Broadly speaking, the RoRo market encompasses four main sub-markets. Firstly, there is the deepsea segment which can be divided into car carrying trades and regular liner trades with RoRo-facilities. Secondly, we have the shortsea segment which can be divided into ferry transport for both passengers (with cars) and rolling freight on the one hand, and freight-only RoRo transport (including containers on mafis) on the other. These four sub-markets will be dealt with in this chapter.

The vehicle manufacturing industry nowadays produces about 65 million units per year and is characterized by chronic overcapacity (plant utilisation is estimated at some 73% of capacity on a global level, whereas according to industry specialists a level of 80% is needed in order to earn a decent profit) and chronically poor return on investment¹⁴. Moreover, just like many other industries, the vehicle manufacturing industry has witnessed an ever-increasing degree of concentration and globalisation over the last few decades. As a result, the industry nowadays counts just 15 (global) players. The top-seven manufacturers, many of which feature a whole myriad of different car brands in their portfolio¹⁵, accounted for 46.7 million vehicles in 2005 or nearly 75% of worldwide production (Table 23). It is assumed that by 2010-2012 the number of global vehicle manufacturers will have further reduced to 9 or 10.

Over the years the car manufacturing industry has witnessed important geographical shifts. While established production markets such as Europe, North America and Japan are nowadays indeed still dominant as far as output is concerned, emerging markets in Asia (China, India and Thailand), the Middle-East, South-Africa¹⁶, Eastern Europe, Russia or South-America have rapidly gained in importance in recent years. Many of these latter regions have welcomed substantial investments in large car assembly factories. As a result their importance for the worldwide car manufacturing market has increased significantly, and is expected to continue to do so in the years to come. As an example, production in South Asia is expected to more than quadruple from 1.3 million units in 2004 to 5.4 million units in 2020.

China is a very special case in point: its light vehicle production (including commercial vehicles) is expected to increase from about 5.2 million units in 2004 to no less than 12 million units by 2015 and 15 million units by 2020. As far as selling strategies are concerned, there exists a big difference between the local car manufacturers: on the one hand, Chinese manufacturers that are developing production lines are doing so independently, with the aim of exporting their products¹⁷. On the other hand, global manufacturers that have set up joint-ventures with Chinese manufacturers are (mainly) producing for the domestic market (Nightingale, 2006:37).

In view of the above, it is hardly surprising that the seaborne vehicle trade has over the years undergone a fundamental geographical shift as well. Whereas today the major seaborne car trades still are Japan-North America, Japan-Europe, Japan-Middle East, South Korea-North America, South Korea-Europe and West Europe-North America, seaborne trades are nowadays increasingly focusing on the developing economies and emerging markets mentioned above. Unsurprisingly, the increasing globalisation trend has also been accompanied by a significant increase in the seaborne shipments of new cars¹⁸. It is estimated that, from the total production volume of nearly 65 million new cars in 2005, about 20-25% (i.e. some 15 million units) ¹⁹ were exported by ship from their country of manufacture, with the remaining 75-80% sold locally or exported overland. Forecasts for 2015 indicate a worldwide production volume of some 80 million units with overseas exports accounting for nearly 20 million units (excluding transhipment). This figure includes about 16.4 million deepsea units and 3.2 million shortsea units.

The European automotive network, as depicted in Figure 3, demonstrates how the main axes of car assembly and supplier activities are increasingly being complemented by strong developments in the periphery²⁰. Figure 4 depicts a typical production process for CBU (Completely Built Up) cars.

To conclude, Table 24 provides an overview of the production and overseas exports of new cars for four selected regions in 2005. As this table indicates, both Europe²¹ and North America took the lead with a production of about 20 million new cars each that year. However, only a relatively small amount of their output was exported overseas. The opposite applies to Japan and South Korea, which produced (far) less new vehicles than Europe and North America but generated much bigger maritime export flows.

Nowadays new and second-hand cars are transported mostly by Pure Car and Truck Carriers or PCTCs, the first of which entered service in 1977 (some ten years later than the first Pure Car Carrier or PCC) ²². By mid-February 2007 the total fleet consisted of 580 vessels with a combined capacity of 2.64m CEU. Just like in many other shipping sectors, the average size of the PCTC fleet is getting bigger and bigger, as indicated by Table 25. Whereas the average PCTC had a capacity of just 4035 CEU in mid-1990, this figure had increased to 4552 CEU (+13%) by mid-February 2007. The lion�s share of the PCTC fleet is currently being deployed on liner-type services, i.e. involving fixed routes with a predetermined schedule. Moreover, there is nowadays an increased emphasis on logistics and supply chain management, so that car carrier operators are now required to do more than simply move vehicles between ports (Drewry Shipping Consultants, 2006:45).

At the end of 2006 the largest PCTCs in service had a capacity of 7200 CEU, only to attain the 8000 CEU barrier about half a year later. And the increase in size will probably not stop there: shipyards have reportedly been asked to consider building PCTCs of up to 11000 CEU capacity. While such giant ships would obviously result in significant cost savings on the sea leg (economies of scale), they could well pose serious problems for car terminals designed to handle ships with a maximum LOA of 200m²³. In fact, the growing shortfall in the number of ports that have the infrastructure for car capacity, including in Europe, constitutes an important problem for car carrier operators. As a result, some of them have taken control of their own destiny by taking a (significant) stake in RoRo terminals in strategic ports (Nightingale, 2007). The takeover of PSA HNN�s RoRo terminals by NYK in early 2007 constitutes a prime example of this. Hence, the parallel with the liner shipping industry, which has also seen increasing investments by shipping lines in (semi-) dedicated facilities in recent years (cf. supra), is striking.

As its name implies, the cargo of PCTCs is not limited to new or second-hand cars alone. Broadly speaking, the cargo carried by PCTCs on deepsea routes nowadays typically includes about 50% new or second-hand cars and 50% (better-paying) "high and heavy" or "out of gauge" cargo. The latter is mainly being carried on ballast legs (i.e. on their way back to the car loading areas) and can include a whole myriad of different cargoes such as agricultural machines, construction equipment, wind mills, helicopters, etc. On some trades (e.g. to the Eastern Mediterranean or West-Africa) vast volumes of second-hand cars are shipped.

From a technical point of view the PCTC with 10-14 cargo decks, of which 4 or 5 hoistable ones, is nowadays the true workhorse in the industry. Many of the recently built PCTCs have a ramp capacity of 150 tons or more, clearly reflecting the increasing importance of "high and heavy" cargoes. In contrast to PCTCs, the Pure Car Carrier (PCC) concept is slowly fading away, due to a number of reasons. Firstly, the clear height of their car decks is not high enough to accommodate mono-volume cars or SUVs, which are becoming increasingly popular among consumers. Moreover, PCCs are far less suited than PCTCs to carry the mix of cargo types mentioned above.

In order to keep up with an expected annual 2% growth in vehicle manufacturing in the coming decade (to reach an estimated 80 million units by 2015), many operators have invested heavily in the expansion of their ship fleets²⁴. At the end of 2006 about 570 car carriers with a capacity of at least 1000 car equivalent units (CEU) were being deployed on worldwide trade routes (never before was the fleet so big) for a combined capacity of about 2.6 million CEU. At that time, a further 160-180 units were on order for delivery during 2007-2010, for a total capacity of nearly 1 million CEU. In recent years prices for PCTC newbuildings have increased steadily: the typical price tag for a 6500 CEU unit ordered at established Japanese or South Korean yards nowadays reaches about USD 75-80 million, while this was just USD 55 million a few years ago. However, shipyards in emerging markets such as Vietnam (Vinashin) or India (Pipavav) nowadays offer more attractive prices.

Although representing some 35% of the current fleet capacity in CEU terms, it is generally assumed that the 1 million CEU order backlog, which is heavily focused towards 6000+ CEU vessels, will easily be absorbed by the market. Indeed, according to some industry observers the market is nowadays (i.e. early 2007) characterized by a shortage of car carrier tonnage of about 10%. This corresponds with roughly 50 ships in the 2000+ CEU segment²⁵. This tight supply situation has a number of important implications. Firstly, according to some industry observers, shipments of new cars from Japan or South Korea are nowadays being delayed by about 1 month on average. Secondly, the lack of capacity has resulted in the fact that many older vessels, which have reached an age at which they could be scrapped, are actually still trading. In fact, just one single car carrier went to the scrapping yard in 2005, while in 2006 no such ship was scrapped at all.

Because of this extremely low scrapping activity in recent years, the car carrier fleet nowadays counts more than 100 vessels of over 25 years old²⁶. Unsurprisingly, these ships are mainly used to transport second-hand cars, as they do not boast the necessary deck configuration or ramp capacity to carry a mix of vehicles and "high and heavy" cargo. Other important markets for older car carriers are the trades between Japan and New Zealand, between the Mediterranean/North Europe and West Africa, and between the Far East and Middle East. It is generally assumed that the older vessels will gradually be phased out as from 2010 onwards, i.e. after the large newbuildings slated for 2007-2010 delivery have poured into the market.

A final implication of the tight supply situation is the fact that charter rates for car carriers have increased significantly in recent years. Current 12-month charter rates for 5000-6000 CEU units are estimated to be in the USD 30,000-40,000 per day region, i.e. about double the rate level of three years ago. In fact, the increases in newbuild prices and charter rates have been much more pronounced than the increase in freight rates obtained from shippers/manufacturers who, following consolidation in the car manufacturing industry, nowadays have substantial bargaining power vis-à-vis operators²⁷. This fact, combined with spiralling fuel costs, has obviously put PCTC operators under significant pressure in recent times. Yet, as manufacturers are well aware of the current tight supply situation and high fuel costs, they are prepared to put freight rates and bunker clauses back up for discussion at the negotiation table. After all, for many of them ocean transport constitutes a vital part of their global supply chains. In fact, as a result of the tight supply situation at the end of 2006, some car manufacturers have signed longer-term contracts than usual with car carrier operators, in order to safeguard required space (Nightingale, 2007).

On the shipping lines side, the industry is dominated by about six major operators based in the Far East and Europe, the largest of them being Japanese NYK (Table 26). This carrier currently has an (owned or controlled) car carrier fleet of some 90 vessels for a combined capacity of some 400,000 CEU and shipped an estimated 3.2 million CEU during 2006, an increase of about 15% compared to 2005²⁸. Driving forces behind the growth in NYK cargo volumes were (and still are) the increasing interest from the North American market for fuel-efficient Japanese cars, as well as good sales in Europe and the Middle East. Apart from NYK, other big deepsea players are Mitsui OSK Lines, K-Line, Eukor, Wallenius Wilhelmsen Logistics and Höegh Autoliners (including Maersk tonnage). These six carriers nowadays directly own about half of the specialist car carrier capacity and operate more than 80% of global fleet capacity (when chartered tonnage is included). Moreover, each of these six companies is also heavily involved in the European shortsea trades (cf. infra). Table 26 also clearly illustrates the dominance of the Japanese "Big Three" (NYK, K-Line and MOL) on the car carrying market. The "Others" line in Table 26 mainly refers to tonnage suppliers. The main players in this segment are Cido Shipping, Ray Car Carriers, Gram Car Carriers, Zodiac Maritime, Vroon and Dyvi. Just like the big operators, these tonnage providers have invested heavily in newbuilding orders in recent years.

In order to supply their global networks, the large vehicle manufacturers are nowadays increasingly obliged to use the services of all these six independent companies, rather than appointing one or two single carriers to carry all their business. In fact, "established relationships from the past have given way to contracts based on the need to supply different trade flows with different carriers. These contracts often imply cargo sharing agreements between carriers" (Drewry Shipping Consultants, 2006:45).

Another major player in the car carrier industry is Italy-based Grimaldi, which is both an operator and tonnage provider. From a geographical point of view, Grimaldi is a special case in point in that it currently mainly focuses on the Mediterranean-North Europe, intra-Mediterranean, Europe-West Africa and Europe-East Coast South America trade routes. As such, it has not (yet) ventured outside the Atlantic for vehicle transport. Secondly, United European Car Carriers (UECC) and K-Line European Sea Highway Service (KESS) are peculiar cases as they are heavily involved in the shorter-haul feedering business for the main deepsea operators (see below). So also here a parallel with the container liner industry can be drawn.

From an operational point of view, in the mid-nineties a fleet of 25-30 ships was regarded as the threshold needed to have the scale and flexibility to serve the car manufactures. The general expectation is that this figure will have tripled to around 80 units by 2010. This probably constitutes an important reason behind the recent team-up between Höegh Autoliners and AP Moller Maersk, which involves a joint commercial management of their respective fleets. The combined Höegh and Maersk fleet consists of 55 + 12 vessels (including chartered tonnage), with a further 15 units to be delivered during the next couple of years. Höegh Autoliners carried about 1.8 million CEU in 2006, of which some 65% consisted of new cars. While the AP Moller Maersk Line fleet (consisting of small 2500 CEU and mid-sized 4800 CEU units) is currently chartered out to Wallenius Wilhelmsen Logistics, NYK and Mitsui OSK Lines, it is expected that most of the vessels will join the Höegh fleet once their charter commitments expire.

As far as the European market is concerned, maritime transport of cars is expected to increase steadily over the years to come, following stable or slightly increasing demand for cars in the mature Northwest European market coupled with strong growth in Russia²⁹, Eastern Europe and Turkey. This latter aspect is expected to boost the import flow of cars in ports in the Baltic and the Black Sea. It will also be to the benefit of intra-European shortsea carriers, because many ports in the Baltic or the Black Sea currently dispose of insufficient draught to handle the large PCTCs, such that they have to be served by transhipment via large ports in the Mediterranean such as, for example, Gioia Tauro. Similarly, Russian volumes³⁰ could be transshipped via Bremerhaven in Germany to, e.g., Kotka in Finland. Of course, this picture would change drastically when car manufacturers would decide to establish factories in Eastern Europe and Russia.

As outlined above, the leading deepsea car carrier operators are also heavily involved in the intra-European shortsea trades. Examples include United European Car Carriers (UECC) which is jointly owned by Wallenius and NYK; K-Line European Sea Highway Service (KESS); Autoliners SAS (a subsidiary of Höegh Autoliners); Euro Marine Carriers (in which Höegh has a minority share); and the European shortsea operations by MOL. This combination of deepsea and regional service provision is part of the general trend towards the "one stop shop/total service logistics package" which operators now have to provide to vehicle manufacturers to retain their business (Drewry Shipping Consultants, 2006).

In more mature European markets, for example the English Channel traffic, car carriers are nowadays facing very strong competition from traditional RoRo operators (such as Cobelfret from Belgium), although the RoRo operators� main business still remains the transport of trucks, trailers and mafis so their ships have relatively limited car capacities. Cobelfret is a particular case in point. Following the commissioning of the massive 4600 lanem PAULINE in 2006, coupled with a consolidation of its terminal interests (Cobelfret owns terminals in Purfleet, Killingholme, Rotterdam, Zeebrugge and Vlissingen), the company can remain ahead of competition from other RoRo operators, as well as feeder lines and shortsea car carriers (ShipPax Information, 2007).

Table 27 provides an overview of the main European ports handling new and second-hand cars. In some of these ports, car carrier operators have invested in dedicated hub terminals from which other destinations are feederised, although a typical roundtrip for a large PCTC nowadays still counts five or six ports in Europe. Apart from the ports listed in Table 27, other important car handling ports which receive regular PCTC calls are Livorno (Leghorn), Flushing (Vlissingen), Vigo, Valencia, Piraeus, Koper, Santander, Göteborg, Copenhagen/Malmö³¹, Marseilles, Pasajes, Almeria, Rotterdam, Ghent and Cuxhaven. Some of these ports are situated in close proximity to car assembly factories of leading manufacturers³².

From an operational point of view, the ConRo concept (which refers to the carriage of containers and RoRo cargo on one single ship) bears a big resemblance with container liner shipping. Cargoes transported include new and second-hand cars, trucks, "high and heavy" cargo and containers. Important shipping lines in the ConRo market segment are Grimaldi Lines, Delmas/OTAL and Nile Dutch Shipping (NDS).

In the past, the deployment of ConRo vessels was very popular on certain liner trades to the Middle East, West Africa, South America and Russia, where facilities to handle ships were rather limited in certain ports. As for today, the ConRo concept has almost completely faded away on the deepsea routes (e.g. to/from the Middle East). It does, however, still survive on certain Western African and South-American trades (e.g. Grimaldi), as well as in the Levant (Beirut/Tripoli). ConRo vessels are nowadays also being deployed on some shortsea routes, by such carriers as Transfennica (Spliethoff Group) or Cobelfret.

Most of today�s ConRo vessels were built in the seventies and eighties, so the fleet has reached a relatively high average age. Construction of new vessels has come to an almost complete standstill as, with the exception of Grimaldi Lines (Naples) (which also owns transatlantic ConRo line ACL), nobody seems willing to invest in new tonnage³³. This has led to a decreased presence of ConRo tonnage on liner trades.

Grimaldi Lines also has a number of dedicated terminals around Europe suited to handle ConRo cargoes, e.g. in Naples, Antwerp and Hamburg. To conclude, it should be mentioned that the United States (for their Naval Reserve Fleet) have absorbed quite a number of ConRo vessels from the market, to deploy them in times of international conflict.

Whereas in the past the main income of ferry operators was related to the transport of passengers (income could typically be broken down into 35-40% "passengers", 35-40% "duty free" and 25-30% "freight"), this picture has undergone a fundamental change in recent years. First of all, ferry operators have been and are still faced with increasing competition from low-cost airlines (which nowadays offer cheap flights throughout entire Europe) and passenger and freight services between the Continent and the UK via the Channel Tunnel. This was compounded by the abolishment of duty-free sales onboard ferries in Europe on 1 July 1999, which resulted in a substantial loss of passenger traffic. On the other hand, freight transport in Europe has enjoyed very high growth rates in recent years.

In view of the above, it is not surprising to see that many ferry operators have felt the need to substitute passenger space for freight space onboard their vessels. As a case in point, SeaFrance�s latest ferries nowadays carry about 120 trucks per voyage between Calais and Dover, compared to 40-50 for vessels of a previous generation. Another example is the Travemünde-Trelleborg service operated by TT-Line. After the abolishment of duty-free sales, TT-Line decided to order new ships with a higher focus on freight and with only very basic passenger facilities (in the past passenger facilities onboard TT-Line ships were much more developed). In addition, Stena Line decided to have its cruise-ferries on the Kiel-Göteborg link rebuilt, to allow for an increase in the freight capacity and a more flexible exploitation of passenger facilities (e.g. less crew in the Winter season and more crew in the Summer holiday season). Another case in point is ferry operator Finnlines, who is introducing ferries with a freight capacity of about 4200 lanemetre and several hundred passengers on the link between Travemünde (North Germany) and Finland. These ferries will temporarily be the largest ferries deployed on European trade routes.

Hence, the general tendency in today�s ferry market is clear - an increasing focus on freight (which has indeed now become the main revenue for most ferry operators in Europe) rather than passengers. Having said this, however, passenger transport still remains a very important business for certain ferry operators on certain markets. A prime example of this is the Sweden-Finland link. Another exception to the increasing focus on freight is formed by the "cruise-ferries" which only carry about 1300-1500 lanemetre of freight. As an example we can mention the cruise ferries of Color Line deployed between Kiel (Germany) and Oslo.

In addition to the traditional services in Northern and Western Europe, another market worth mentioning is the Mediterranean, which has witnessed a tonnage rejuvenation in recent years. The traffic is very much "North-South" oriented rather than "East-West", with large ferries being deployed on services linking e.g. Northern Italy with Sicily, Italy with Greece, and France with Corsica or North Africa. Service speed is a crucial issue in these markets, and several ferry operators are nowadays deploying conventional RoPax vessels with speeds of up to 30 knots. To some extent, this phenomenon of fast conventional RoPax ferries is being copied in Northern Europe. As a case in point, Color Line will soon be introducing fast conventional vessels replacing ageing passenger RoRo vessels on services linking Denmark with Norway.

Finally, other important ferry links are Germany/Sweden, Denmark/Sweden, England-Wales/Ireland, England/Scandinavia, Calais/Dover, Valencia/Barcelona to the islands and North Africa, Marseilles/Corsica and North Africa, Sicily/Sardinia, Greece, the Adriatic Sea and Tunisia/Algeria/Morocco. Generally speaking, all these markets are being characterized by an increasing focus on freight transport (and thus a reduced focus on passenger transport), less duty-free sales and the deployment of faster and more modern ships.

Table 28 provides an overview of world ferry traffic in 2005 for passengers and three types of cargo. As this table indicates, Europe accounts for about one third of global passenger traffic by ferries, but its share of commercial ferry cargoes is much higher. The Baltic is a very important market for car and bus traffic, while the North Sea area is the dominant geographical region for trailer traffic.

At 01/01/2006, the total number of ferries amounted to 1,162 units, with a combined capacity of 1.16 million passengers (of which some 320,000 accommodated in cabins), 266,210 cars or 769,210 lanemetres of commercial vehicles. The fleet had a combined Gross Tonnage (GT) of 12,816,377 GT and an average age of 21 years. As Table 29 indicates, the top-15 European ferry operators accounted for just over half the GT of the total ferry fleet.

The market for unaccompanied freight transport, the fourth of the RoRo markets discussed here, is "booming business" for all players involved in it, and for most geographical regions is being characterized by scale increases (larger vessels), a shortage of vessels and a rather old age profile of the fleet. Vessels active in these trades are only allowed to carry a maximum of 12 drivers.

The market for unaccompanied freight transport is of crucial importance to many ports in Scandinavia, through which substantial

volumes of paper and forest products from local manufacturers (such as Stora Enso, UPM Kymmene, SCA or Norske Skog) are exported. Many of these export cargoes are loaded on mafis and then transported via RoRo vessels to destinations all across Europe. However, in order to reduce their dependency on pure RoRo freight, many shipping lines also increasingly reserve space for containers, for example for northbound cargoes going back to Scandinavia (e.g. 600-700 teu on the new Transfennica vessels). Important shipping lines in the Scandinavian RoRo market include DFDS Tor Line, Transfennica (Spliethoff Group) and Finnlines.

Another major market for unaccompanied RoRo freight transport is the North Sea, for example from Benelux ports (Rotterdam (Europoort and Hoek van Holland), Flushing, Zeebrugge and Ostend) to ports along the Humber and Thames (Hull, Killingholme, Dartford, Purfleet, Dagenham and Immingham). Other important UK ports are the so-called "Haven Ports" of Felixstowe (Norfolkline), Harwich (Stena Line) and Ipswich (Ferryways). On the North Sea market, containerized cargo is playing an increasingly important role (e.g. Cobelfret already carries more containers than trailers on some of its North Sea services). Without a doubt, the container will increase its penetration on other trade routes than the North Sea as well.

In contrast to the two above-mentioned markets (i.e. Scandinavia and North Sea), the market between North Europe and the Mediterranean is a very difficult market for unaccompanied RoRo transport, due to the very heavy competition from road transport. As a matter of fact, volumes on the North Europe to Mediterranean trade are nowadays not yet sufficient for a cost-effective exploitation of services with a high frequency (which is a crucial aspect if one wants to compete head-on with road transport). As a result, services on this trade route are only viable if they enjoy financial support such as under the "Motorways of the Seas" programme of the European Commission.

Finally we have the intra-Mediterranean market, but this market is nowadays more focussed towards RoPax vessels, while trailer transport is only of secondary importance. In the intra-Mediterranean market, UN RORO offers very specific concepts between Italy and Turkey (2 RoRo sailings per day between Trieste and Istanbul).

Table 30 provides an overview of RoRo traffic handled in a selection of European seaports. The table was drawn from a large Eurostat database containing about 260 ports handling RoRo traffic, for a total throughput of 415 million tons in 2005. However, for the present Report we have limited ourselves to those seaports which handled at least 200,000 tons of RoRo traffic. This resulted in a total ports sample of about 160 individual ports spread across 22 different countries, as shown in Table 30. Their combined RoRo throughput amounted to 407 million tons in 2005, effectively representing 98% of the total RoRo throughput of the 260 ports in the Eurostat database.

As can be seen from Table 30, ports in the United Kingdom handled nearly 100 million tons of RoRo traffic in 2005, i.e. about one quarter of the total RoRo traffic handled in European seaports. Next, with a total traffic of more than 50 million tons, Italy is also a major import/export country for RoRo cargo in Europe, although none of its ports handled more than 8 million tons of RoRo traffic in 2005. Other important players on the European RoRo handling scene are Sweden, Germany, Belgium, Denmark, France, Greece, Spain, the Netherlands, Finland and Ireland, each handling more than 0 million tons in 2005. Between them, these 12 countries handled about 391 million tons of RoRo traffic in 2005, giving them a combined market share of nearly 95%.

On an individual port basis, the biggest RoRo port in Europe is Dover, with a total traffic of more than 20 million tons in 2005. This represents about 5% of the combined RoRo throughput of the 260 ports in the Eurostat database. Other major RoRo ports, handling more than 10 million tons per year, include Calais (France), Zeebrugge (Belgium), Lübeck (Germany), Immingham (UK), Rotterdam (the Netherlands), Trelleborg and Göteborg (Sweden). At the other end of the spectrum, more than 160 European seaports handled less than 1 million tons of RoRo traffic in 2005.

As Table 30 indicates, the Eurostat database makes a distinction between two kinds of RoRo cargo, namely mobile self-propelled units on the one hand and other (non-self-propelled) RoRo cargo on the other. These two categories correspond to Codes 5 and 6, respectively, as defined in Annex II of Council Directive 95/64/EC of 8 December 995 on statistical returns in respect of carriage of goods and passenger by sea, as modified by Commission Decision 2005/366/EC of March 2005 (see OJ L123 of 17/5/2005 page 5).

The first RoRo category (mobile self-propelled units) includes (1) road goods vehicles and accompanying trailers, (2) passenger cars, motorcycles and accompanying trailers/caravans, (3), passenger buses, (4) trade vehicles (including import/export motor vehicles), (5) live animals on the hoof, and (6) other mobile self-propelled units. The total throughput in the selected European ports amounted to some 243 million tons in 2005. The United Kingdom was by far the market leader with a cargo volume exceeding 50 million tons, followed by Sweden (26.8 million tons), France (22.5 million tons) and Germany (21.4 million tons). On an individual port basis, Dover was by far the biggest port, followed by Calais, Rotterdam, Trelleborg, Lübeck, Rostock and Rodby.

The second RoRo category (non self-propelled cargo) includes (1) unaccompanied road goods trailers and semi-trailers, (2) unaccompanied caravans and other road, agricultural and industrial vehicles, (3) rail wagons, shipborne port-to-port trailers, and shipborne barges engaged in goods transport, and (4) other mobile non-self-propelled units. Here, the total throughput reached about 172 million tons in 2005. The United Kingdom was again by far the market leader with a cargo volume of nearly 50 million tons, followed by Italy (32 million tons), Belgium (20.4 million tons), Germany (15.3 million tons) and Sweden (13.9 million tons). On an individual port basis, Zeebrugge was the biggest port for this type of RoRo cargo, followed by Immingham, Lübeck, London, Göteborg and Genova.