an appropriate temperature during transport, has been growing steadily for over 20 years. This is due to many factors, including the increasing wealth of societies, which translates into an increased demand for a healthy and varied diet rich in products from around the world. There is also a noticeable positive correlation between income per capita and the size of refrigerated cargo in food imports [22].

The Combined Nomenclature (CN), introduced on 23 July 1987 by the Council of the European Communities, aims to collect and exchange statistical data on the external trade of the Community. In order to simultaneously meet statistical and tariff requirements, it replaced the existing Common Customs Tariff and the Nimexe nomenclature [23]. According to the CN classification, also adopted in the statistics of Polish external trade, the following cargo groups are classified as refrigerated cargo (numerical codes, such as (02) and (03), represent specific product categories under the CN classification system):

• Meat and edible products (02);
• Fish and crustaceans, mollusks and other aquatic invertebrates (03);
• Dairy products, birds’ eggs, natural honey, edible animal products (04);
• Vegetables and certain edible roots and tubers (07);
• Edible fruit and nuts, peel of citrus fruits or melons (08);
• Animal or vegetable fats and oils and their decomposition products; prepared edible fats; animal or vegetable waxes (15);
• Meat, fish or crustaceans, mollusks, or other aquatic invertebrates preparations (16);
• Vegetables, fruits, nuts, or other parts of plant preparations (20);
• Various food preparations (21);
• Non-alcoholic, alcoholic beverages and vinegar (22);
• Pharmaceutical products (30).

Considering the wide range of products classified as refrigerated cargo, this research focused on group 08 fruits and 07 vegetables, and the research was based on the analysis of fruit transport. Of all cargo requiring controlled temperature, 37% are frozen cargo, while 63% are refrigerated cargo [24]. There are five categories of refrigerated cargo with different temperature sensitivity [25]:

• Deep-frozen (−30 °C to −28 °C)—this category includes the lowest possible temperatures and the transport of seafood.
• Frozen (−20 °C to −16 °C)—this category refers mainly to the transport of meat.
• Chilled (2 °C to 4 °C)—medium temperature category; at this temperature, fruits and vegetables are stored, which ensures their optimal freshness.
• Pharmaceuticals (2 °C to 8 °C)—this cargo category refers mainly to pharmaceutical goods, e.g., vaccines.
• Bananas (12 °C to 14 °C)—this is a temperature range that ensures controlled ripening of bananas.

The research conducted in this article refers to the part of refrigerated cargo classified as food. National and EU regulations regulate the transport of food products. The main legal regulations include the following [26]:

• Act of 25 August 2006 on food and nutrition safety [27];
• Act of 6 September 2001 on road transport [28];
• Agreement on the international carriage of perishable foodstuffs and on the special equipment to be used for such carriage (ATP) [29];
• Act of 25 August 2006 on food and nutrition safety [30];
• Regulation (EC) No 852/2004 of the European Parliament and of the Council of 29 April 2004 on the hygiene of foodstuffs [31];
• Regulation (EC) No 853/2004 of the European Parliament and of the Council of 29 April 2004 laying down specific hygiene rules for food of animal origin [32];
• Regulation (EC) No 178/2002 of the European Parliament and of the Council of 28 January 2002 laying down the general principles and requirements of food law, establishing the European Food Safety Authority, and laying down procedures in matters of food safety [33].

The regulations of the international ATP agreement [34] govern the transport of goods requiring controlled temperature. The guidelines are divided into two groups: provisions relating to frozen or deep-frozen goods and goods that are neither deep-frozen nor frozen.

The regulations of the ATP agreement apply to the following [34]:

• Road and rail transport (also combined) using refrigerated, heated, or thermally insulated means of transport for the transport of food;
• Sea transport (distances not exceeding 150 km), if the delivery by land transport takes place without reloading.

They do not apply to container land transport without reloading when previously transported by sea. In the study, the authors analyze the external costs of transporting refrigerated cargo on a sea section longer than 150 km for which the provisions of the ATP agreement are not required. In transporting goods requiring controlled temperature not regulated by the ATP agreement, the User Requirements Specification (URS) is used, which is an agreement between the carrier and the customer. It specifies all parameters, levels, and instruments that should be implemented and observed during transport, such as the following:

• Temperature and humidity parameters that must be maintained for each product or product type;
• Mode of transport and/or vehicles to be used;
• Level of service required;
• Acceptable levels of risk for the product;
• Types of packaging;
• Type of temperature and humidity monitoring devices to be used and acceptable level of accuracy of those devices;
• Specific service activities (go/not-go decision-making for temperature exceedances, more complex analytical data collection, reporting requirements).

One of the methods of transporting temperature-controlled cargo is their handling in the sea–land system. Within the sea–land transport chains in the sea section, refrigerated cargo can be transported using a container ship and a specialized refrigerated ship. Research results present the economic aspects of choosing a ship—a classic container ship with connections for refrigerated containers and a specialized refrigerated vessel. According to the results of the analysis conducted and published by P. Cudina and A. Bezic [35], a refrigerated ship is more energy-efficient and can reach a higher speed, so the journey takes less time. On the other hand, loading cargo takes more time, and the insulation of the hold during transshipment operations can be damaged, especially during rain. A container ship can be loaded or unloaded faster and easier, but it has lower energy efficiency and usually moves at a lower speed, so transport takes more time. It should be remembered that the design of a container ship must be adapted to the transport of refrigerated containers:

• The holds must be equipped with electrical sockets for refrigerated containers.
• They must also be mechanically ventilated to remove waste heat from the refrigerated containers.
• The ship’s engine must cover the total energy consumption, including the refrigerated containers and the ventilation in the hold.

All of these factors affect the economy of the ship’s operation, which is why choosing the most optimal type is not easy [35] (pp. 129–141).

The storage of refrigerated cargo in seaports takes place in cold store warehouses or storage yards equipped with devices for connecting the power supply of a refrigerated container. The increased dynamics of the transport of refrigerated cargo in containers will force the development of logistics functions. Currently, managers of refrigerated cargo expect a new quality of services in the port and its surroundings [36], which can be ensured by activities carried out within efficiently managed logistics centers located near the seaport.

In the land section, refrigerated vehicles with controlled temperatures are often used to transport refrigerated cargo or tarpaulin vehicles without a refrigeration unit. The latter are used, for example, to transport lemons (less sensitive to higher temperatures) at lower temperatures to reduce transport costs.

As already mentioned, one of the methods of transporting temperature-controlled cargo in the sea–land system is using refrigerated containers, which are designed to maintain the appropriate temperature during transport. Refrigerated containers are insulated containers equipped with an integrated cooling unit for temperature regulation and control [37]. Temperature-sensitive goods require almost constant cooling to maintain their quality over time. Such a process is often called a “cold” supply chain or cold chain among practitioners. In the case of sea transport, this type of refrigerated container has become a standard solution. The integrated refrigeration unit cools the air that circulates thanks to two fans. Cold air flows into the hold from the bottom of the container through the floor profile, and warmer air is directed back to the refrigeration unit at the top, constantly pumping the cooled air through the container and around its contents. The temperature of the warmer air fed back to the refrigeration unit is monitored so that the refrigeration unit maintains the temperature of the cargo at the desired level. To ensure that the ambient temperature does not affect the load, the containers themselves are well insulated and painted with clear white paint to limit the impact of solar radiation on the temperature maintained inside them [37].

5. External Cost Analysis

This section presents a detailed analysis of the external costs associated with both road and sea–land transport options. The analysis follows a structured approach: we first calculate the total and average external costs for a certain type of cost (such as accident costs, air pollution, noise, etc.) using the methods outlined in the Handbook on the External Costs of Transport [19] and present these more general cost values in the first table. Next, we calculate and compare the external costs for road transport and the sea–land transport chain specifically for the Valencia–Warsaw route in the second table, highlighting the differences in costs between the two transport scenarios. This step-by-step approach allows for a clear and systematic comparison of the two modes of transport across the various external cost categories.

The analysis included eight types of costs: accidents, noise, environmental pollution, climate, congestion, energy production, habitat damage, and infrastructure [19]. We begin with the calculations of accident costs.

Accidents occur in connection with the use of any means of transport and cause significant costs, including material costs (e.g., vehicle damage, administrative costs, and medical costs) and intangible costs (e.g., shortened life, suffering, or pain). Using market prices to calculate material costs is possible, but these mechanisms will not work for intangible costs. In addition, part of the total costs of accidents is already internalized through, for example, insurance premiums or by adequately considering the degree of risk. Although there is no single comprehensive definition of external accident costs, they are often defined as the social costs of road accidents that are not covered by insurance premiums that consider the risk of accidents [19]. The insurance system determines the share of accident costs that are considered internal. Therefore, any costs covered by insurance are not considered external to the individual (which means that they are internalized). Costs that are not covered by insurance are external in nature. The total and average costs of accidents were determined using a top-down approach. (In the top-down approach, a complex problem is divided into smaller sub-problems, which are then analyzed one by one. In the case under consideration, we started with the total number of accidents and then assigned them to individual types of vehicles (road transport or sea transport). The average costs of accidents in road transport are the ratio of the total costs of accidents to the transport work performed. (

Table 2. Total and average external costs of road transport accidents.

Type of Transport	Total Costs EU28	Average Costs
Freight transport	Billion EUR	EUR-cent/tkm	EUR-cent/vkm
LCV	19.8	6.0	4.1
Truck	23.0	1.3	15.5
Total road freight transport	42.8

Source: own study based on [19] (p. 38).

)).

In road freight transport, trucks generate more than 53% of the total external costs of accidents in the EU, while LCVs generate 47%.

The total and average costs of accidents in maritime transport are listed below in

Table 3. Total and average external costs of maritime transport accidents.

Type of Transport	Total Costs EU28	Average Costs
Freight transport	Billion EUR	EUR/port call	EUR/1 million tonnes
cargo ship	63.3	318	36,524

Source: own study based on [19] (p. 39).

. Costs for cargo ships are presented per port call and per million tons.

Below, in

Table 4. Calculation of external costs of road and sea–land transport accidents.

Road Transport [Valencia–Warsaw]
	Transport Performance (thousand tkm)	Average Cost (EUR-cent/tkm)	External Cost (thousand EUR)
accident costs	429,000	1.36	5834
Sea–Land Transport—Sea Section [Valencia–Gdańsk]
	Transport (thousand tonnes)	Average cost (EUR/million tonnes)	External cost (thousand EUR)
accident costs	156	36.52	0.006
Sea–Land Transport—Land Section [Gdańsk–Warsaw]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
accident costs	54,600	1.36	743

Source: own study based on [19] (p. 39).

, the external costs of road and sea–land transport accidents for the Valencia–Warsaw route have been calculated. The sea–land transport chain has been divided into two sections: the sea section from Valencia to Gdańsk and the land section from Gdańsk to Warsaw.

The calculation results showed that the external costs of accidents for road transport are more than seven times higher than those that occur during the implementation of the sea–land transport chain. The sea section is characterized by a negligible level of external costs in terms of accident costs.

Another type of cost included in the calculation of external costs is noise. Noise can be defined as unwanted sounds of varying duration, intensity, or quality, which can cause physical or psychological harm to people [38]. Road noise is generally perceived as a significant nuisance, which is also associated with high costs. Noise emission in road traffic is a growing environmental problem due to the simultaneous trend of increasing urbanization and continuous growth of traffic intensity. Increased traffic intensity results in higher noise levels, while increased urbanization results in more people experiencing nuisance because of it. This relationship affects the forecast increase in road noise costs, even despite the implemented improvements in vehicles, tires, and road infrastructure in the scope of noise reduction [39].

The noise costs for road transport will be presented below. The noise costs related to maritime transport have not been included, as they were considered insignificant or non-existent in the adopted research methodology [19]. This transport takes place in uninhabited areas by design, and the noise emission coefficients for this type of transport are also relatively low. The thresholds above which noise is considered bothersome result from many scientific studies [40]. Reports of the European Commission from previous years specified the thresholds at 50, 55, and 60 dB. It should be noted that the choice of the threshold level significantly impacts the marginal noise costs (in the report [19], a threshold of 50 dB was assumed for cost estimation purposes).

Total and average noise costs are estimated using a bottom-up approach. The bottom-up approach involves projecting the most basic data, which are combined to make a higher-level module (For example, the primary input data were used to estimate the total and average noise costs: the number of people affected by noise by type of transport in the individual countries and the cost of irritation and damage to health per person. In this way, the total noise cost was calculated by type of transport and country. In the last step, the total noise cost was allocated to the individual vehicle categories using the established weights, and the total and average noise costs for the particular vehicle categories in the individual countries were estimated).

Two types of input values were used: the number of people exposed to noise for each mode of transport and the noise costs per exposed person (The noise classes to which people are exposed are classified in intervals, e.g., 5 dB. For each noise class and means of transport, the total number of people exposed is estimated. The second input value, the noise costs per exposed person, consists of two values: a nuisance value and a health value. The sum of the health and nuisance values gives the total noise cost per exposed person. This cost is multiplied by the number of people exposed to the relevant noise level, and the sum of these costs gives the total external noise costs for the appropriate means of transport. The average noise costs are estimated by dividing the total costs by the total transport efficiency (i.e., pkm, tkm, etc.)). The results are shown in

Table 5. Total and average external noice costs of road transport.

Type of Transport	Total Costs EU28	Average Costs
Freight transport	Billion EUR	EUR-cent/tkm	EUR-cent/vkm
LCV	5.4	1.6	1.1
3.5–7.5 t truck	1.0	1.2	4.0
7.5–16 t truck	1.8	0.8	5.7
16–32 t truck	3.0	0.4	6.5
>32 t truck	3.2	0.4	7.2
Total road freight transport	14.5

Source: own study based on [19] (p. 81).

.

After the implementation of restrictive regulations on noise emission reduction standards EURO 5 and EURO 6 for trucks in the EU, LCVs generate the highest average external noise cost. The external cost of noise for trucks per tkm decreases with the increase in the vehicle’s load mass and ranges from 0.4 to 1.2 EUR-cent/tkm. Below, in

Table 6. Calculation of external costs of noise in road and sea–land transport.

Road Transport [Valencia–Warsaw]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
noise costs	429,000	1.60	6864
Sea–Land Transport—Sea Section [Valencia–Gdańsk]
	Transport (thousand tonnes)	Average cost (EUR/million tonnes)	External cost (thousand EUR)
noise costs	no impact on society
Sea–Land Transport—Land Section [Gdańsk–Warsaw]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
noise costs	54,600	1.60	874

Source: own study based on [19] (p. 81).

, are calculations of external noise costs in road and sea–land transport for the Valencia–Warsaw route. The calculation has been made only for land sections.

Noise costs for maritime transport are usually considered negligible or non-existent, as they usually occur in sparsely populated areas, and the noise emission factor for this mode of transport is relatively low. However, the calculations indicated noise costs of over EUR 6800.

Air pollution costs are one of the most frequently analyzed categories of external costs. Unfortunately, few comprehensive international studies cover the entire impact path from emissions to direct air pollution impacts and generated costs. To estimate external costs related to air pollution from transport, the aforementioned European Commission report [19] analyzed four types of impacts: health effects, crop losses, material and building damage, and loss of biodiversity. Total and average costs related to air pollution are estimated using a bottom-up approach. There are two main types of input values: emissions and cost factors per tonne of pollutants.

Table 7. Total and average external costs of air pollution from road and sea transport.

Type of Transport	Total Costs EU28	Average Costs
Freight transport	Billion EUR	EUR-cent/tkm	EUR-cent/vkm
LCV	15.49	4.68	3.24
LCV—petrol	0.33	1.72	1.17
LCV—diesel	15.16	4.86	3.37
truck	13.93	0.76	9.38
Total road freight transport	29.42
sea vessel	29.00	0.4	-

Source: own study based on [19] (p. 50).

presents total costs and average costs per vkm and per pkm or tkm.

Total and average external costs of air pollution for individual means of transport used in road transport significantly exceed the costs generated by sea transport. A seagoing vessel generates 47% less external costs per tkm than a truck. The means of transport generating the highest external costs of air pollution is an LCV powered by a diesel engine.

Table 8. Calculation of external costs of air pollution in road and sea–land transport.

Road Transport [Valencia–Warsaw]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
air pollution costs	429,000	0.76	3260
Sea–Land Transport—Sea Section [Valencia–Gdańsk]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
air pollution costs	708,240	0.40	2833
Sea–Land Transport—Land Section [Gdańsk–Warsaw]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
air pollution costs	54,600	0.76	415

Source: own study based on [19] (p. 50).

calculates the external costs of air pollution in road and sea–land transport for the Valencia–Warsaw route.

The analysis shows that the external costs of air pollution for direct road transport in relation to transport carried out within the sea–land transport chain are comparable. For both variants, it is around EUR 3,250,000. The truck assumed in the calculation is characterized by a relatively low average cost of air pollution, which translates into a competitive amount of this cost to sea transport.

Another type of cost included in the calculation of external costs is climate costs. Because the effects of climate change are global and long-term and are characterized by risk patterns that are difficult to predict, determining these costs is highly complex. Research in this area that was conducted by, among others, D. Bernacki, Ch. Lis presents climate costs as one of the most essential types of external costs in land transport [41]. The transport sector causes CO₂, N₂O, and CH₄ emissions. All these compounds belong to the group of greenhouse gases contributing to climate change. For this reason, determining the costs of the impact of transport on the climate is extremely important (The emission of greenhouse gases into the atmosphere leads to global warming and climate change. According to IPCC estimates [42], establishing appropriate climate policy priorities is necessary for a significant increase in the average temperature on Earth to be expected by the end of the 21st century. Such a radical change will significantly and largely irreversibly impact ecosystems, human health, and societies). Climate change costs are defined as costs related to all the effects of global warming, such as sea level rise, loss of biodiversity, water management problems, frequent and extreme weather conditions, or crop failures.

Three input indicators were used to determine the costs: greenhouse gas emission factors of vehicle types, vehicle performance data, and climate change costs per tonne of CO₂ equivalent (Greenhouse gas emissions by vehicle type can be calculated by multiplying the number of kilometers of a specific vehicle type in each country by the vehicle emission factors (in g/km) for each of the different greenhouse gases (emissions of CO₂, N₂O, CH₄, etc.). To obtain the total costs of climate change, it is possible to sum up the greenhouse gas emissions using the Global Warming Potential (GWP) factor, resulting in the total greenhouse gas emissions in CO₂ equivalent. This result is multiplied by the climate change costs per tonne of CO₂ equivalent to determine the total climate change costs for the relevant vehicle types. The total costs are divided by the number of pkm or tkm to obtain the average climate change costs, depending on the vehicle type).

The cost factors (baseline values) for climate change costs by transport mode and vehicle type are listed below in

Table 9. Total and average external costs of climate change in road and maritime transport.

Type of Transport	Total Costs EU28	Average Costs
Freight transport	Billion EUR	EUR-cent/tkm	EUR-cent/vkm
LCV	13.17	3.98	2.75
LCV—petrol	0.71	3.76	2.56
LCV—diesel	12.45	3.99	2.77
truck	9.63	0.53	6.48
Total road freight transport	22.79
sea vessel	11.0	0.16	-

Source: own study based on [19] (p. 66).

. The table includes total costs and average costs per vkm and pkm or tkm.

As the above list shows, a seagoing vessel generates almost 70% less average external costs of climate change per tkm than a truck. An LCV generates the highest external costs of climate change, like the costs of air pollution.

The calculation results in

Table 10. Calculation of external costs of climate change in road and sea–land transport.

Road Transport [Valencia–Warsaw]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
climate change costs	429,000	0.53	2274
Sea–Land Transport—Sea Section [Valencia–Gdańsk]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
climate change costs	708,240	0.16	1133
Sea–land transport—land section [Gdańsk–Warsaw]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
climate change costs	54,600	0.53	289

Source: own study based on [19] (p. 66).

show that the external costs of climate change within the sea–land transport chain on the analyzed route are 37% lower than the analogous costs generated by direct road transport.

Congestion costs are another type of costs included in calculating external costs. Congestion is defined as a state in which vehicles are delayed during their journey [43]. Congestion costs arise significantly when an additional vehicle reduces the speed of other vehicles during the journey, thus increasing the travel time. Congestion costs can be determined based on the speed-to-flow ratio, for example, in an urbanized or intercity area. This approach does not apply to other modes of transport, such as rail or air, because they generally provide scheduled services and are planned based on the capabilities of the entire network and individual nodes. Another definition of congestion was proposed by T. Wilk and P. Pawlak [44], defining it as an external traffic jam, i.e., the resistance that vehicles impose on each other when the traffic flow volume approaches the network’s maximum capacity. Preventing this phenomenon and attempts to reduce it are very important due to the costs it generates. These are both costs related to vehicle users themselves, as well as costs affecting the natural environment. It is also worth noting that road congestion can also affect other external costs. Changing traffic intensity on roads entails, for example, different levels of pollutant emissions (local and global), road accidents, and, consequently, the external costs associated with them.

The total annual costs of road congestion in individual countries can be determined based on a bottom-up approach in terms of delay costs and deadweight loss. The estimation is based on the values of delay cost per vkm and deadweight loss per vkm, estimated for different types of road infrastructure. Costs of delays and deadweight loss were calculated for each type of transport and are presented in

Table 11. Total and average external costs of congestion generated by individual types of vehicles.

Type of Transport	Costs of Delays			Deadweight Loss
Freight transport	Billion EUR	EUR-cent/tkm	EUR-cent/vkm	Billion EUR	EUR-cent/tkm	EUR-cent/vkm
LCV	55.5	16.8	11.6	9.4	2.8	2.0
urban roads	46.5	39.6	27.4	8.0	6.8	4.7
intercity roads	9.0	4.2	2.9	1.4	0.7	0.5
truck	14.6	0.8	10.9	2.5	0.1	1.8
urban roads	11.6	2.5	34.1	2.0	0.4	6.0
intercity roads	3.0	0.2	3.0	0.5	0.0	0.5
Total	70			11

Source: own study based on [19] (p. 96).

.

Based on the above-mentioned costs,

Table 12. Calculation of external costs of congestion in road and sea–land transport.

Road Transport [Valencia–Warsaw]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
congestion costs *	429,000	0.90	3861
Sea–Land Transport—Sea Section [Valencia–Gdańsk]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
congestion costs	no impact on society
Sea–Land Transport—Land Section [Gdańsk–Warsaw]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
congestion costs	54,600	0.90	491

* The cost of congestion is the sum of the costs of delays and deadweight loss. Source: own study based on [19] (p. 96).

presents the calculation of external costs of congestion in the analyzed case for road and sea–land transport.

In the above calculation, congestion costs were estimated only for land sections. According to the methodology adopted in the work for sea transport, congestion costs are not determined. It is assumed that the sum of delay costs and deadweight loss is zero. However, considering the analyzed land section in sea–land transport, congestion costs are almost eight times lower than in the case of road transport.

All types of costs discussed above refer to the direct effects of undertaking transport operations. However, there are many other upstream and downstream processes directly related to transport, which also lead to negative impacts and generate external costs. Energy production, vehicle production, and infrastructure construction also lead to the emission of air pollutants, greenhouse gases, toxic substances, and other negative effects on the natural environment. By far, the most significant effects are emissions from energy production. These costs are directly related to transport activities and can be precisely estimated. The cost of energy production includes generating energy from various sources, which leads to emissions and other external effects. Extraction of energy sources, their processing (e.g., refining or electricity production), transport and industry, construction of power plants, and other infrastructure—all these processes lead to the emission of air pollutants, greenhouse gases, and other substances. Emissions during the production of energy sources are significant from the point of view of total external costs. The effects of energy production play an important role, especially in the case of means of transport powered by electricity, because energy consumption is practically emission-free [19,45].

The methodology for calculating energy production costs is the same as for air pollution costs and climate change costs. The estimation is based on the same cost factors shown in

Table 13. Total and average external costs of energy production in road and sea transport in inland waterways.

Type of Transport	Total Costs EU28	Average Costs
Freight transport	Billion EUR	EUR-cent/tkm	EUR-cent/vkm
LCV	3.79	1.15	0.79
LCV—petrol	0.22	1.18	0.81
LCV—diesel	3.57	1.14	0.79
truck	3.71	0.20	2.50
Total road freight transport	7.50
sea vessel	4.00	0.06	-

Source: own study based on [19] (p. 108).

.

The average external energy production costs for a seagoing vessel were determined at 0.06 EUR-cent/tkm and are more than three times lower than for a truck and more than five times lower than for an LCV. This shows that a seagoing vessel is characterized by a very low level of generating external energy production costs.

Table 14. Calculation of external costs of energy production in road and sea–land transport.

Road Transport [Valencia–Warsaw]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
energy production costs	429,000	0.20	858
Sea–Land Transport—Sea Section [Valencia–Gdańsk]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
energy production costs	708,240	0.06	425
Sea–Land Transport—Land Section [Gdańsk–Warsaw]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
energy production costs	54,600	0.20	109

Source: own study based on [19] (p. 108).

presents the calculation of external energy production costs for the analyzed case.

The calculation results showed that the external costs of energy production for the sea–land transport chain are almost 50% lower than those for direct road transport.

Another type of cost included in the calculation of external costs of transport is the cost of habitat damage. The transport sector has various impacts on nature, landscape, and natural habitats. The main effects described in the literature are habitat loss (loss of ecosystems), habitat fragmentation, and negative effects on ecosystems caused by the emission of air pollutants (resulting in, among others, a reduction in biodiversity). However, the negative impact of transport on nature and landscape is included in a few studies of external costs. The total and average costs of habitat damage are presented in

Table 15. Total and average external costs of habitat damage from road transport.

Type of Transport	Total Costs EU28	Average Costs
Freight transport	Billion EUR	EUR-cent/tkm	EUR-cent/vkm
LCV	4.4	1.35	0.9
truck	3.6	0.19	2.4
Total road freight transport	8.0

Source: own study based on [19] (p. 121).

.

The total and average external costs of habitat damage were estimated only for road transport. For this group of costs, society is assumed to have no impact on maritime transport.

Table 16. Calculation of external costs of habitat damage due to road and sea–land transport.

Road Transport [Valencia–Warsaw]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
costs of habitat damage	429,000	0.19	815
Sea–Land Transport—Sea Section [Valencia–Gdańsk]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
costs of habitat damage	no impact on society
Sea–Land Transport—Land Section [Gdańsk–Warsaw]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
costs of habitat damage	54,600	0.19	104

Source: own study based on [19] (p. 121).

presents a detailed calculation of the external costs of habitat damage.

The calculation results show that direct road transport generates over 7 times higher external costs of habitat damage than transport carried out within the sea–land transport chain. This high difference undoubtedly results from the lack of costs for the sea section, which is the main element of the sea–land transport chain.

Infrastructure costs are another crucial external cost related to the functioning of transport. For this analysis, a broad definition of transport infrastructure is adopted as the physical and organizational network that enables movement between different locations [46,47]. This definition, in addition to the physical infrastructure, also includes organizational aspects (e.g., traffic police, traffic management). Infrastructure costs can be defined as direct expenses plus financing costs or, from another perspective, opportunity costs associated with not spending the funds on more profitable purposes. Financing costs (or opportunity costs) are expressed as interest in capital [19]. Regarding expenses, a distinction can be made between investments (further distinguished into modernization and renewal costs) and operation and maintenance costs (divided into operating and maintenance costs). Infrastructure costs can be further classified based on the way they are affected by their use, i.e., the volume of transport [48]:

• Variable costs: costs that vary with the volume of transport while the functionality of the infrastructure remains unchanged; some maintenance and renewal costs fall into this cost category,
• Fixed costs: costs that do not vary with the volume of transport while the functionality of the infrastructure remains unchanged or costs that increase the functionality of the infrastructure; all improvement costs and operating costs are fixed infrastructure costs; some maintenance and renewal costs are (partly) fixed costs.

Average and total infrastructure costs are expressed in EUR 1000 per pkm, tkm, or vkm in

Table 17. Total and average external costs of road transport infrastructure.

Type of Transport	Total Costs EU28	Average Costs
Freight transport	Billion EUR	EUR/1000 tkm	EUR/1000 vkm
LCV	19.4	-	40.6
truck	42.8	23.4	288.2
Total road freight transport	62.2

Source: own study based on [19] (p. 43).

, which shows the total and average costs of road transport infrastructure.

The total and average external infrastructure costs are generated by both road and sea transport, as shown in

Table 18. Total and average external costs of maritime transport infrastructure.

Type of Transport	Total Costs EU28	Average Costs
Maritime transport	Billion EUR	EUR/1 passenger	EUR/tonne
sea freight transport	1.0	-	1.0
RoPax Ships/cruise ships	0.4	10.3	-
Total	1.4

Source: own study based on [19] (p. 43).

.

Based on these costs, external infrastructure costs were calculated for the analyzed section, as presented in

Table 19. Calculation of external costs of infrastructure in road and sea–land transport.

Road Transport [Valencia–Warsaw]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
infrastructure costs	429,000	2.34	10,039
Sea–Land Transport—Sea Section [Valencia–Gdańsk]
	Transport (thousand tonnes)	Average cost (EUR/tonne)	External cost (thousand EUR)
infrastructure costs	156	1.00	156
Sea–Land Transport—Land Section [Gdańsk–Warsaw]
	Transport performance (thousand tkm)	Average cost (EUR-cent/tkm)	External cost (thousand EUR)
infrastructure costs	54,600	2.34	1278

Source: own study based on [19] (p. 43).

.

After analyzing the calculation results, it can be concluded that direct road transport incurs almost seven times higher external infrastructure costs than transport within the sea–land transport chain.

Table 20. External costs for road transport on the Valencia–Warsaw route.

Road Transport [Valencia–Warsaw]
	Transport Performance (Thousand tkm)	Average Cost (EUR-cent/tkm)	External Cost (Thousand EUR)
accident costs	429,000	1.36	5834
noise costs	429,000	1.60	6864
air pollution costs	429,000	0.76	3260
climate change costs	429,000	0.53	2274
congestion costs	429,000	0.90	3861
energy production costs	429,000	0.20	858
costs of habitat damage	429,000	0.19	815
infrastructure costs	429,000	2.34	10,039
Total			33,805

Source: own study.

summarizes the partial analyses by summarizing the external costs for road transport and the sea–land transport chain.

Almost 1/3 of all external costs for road transport are infrastructure costs. This value shows a significant share of infrastructure costs in the overall structure of external costs. Considering the progressive internalization of infrastructure costs and the relative ease of estimating them, this value should decrease in the coming years. The estimated total external costs for the Valencia–Warsaw section amount to 33,805 thousand euros.

As mentioned earlier, sea–land transport was divided into two sections—sea: Valencia–Gdańsk (4540 km) and road: Gdańsk–Warsaw (350 km). The total external costs for the sea section amount to EUR 4547 thousand, as presented in

Table 21. External costs for sea–land transport—sea section of the Valencia–Warsaw route.

Sea–Land Transport—Sea Section [Valencia–Gdańsk]
	Transport Performance/Transport	Average Cost	External Cost (Thousand EUR)
accident costs	156 thousand tons	EUR 36.52/million tonnes	0.006
noise costs	no impact on society
air pollution costs	708,240 thousand tkm	0.40 EUR-cent/tkm	2 833
climate change costs	708,240 thousand tkm	0.16 EUR-cent/tkm	1133
congestion costs	no impact on society
energy production costs	708,240 thousand tkm	0.06 EUR-cent/tkm	425
costs of habitat damage	no impact on society
infrastructure costs	156 thousand tons	EUR 1.00/tonne	156
Total			4547

Source: own study.

. The total external costs for the land section amount to EUR 4302 thousand, as presented in

Table 22. External costs for sea–land transport—land section of the Valencia–Warsaw route.

Sea–Land Transport—Land Section [Gdańsk–Warsaw]
	Transport Performance (Thousand tkm)	Average Cost (EUR-cent/tkm)	External Cost (Thousand EUR)
accident costs	54,600	1.36	743
noise costs	54,600	1.60	874
air pollution costs	54,600	0.76	415
climate change costs	54,600	0.53	289
congestion costs	54,600	0.90	491
energy production costs	54,600	0.20	109
costs of habitat damage	54,600	0.19	104
infrastructure costs	54,600	2.34	1278
Total			4302

Source: own study.

.

Not only infrastructure costs but also accident costs, noise costs, and costs related to congestion constitute the largest share of external costs of the route under study. Additionally, the conducted analysis shows how many negative effects on the natural environment and human health and life are generated by road transport. The 350 km road section between Gdańsk and Warsaw generates a similar amount of external costs as the over 4500 km sea section between the port of Valencia and the port of Gdańsk. The total external costs for the sea-land transport on the Valencia-Warsawa route are presented in

Table 23. External costs for sea–land transport on the Valencia–Warsaw route.

Sea–Land Transport
External cost	thousand EUR	%
accident costs	743	8.4
noise costs	874	9.9
air pollution costs	3248	36.7
climate change costs	1422	16.1
congestion costs	491	5.5
energy production costs	534	6.0
costs of habitat damage	104	1.2
infrastructure costs	1434	16.2
Total	8850	100.0

Źródło: Source: own study.

.

6. Conclusions

In summary, the external costs of direct road transport on the Valencia–Warsaw route significantly exceed the external costs for the sea–land transport chain running along this route, as is also the case in other studies [49]. A comparison of the external costs generated in the transport service of cargo for the examined route is presented in

Table 24. External costs between Spain and Poland on the Valencia–Warsaw route.

Transport	External Cost (Thousand EUR)
road	33,805
sea–land	8850

Source: own study.

.

External costs in the case of sea–land transport are as much as 73.8% lower on the route under study than in the case of road-only transport. Internalizing at least some of these costs would significantly increase the competitive advantage of sea–land transport chains.

Table 25. The amount of social costs between Spain and Poland on the Valencia–Warsaw route.

Transport	Transport Price (Thousand EUR)	External Cost (Thousand EUR)	Total (Thousand EUR)
road	11,551	33,805	45,356
sea–land	15,679	8850	24,529

Source: own study.

summarizes all costs generated on the analyzed Valencia–Warsaw route.

Taking into account both the transport price and all the external costs analyzed on the Valencia–Warsaw route, the cost of sea–land transport is 46% lower than that of road transport. This not only indicates the high potential of sea–land transport chains but also the significantly lower environmental pressure of this type of transport. Taking into account the external costs of various transport solutions in the decisions of cargo operators significantly affects the cost competitiveness of sea–land transport chains in relation to direct road transport. The adoption of the social cost criterion in these decisions on the exemplary routes from Spain to Poland showed how the cost advantage of direct road transport (taking into account only internal costs) gave rise to the cost advantage of sea–land solutions.