Recent developments and prospects at UK container ports

Table of contents

• Introduction
• Summary
• Table 1: Container Traffic at UK Ports, 1988 to 1999
• Coverage
• Trade
• Data
• The last 10 years
• Interpretation
• Containerisation
• Direct Trade & Transhipment
• Forecasts - Methodology
• European and World Forecasts
• Ports
• Throughput
• Ship Size
• Productivity
• Quay Length
• Quality of Service
• Dedicated Terminals/Berths
• Land Utilisation
• Automation
• Balance of Supply and Demand
• Inland Distribution
• Conclusion
• Further Work
• Glossary & Abbreviations
• References
• Major North European Container Terminals

Introduction

1. This paper forms part of the commitment made by the Department of the Environment Transport and Regions (now Department for Transport, Local Government and Regions) in Modern Ports: A UK Policy to have a clear picture of trends affecting the ports industry, and especially of the potential need for port investment including new development. It is the Department's policy that, for environmental and other reasons, ports, as with other transport modes, should make the best use of existing natural and man-made capacity in preference to new infrastructure. The paper concentrates on recent developments and likely future growth in the demand for container movements at UK ports and factors affecting the current and likely future capacity to meet that demand.

2. The paper is based on discussions with ports, shipping lines, consultants, academics and some internal DTLR research. It is intended to highlight important trends in containers handled at UK ports and the implications for future capacity requirements. The paper does not set out government policy on ports and port capacity. This is described in Modern Ports: A UK Policy, in particular section 2.4. Nothing in the paper should be construed as an official endorsement of any particular view or statement. It is also important to note that any proposed port project will be subject to the usual planning controls and procedures and that this paper will not affect or pre-empt any decision made by the Secretary of State in exercising his planning functions.

Summary

3. The paper describes the recent trends in container units handled at UK ports, and sets these trends in the context of developments in world trade, container shipping and the increasing competition between UK and European ports for transhipment traffic.

4. In the last few years UK container ports, and in particular Felixstowe, have experienced quite rapid growth (see table 1). Some of this growth is related to the performance of the UK economy, but there has also been a substantial growth in transhipment traffic. This reflects the competitiveness of Felixstowe and its use by major world shipping lines as an important European hub port. In 1989 Felixstowe already accounted for 35% of UK container movements, but by 1999 this had grown to nearly 41%, though its share of containers carrying UK deep-sea exports and imports has probably not grown.

5. All commentators expect growth in containers moved at UK ports to continue at rates well above UK GDP growth. Shipping lines will continue to serve UK imports and exports by direct calls, but there is more uncertainty about future levels of transhipment through UK ports. Competition for transhipment traffic from new and existing facilities on the Continent is likely to be intense. As much port infrastructure has a long useful life and usually has little alternative use, the case for investment in such infrastructure needs to take into account forecasts of demand well into the future.

Table 1: Container Traffic at UK Ports, 1988 to 1999

Source: Maritime Statistics 1999 (DETR, 2000) and previous editions

6. There is uncertainty about the timing of the requirement for new capacity. This is partly due to uncertainty about future growth of traffic, but also about the potential for productivity improvements at existing ports. If productivity can be significantly increased by adopting best practice methods and modern equipment and making better use of existing infrastructure, then the requirement for new capacity may be delayed for a little while. Alternatively, if productivity can only be increased at a rate slower than trade growth the requirement for new capacity is virtually immediate. The time required for planning and obtaining approval has an important impact on when additional capacity is likely to become available. UK ports generally perform well in international comparisons of aggregate measures of productivity, but shipping lines find that quality of service indicators show UK ports to be poorer than European or Far East ports. The paper does not come to a definite view on productivity and the trade-off with quality of service.

7. The paper also considers the inland mode of container distribution. Ports and shipping lines are keen to increase the rail share of inland distribution, but pointed to capacity problems on the rail network as a constraint on their plans. The 10 Year Transport Plan will address these capacity problems. Under present conditions, ports and shipping lines did not see a significant role for distribution by coastal services. This is reflected in the small number of units moved by coastal container or ro-ro services. The recently published Planning and Policy Guidance on Transport (PPG 13) sets out a range of policies to promote carriage of freight by rail and water, including freight handled by ports.

Coverage

8. The paper concentrates on container traffic. The main reason for doing so arises from the conclusions of a previous and very detailed study of demand and supply at UK ports, Port Development and Nature Conservation, published by the Royal Society for the Protection of Birds. This report concluded that not only is container traffic likely to grow significantly, but also the increasing size of container ships indicated that only a handful of ports would be capable of dealing with this growth. By 2010 there was a significant predicted shortfall in capacity for deep-sea container cargo. Some other types of cargo, notably ro-ro were also growing, but the port capacity constraints were less because there was a greater range of ports to cater for that demand.

9. Since the RSPB report appeared (published in 1997, but based on 1994 data) the trends in UK port traffic have reinforced the reasons for concentrating on container movements. Container traffic itself has grown faster than predicted, partly because of buoyant economic growth, but also because of increased transhipment movements, mainly at Felixstowe. It is now clear that there will a significant fleet of very large container ships serving the main routes, which can only be accommodated at a small number of ports. Ro-ro traffic has also increased, but a significant share of cross-Channel traffic has been taken by the Channel Tunnel. Cross-Irish Sea ro-ro traffic has grown rapidly. Other cargoes are not growing significantly or are even falling and there is little pressure on capacity, but this does not rule out the need for port expansion to cater for particular cargoes or at particular locations, for example to provide for more sophisticated storage and handling methods.

10. The paper concentrates on UK ports, but any consideration of their prospects must take account of the capacity of European hub ports competing for transhipment cargo. There are also wider developments in the world economy, in world shipping, and new methods and technology in terminal operation, which need to be considered.

Trade

Data

11. Container traffic at UK ports is recorded by DTLR statistics in number of units and in tonnes of goods. The number of units, full and empty, and the tonnage by size of unit is also known, but there are some uncertainties about the mix of 20' and 40' boxes at UK ports in the mid-1990s. It is estimated that the proportion of 40' boxes rose from about 35% in 1990 to about 45% now, but the proportion is probably higher in deep-sea trades. For some purposes the TEU measure ("twenty-foot equivalent units") is a useful indicator of demand. There is no direct link to customs statistics, so that making commodity data and the first origin or final destination of trade consistent with port data on container traffic is a complicated exercise, especially with the increase in transhipment and carriage of semi-bulks in containers on lightly loaded legs. Information from the new systems set up under the Maritime Statistics Directive has recently become available. This gives some insights into the movements of containers between different ports by different carriers, but because of its very recent inception, there is no time series of data.

The last 10 years

12. Over the period 1989 to 1999 the total number of container units handled at UK ports (foreign and coastal) grew by 61.7% or 4.9% pa, and the tonnage by 77% or 5.9% pa. The rate of growth has increased in the last few years - units were up 7.2% in 1997, 5.3% in 1998, and 5.2% in 1999 (see table 1 above). However, it is somewhat misleading to look at this overall total as the deep-sea (essentially outside Europe) and near and short-sea (European) container markets have different demand and supply characteristics. Demand trends, competition from other modes, vessel size, mix of ports and dwell times are significantly different between short and deep-sea. Coastal traffic is less than 5% of the total and falling. UK port figures suggest that the movement of deep-sea containers more than doubled in the last 10 years (average annual growth of 7.6%), whereas near and short-sea units grew by just 2.6% pa and 4% pa respectively (see Table 2 below).

Source: Maritime Statistics, 1999 (DETR, 2000)

Notes
[ 1 ]. Near-sea covers trade with ports in the Irish Republic, Germany, Netherlands, Belgium and France.

[ 2 ]. Short-sea covers trade with ports in Denmark, Scandinavia and Baltic, Spain and Portugal, Italy and other Mediterranean.
[ 3 ]. Deep-sea covers all other foreign trade.

13. However, because these figures do not identify transhipment of loaded and empty containers they can be misleading as to the final destination or first origin of trade. Unofficial sources suggest that containers with a deep-sea origin or destination (including transhipment) have grown much faster than wholly short-sea traffic. Ro-ro services have carried the majority of the growth in near-sea traffic between UK and the Continent.

14. The number of boxes handled by ports includes empty boxes. In 1999 empty boxes made up 18% of total boxes handled, a similar percentage to that for 1998, but well above that for the previous three years. The proportion of empty boxes is affected by the balance between exports and imports in containers on the major legs of UK trade. In recent years the imbalance has been particularly severe on trade links with the Far East. When measured in tonnes the imbalance is probably greater still. On deep-sea trades, it is estimated that the tonnage of UK unitised imports is 66% higher than unitised exports, and this gap has been widening. In 1988 imports were only 30% higher than exports. The actual imbalance on the mainline routes is affected by trade of the whole of North-West Europe, and the ability to fill empty containers with non-traditional cargoes.

15. According to Containerisation International figures, based on returns from terminals, the number of TEUs handled by West European container ports was about 36 million in 1997 or 22% of the world total. This share has been falling as economic activity and trade moves to the Far East. UK ports, taken together, handled more than any other country in Western Europe (source Containerisation International Yearbook, 1999).

Interpretation

16. UK container traffic does not exist in a vacuum, but also responds to developments in European and world trade. Many of the factors affecting UK container traffic are related to developments in the world economy and world shipping which would have the same magnitude of effect wherever the port is located. There are also specific factors affecting UK ports arising from recent history, geography, trading patterns, and institutions.

17. Growth in world trade has tended to outpace growth in the world economy. Figures from the IMF suggest that in the 1980s world trade in goods grew 1.2 percentage points pa faster than the world economy. During the 1990s this gap widened to 3.2 percentage points (source IMF Economic Outlook). These trends can be attributed to trade liberalisation and the development of global production strategies utilising different countries' comparative advantages. The part of world trade that is containerised is likely to grow faster than the rest of world trade as traditional industries with integrated production are replaced by dispersed component manufacturers and assembly of goods for final consumers. In spite of, or even because of, the growing use of the internet, consumers demand ever more sophisticated tangible products supplied from a greater variety of distant locations.

18. These demands are increasingly supplied from a chain organised by global companies, whereby, for example, the same components are made in many different countries and assembled into the same products in other countries. It has been estimated that over 50% of intercontinental containers are now shipped by large multinational or global companies. By 2010 this percentage is expected to grow to 70% (source Ernst Frankel in Lloyds Shipping Economist March 1999). This process has moved trade growth rates above those from traditional exchange of consumer or industrial goods manufactured in one country and supplied to other countries.

19. The big question for ports is whether these higher trade growth rates can be sustained. It is generally assumed that they will be as there is still scope for new markets, decentralising production in many activities and the powerful effects of globalisation.

Containerisation

20. During the interviews with representatives of shipping lines and ports the effect of the containerisation of break-bulk and semi-bulk cargoes on the growth of containers handled was discussed. There is some divergence of view as to the extent of this effect in the immediate past and the likely effect in the future. Given that the UK was one of the first (international) trades to be containerised it might be surmised that the extent of further containerisation must have been diminishing for some time. It is also difficult to establish from trade and port data a reliable and usable indicator of containerisation. There will be some effect from trades becoming containerised as land-side infrastructure comes on stream in areas such as India, China, South America and the former Soviet states. Containerisation of specific products is also a factor, for example timber products and refrigerated goods. It is noticeable that new container ships have a large number of reefer slots. These slots will be competing with air cargo and specialist reefer ships. The carriage of cars in containers is also a factor. This could be an important market where specialist, high specification and customised cars are involved.

21. Increased containerisation may also be a factor where semi-bulk goods, normally carried in bulk mode, are instead carried in containers. Such goods are attracted by the low freight rates for what would otherwise be empty boxes returning to load areas. Given the recent large imbalances on certain trades, rates on export boxes from the UK have been reported at very low rates. The carriage of unfamiliar cargo is clearly of importance to the overall profitability of shipping lines. However, in terms of port throughput and container handling there appears to be little, if any, impact. In the UK, box movements are driven by imports.

Direct Trade & Transhipment

22. World container traffic, as measured by the number of TEUs handled at ports, has been growing at 8-9% pa over the last decade. This growth rate is greater than the growth in the number of units because of the gradual increase in the proportion of 40' containers. Also 40' boxes tend to handle bulkier, but not necessarily heavier items. The growth rate in units is also greater than the tonnage or real value of cargo recorded in trade data, because of the increase in multiple handling of containers by different ports across the world. As shipping lines deploy larger ships on the main trade routes, often in consortia or alliances with other lines, it is either practically impossible or economically unattractive for them to call at as many ports as in the past. Hence the increase in transhipment between hub and spoke ports. For some major container ports such as Colombo and Algeciras transhipment traffic makes up 70-80% of their total throughput. It has been estimated that 26% of world port movements are transhipment (Cargo Systems March 2000).

23. Container volumes at UK ports are, like world volumes, driven by general economic and shipping factors. In the absence of other factors a rule of thumb used in the industry suggests that for the UK, direct cargoes grow at something like twice the rate of growth of GDP. However, there are some important historical, institutional and other factors that need to be taken into account. These factors are particularly important where they affect transhipment cargo.

24. Transhipment takes various forms. The traditional and most easily identifiable occur where export cargo is taken by feeder vessel or cargo ferry from one country to another, usually in the same continent, for onward shipment on another vessel to a third port usually in another continent. Import cargo follows the reverse process. This type of transhipment can be controlled by either the shipping line or by the shipper. Line transhipment arises from co-ordinated schedules of mainline and feeder vessels controlled by the line. Shipper transhipment is more opportunistic, exploiting the services of different lines or modes to reduce transit times or costs. A different form of transhipment - relay - is wholly controlled by the individual shipping line, consortia or alliance concerned. This involves cargo carried on one main line vessel relayed to another main line vessel at a hub port. An example would be cargo from East Africa to North America, being carried on a mainline vessel on the trade leg from East Africa to Europe, and then transhipped to another mainline vessel on the trans-Atlantic trade leg. Co-ordination of sailing schedules and concentration of calls at one port are needed to make this operation efficient and attractive to shippers. While all transhipment is more sensitive to port costs and efficiency of handling than direct cargo, relay traffic is particularly sensitive, as the range of ports which can act as a hub can extend over a wide area. An alternative description is to distinguish natural and discretionary transhipment. Natural transhipment occurs where boxes are incidental to a shipping lines' main origin and destination cargo, with feeder vessels supplying specific markets. Discretionary transhipment occurs where boxes are moved on main-line vessels as part of a major operation organised by the line, but which could be handled at a selection of ports.

25. In the late 1980s 10-15% of UK container trade was being transhipped at European ports and very little, if any, European trade was transhipped at UK ports (source Transhipment of UK Deep-Sea Trade 1976-1987, Department of Transport & British Ports Federation (1990)). Since then the position has changed radically. It was suggested to us that transhipment of UK trade has declined in relative, and possibly absolute, terms. However, transhipment of European and intercontinental trade at UK ports has grown very rapidly.

26. Transhipment at UK ports is not recorded in DTLR statistics. According to the port, transhipment at Felixstowe amounts to 35% of throughput. Mediterranean Shipping Company use Felixstowe and Antwerp as hub ports for trade with North West Europe, with Antwerp being the main import port and Felixstowe the main export port. Maersk Sealand also tranship at Felixstowe, though their pattern of trade is rather different, using feeder vessels rather than main line vessels. Transhipment at UK ports other than Felixstowe is thought to be much less significant. Southampton tranships some trade for southern Europe.

27. A plausible interpretation of the growth in transhipment cargo at UK container ports is that they are catching up with a natural level of transhipment that would be expected for a country with the largest import market of unitised cargo in Europe. An illustration of the size of the UK deep-sea container market can be seen in chart 1 below. It shows the percentage of unitised deep-sea cargo taken by different countries in Western Europe. The UK has the largest import market. Shipping lines serving routes between Europe and deep-sea locations would therefore want to call in the biggest market for direct cargoes. Other things being equal, this would make the UK an attractive base for transhipment.

Source MDS Transmodal (unpublished), estimated on the basis of unitisable cargo traded by each country.

28. Most commentators believe that the various deregulation and privatisation measures affecting ports in the last 15 years appear to have significantly improved the competitiveness of UK ports against their European rivals. There is no publicly available and consistent data on handling costs, but conversations with those in the industry suggest reductions in handling charges of the order of 50% in nominal terms over the past 20 years. Large reductions in handling charges also seem to have occurred at European ports. It was suggested to us that state funding of continental ports causes some distortion of competition with UK ports. On the other hand there is a view that where this funding is concentrated on infrastructure rather than port operation there is less effect on competition for transhipment traffic.

29. The UK also has a favourable geographical location to supply the transhipment needs of other countries not well served by the larger deep-sea vessels, for example Ireland, Scandinavia and Iberia.

30. Transhipment is also different from direct trade in that it creates different demands on the capacity of terminals. Transhipment across the quay removes the requirement for inland distribution, but terminals with high proportions of transhipment have other operational requirements (see discussion in paragraph 52 below).

31. Transhipment itself is likely to grow. The physical size of main-line ships is still increasing, so that navigational limitations and port infrastructure requirements dictate that fewer ports can cope with the larger vessels. The economics of the operations of larger ships also makes it more likely that there will be fewer port calls in each loading/unloading area.

32. In the absence of detailed data interpretation of transhipment is not straightforward. Nevertheless the growth in containers handled at UK ports has been significantly affected by the growth of transhipment, and in particular relay transhipment. Any serious analysis of the demand for container port capacity cannot ignore the transhipment factor and must make some assumption about its future levels when making forecasts.

Forecasts - Methodology

33. The DTLR does not make its own forecasts of port traffic, but is aware of forecasts made by others of container traffic at UK ports. They are often related to proposals or plans for new container port infrastructure. The methods for generating the forecasts appear to vary, although the results lie within a fairly narrow range. The central case for assessment is about 4-5% pa over the medium term, with the possibility of faster growth in the immediate future. Some forecasts also investigated a range around a central growth rate. This forecasted growth of 4-5% pa is similar to the average experienced over the last 10 years, although slower than in the immediate past.

European and World Forecasts

34. A comparison of forecasts for UK and European container traffic can be made. The Rotterdam port website has a summary of 2020: Integrated Projections for Port and Industry. There are two scenarios based on macro-economic developments with the titles "Global Competition" and "Divided Europe". The former assumes favourable economic developments, especially trade growth due to increasing international specialisation and competition. The latter has much lower trade and economic growth due to regulation and lack of competition. The forecasts of container traffic at Rotterdam based on these two scenarios are summarised in table 3 below.

Source: 2020 Integrated Projections for Port and Industry (Rotterdam Port)

35. The forecasts appear to assume that transhipment containers and containers for onward distribution in Europe grow at about the same rate. Modal split of onward distribution is assumed to favour inland waterways and rail at the expense of road.

36. A recent article by Martin Stopford in Containerisation International (January 2001) noted that the consensus among analysts is that world container traffic would grow at 5-6% pa. This is based on the continuation of trends noted above, ie. the broadening of the trade matrix and the sophistication of consumer demand.

Ports

Throughput

37. The major UK container ports are at Felixstowe, Southampton, Thamesport, Tilbury and Liverpool. These handle nearly all the UK's deep-sea containers (and some short-sea). Containers handled at other ports like Hull, Tees, Immingham, and Belfast are nearly all short-sea or coastal movements (see table 4).

Table 4: Container Movements at UK Ports by Main Port 1989, 1994, 1999

Source: Maritime Statistics, various issues (DETR)
Note Medway includes Thamesport, and London includes Tilbury

38. In North West Europe the largest players are Rotterdam, Hamburg, Antwerp, Bremerhaven and Le Havre (see table 5).

Table 5: Container Movements at Major North West European Ports

Source: Cargo Systems July 2000

39. In addition to existing ports there are a number of proposed or planned developments in the UK and North-West Europe. Some of these are being built, while others are being planned or under consideration. There is some uncertainty as to whether all proposals will go ahead as planned. Extensions to existing sites may be delayed or brought forward as demand or commercial considerations dictate. Table 6 summarises the expansion plans at UK ports. Note that these figures do not include increased throughput that could be achieved on existing land and quays.

Table 6 Potential Capacity Expansion at UK Container Ports

Sources: Environmental Statements, P&O Ports website, Containerisation International (October & December 2000), Scapa Flow Container Transhipment Terminal

Ship Size

40. A major consideration for modern container ports is the degree to which they can accommodate the large 7,000 TEU vessels currently being introduced on the major east-west routes and the prospects of even larger vessels in the near future. It was pointed out to us that 25 years ago the largest ships were about 3,000 TEU. Until 1988, all container ship owners chose to limit ship parameters to that compatible with the Panama Canal. The decision to build ships beyond the Panamax dimensions led to a rapid "catching up" in optimum ship size. Even so, the maximum size of container ships has increased two or three-fold but trade has grown probably at least twice as fast. Therefore larger ships could be justified on trade grounds (ie. maintaining the same frequency and geographical coverage). However, port restrictions become more important.

41. Size has a number of important effects on port operation. The major one is draft on the channels to the berth and at the berth. Given the need to keep to tight schedules, container ships, unlike bulk ships, could not be expected to wait for tides. Larger ships could be built with shallower draft, but this would have significant fuel cost penalties.

42. The length of ships has implications for quay length, and their beam, or number of rows of containers, has implications for the outreach of cranes. Total number of TEUs also affect the amount of terminal space required to stack containers for loading and unloading.

43. Given these size considerations the number of ports that can handle the largest ships becomes more limited or requires major dredging and port infrastructure developments to cater for them.

Productivity

44. One of the main points of interest in this study is the amount of capacity available at existing and planned terminals. Capacity cannot necessarily be equated with current throughput or the application of current productivity levels. This topic produced some divergent views on both the measures of productivity and, on any one measure, what productivity levels are possible to achieve in general and at specific terminals. A model, or models, that directly linked the inputs or factors of production at each container port - labour, capital equipment and land - to its outputs could give some insights into productivity performance. However, this approach is not generally used in the industry, but rather container terminal management is very much a practical science driven by experience rather than theory.

45. The measures used by the industry tend to be either aggregate comparisons of throughput or storage per unit of input or measures relating to the quality of service to the shipping line or haulier. The productivity and quality measures can sometimes be complementary, for example crane productivity and on-berth times. But they can also be in conflict, for example efficient use of storage against speed of access to boxes for hauliers.

Quay Length

Source: Based on Containerisation International Yearbook and other data. Derived units are TEUs divided by 1.5.

50. Interpretations of the differences in productivity need to be made with care. Low levels of throughput per quay length may not be an indication of poor productivity as such, but rather of other short-term factors, such as opening of additional facilities ahead of demand for their full capacity. Also, redundant capacity may be counted in the quay length, so depressing the measure of productivity.

Quality of Service

51. The aggregate productivity of the use of the quay length is a helpful measure, but can be achieved in different ways. A combination of crane productivity, number of cranes in use per metre of quay, hours of working, efficiency of berth use and other factors may come into play. Shipping lines tend to concentrate on measures such as container lifts per hour as this directly affects the amount of time a ship spends on the berth. On this measure, comparing the same ships in the same service, Felixstowe had a somewhat poorer performance than Rotterdam, and significantly worse than some other terminals in the Far East. One port consultant stated that 40 crane moves per hour are possible at some North American ports compared with about 20 currently at Felixstowe and 30 at Rotterdam and Antwerp. Felixstowe's relatively poor crane productivity is magnified by apparently more idle ship time in port when the movement of containers has finished. The cause or responsibility for this idle time is not clear.

52. It is not immediately obvious why Felixstowe's productivity record should be so apparently good at the aggregate, at least when compared with other European terminals, but poor when judged at the micro level. It might be something to do with hours of operation, whereby longer hours make up for slower working or number of cranes and other equipment employed per metre of quay. Mix of cargo may also be relevant. The proportion of container moves associated with transhipment may also affect the performance of ports on standard productivity measures. It is not possible to come to a definite opinion on this question. One view is that terminals with high levels of transhipment could achieve apparently much higher across the quay productivity than terminals with mainly direct trade because (effectively) the same container is counted twice. However a major transhipment operation has different but very challenging operational requirements. In normal import/export operations all inbound and all outbound containers can be stored in separate areas of the yard. Transhipment containers complicate both yard and ship operations as they are both inbound and outbound containers. Experience suggests that transhipment containers do not stay in the yard any less time than imports or exports.

53. A more general explanation of the variation in rankings shown by port productivity measures might arise from the way in which high throughput per length of quay is achieved at the expense of delays to ships. Given that ship arrivals are unlikely to be evenly spaced throughout the day or week, it is not easy to match capacity to throughput. Congestion at the port can lead to costly delays to ships. For example at Rotterdam, where because of delays to Grand Alliance ships, services had to be diverted to Antwerp. Charleston had high productivity once a ship was on berth, but some lines found that Savannah had fewer delays. These comments suggest that high terminal productivity might occur at the expense of delays to ships. For the shipping lines and ports there is balance between capacity utilisation (and hence unit cost of handling) and quality of service. This raises the question of whether dedicated berths or terminals might be attractive to shipping lines.

Dedicated Terminals/Berths

54. The interviews with representatives of shipping lines discussed whether they would prefer to have dedicated terminals or berths. Some said that they would prefer to have dedicated facilities, and claimed that where this is the case, they had a better service, because they are more in control of the service. They also said that they treated their own terminal operations as part of the overall transport chain rather than as a separate profit centre. Other lines could see advantages, but they pointed out that a dedicated terminal would only work where the line, or its partners, could make full use of the facility. Dedicated berths within a common user terminal might be an attractive alternative, avoiding the risks of taking on a complete terminal. All the lines would be concerned if there were further concentration of container terminal ownership.

55. The advantages and disadvantages of dedicated facilities compared with common-user facilities have been debated in the academic literature. A recent article by Hugh Turner, "Evaluating Seaport Policy Alternatives" (Maritime Policy and Management July 2000) reports the results of applying inventory theory to a simulation of terminal performance at the Port of Seattle with different ownership structures of the terminals. Turner points out that when evaluating the performance of a container port it is essential to recognise the interdependence of terminals and carriers rather than treating the seaport as a collection of independent terminals and carriers. Theory suggests that where the variation in demand of the different carriers is independent, common-user terminals can have lower total carrier cost than a regime that restricts specific carriers to specific terminals. These are aggregate results, and it might be the case that individual carriers would be better off with a dedicated terminal even though that imposes costs on other carriers.

56. Turner points out that though the common-user terminal may be the most desirable outcome, pressures from the larger carriers may force ports to make dedicated facilities available. He also notes that the scale of new developments makes joint funding between lines and terminal operators an attractive proposition, but the lines' price is exclusivity. Competition issues are also relevant, especially if terminal capacity is constrained. In these circumstances exclusivity might lead to higher costs and poorer services for smaller lines.

Land Utilisation

57. Efficiency of yard storage is an important, though possibly secondary issue in a European context. In the short term, storage area can limit effective capacity of the terminal and affect overall efficiency. At the aggregate the measures normally used are TEUs per hectare, turnover rate, and dwell time. In a storage context, TEUs are the relevant measure rather than units. Larger storage rates can be achieved by storing containers in higher stacks (5 or more high) and the application of automation methods. However, high stacking also has implications for time and cost of accessing containers.

58. The turnover rate is interpreted as the number of times each slot of storage capacity is used each year. Increasing the turnover rate can increase the throughput of the storage yard on the same land area, but to a large extent turnover rates and stack heights are in conflict, as the higher the stack the more difficult it is to access containers within the stack. This gives rise to wasteful, or 'unproductive moves.' The trade-off between stack heights and turnover can be influenced by management action and application of new technology.

59. Research carried out at the University of Wales makes various comparisons of the efficiency of land use at terminals in different parts of the world. Achieved storage capacity per hectare and turnover rates at Asian terminals are said to be much higher than in Europe and North America (see table 8 below). These differences may in part be explained by a different mix of traffic (including higher levels of transhipment), shortage of land and cultural factors, but the results do suggest there may be some room for improvement in Europe. This would be especially the case if land for storage is in short supply. However, it should be noted that some European terminals do appear to perform much better than the average.

Source Chen, Land Utilisation in the Container Terminal, Maritime Policy and Management October-December 1998. Data from 20 Asian terminal operators, 12 West Europe and 23 North America from Containerisation International Yearbook

60. One other important consideration affecting the potential capacity of terminals is the quantity and quality of the land-side access. If terminals served a large inland market, it is possible that the concentration of inland distribution moves around a port might become a constraint on size. Thus it is difficult to envisage one container hub port serving the whole of North-West Europe. It appears unlikely that land-side constraints affect UK ports at present.

61. Internal DTLR estimates of the proportion of traffic in the vicinity of Felixstowe which is generated by the port (these include Felixstowe's ro-ro as well as container traffic) have been made for 1998. On the A12 and A14 to the south and west of Ipswich, about 60% of the heavy goods vehicle traffic (10% of the total traffic) is port related. This traffic quickly disperses over the network. For example, to the west of Peterborough, it is estimated that the Felixstowe-related traffic only contributes about 20% of the goods vehicle flows and 4% of the total flow. The weekday peak hour flow at the port gate is about 400 vehicles per hour.

Automation

62. Reports and proposals by consultants, and comments by those interviewed, suggest that there is scope for further automation and different working practices at terminals. Perhaps not surprisingly the consultants who are promoting these new ideas are confident that major gains in productivity and savings in costs could be achieved. Some of the port operators are rather more sceptical that such gains could be achieved or are in fact needed. It was said that container terminal operation is hardly "rocket science". Productivity is improved in a piecemeal fashion. Others pointed out that previous attempts at automation had not been very successful. Automation might have been a response to constraints on the use of labour that no longer applied. Greater flexibility in the use of port labour reduced the incentive to implement full automation.

Balance of Supply and Demand

63. The various analyses and projections of demand and supply produce estimates of current and future capacity utilisation. There is agreement that new capacity will be required, though the suggested timing of such additions is variable. Most analyses of the balance do not state what would be a feasible level of utilisation of capacity. Due to anticipated and random fluctuations in demand it might be the case that 100% utilisation is not actually possible without unacceptable delays to ships. A further consideration is that competition between ports is likely to be greater where shipping lines can switch between ports with spare capacity. At the same time there is a balance to be struck between capacity and level of service. More capacity might enhance the level of service to shipping lines, but it increases costs to port developers and also has environmental impacts.

64. For deep-sea container traffic it is generally accepted in the industry that the main constraint is the small number of current terminals which can handle the larger vessels which are expected to dominate these trades. Vessels of 7,000 TEU with 14-metre draughts will shortly become the usual unit for the large operators on the major routes. Major lines are taking control of networks and alliances. They need ports to accommodate the most efficient ship size. These ports will also dominate the transhipment trade that arises from the larger vessels.

Inland Distribution

65. The split of rail and road traffic at the main UK container terminals is estimated to be approximately 25% rail and 75% road, but this varies by port, with Southampton having the largest rail share. Onward distribution by sea in the UK is insignificant. Most of the container movements through UK ports described as coastal are cross-Irish Sea services where competing land base modes are not relevant.

66. Ports are keen to increase the proportion of rail traffic through their terminals. They have recently completed, or have included in their plans, infrastructure to deal with more rail services within their terminals. Some lines have inland distribution depots connected to ports by regular and frequent rail services. However, there is a general view among shipping lines and ports that there are capacity constraints on the rail network that would hamper increases in carrying containers by rail. For some terminals there are constraints in the connection with the main rail network, for example Thamesport, or constraints in the nearby network, for example the London area. Certain pinch points on the rail network affect the ability to run scheduled container services from any port. There is also a general perception that passenger services are given priority where there are conflicts in the use of the same capacity.

67. The Government's strategy, described in Transport 2010 - The 10 Year Plan, sets out a £60 billion investment plan for rail of which £4 billion is to be allocated, specifically, for freight projects. The Strategic Rail Authority (SRA) will work with Railtrack and others to bring about the necessary improvements to the network to achieve the government's objective of 80% growth in rail freight by 2010. A key part of the SRA's strategy is concerned with links to ports and the Channel Tunnel to ensure that there is sufficient capacity to accommodate the increase in traffic flows and that the loading gauge is appropriate to facilitate the movement of the taller containers and swap-bodies currently being introduced.

68. Planning and Policy Guidance 13 on Transport, issued in March 2001, sets out a range of policies to promote carriage of freight by rail and water. This includes protecting sites and routes, both existing and potential, to develop infrastructure including inter-modal freight interchanges; and locating freight-generating developments where they can be served by rail and water. Development plans set out policy and proposals for ports at the local level, and regional planning guidance provides a strategic steer on the role and future development of ports.

69. None of those interviewed are optimistic that coastal shipping could make a significant impact on either shipping containers or as ro-ro traffic to and from ports. Apart from movements to Scotland the distances are generally not long enough to justify the extra handling costs of coastal movements. There is also a perception that shipping services have to establish a sufficient frequency to compete with the lorry, and that shippers are not large enough to provide the necessary guarantees. Shipping lines therefore would incur losses to set up such a service. If the service is a success other lines might enter the market, so that the first mover may not offset the initial losses. These views explained, in part, a reluctance to enter this market.

Conclusion

70. This paper has concentrated on container traffic and specifically deep-sea containers at UK ports. This reflected a perception, supported by others, that the growth in container traffic and the availability of capacity to deal with the ships employed on deep-sea trades are important catalysts for decisions by port operators about developments at UK container ports. The widespread use of large, 14-metre draught, 7,000 TEU vessels, and the prospect of even larger vessels is significant, because few terminals can easily cope with such vessels. For other trades and cargo types the capacity problems are not so immediately pressing.

71. The number of container units handled at UK ports has been growing very strongly particularly since 1996. To some extent this growth is a reflection of the strength of the UK economy and long-term trends in consumer tastes and the way in which those demands are supplied, but also the use of Felixstowe as a major transhipment hub. Growth related to trade and income is expected to continue, but the growth of transhipment is more difficult to predict. Transhipment traffic is much more dynamic and subject to the decisions of a few major lines and competition between major terminals in North-West Europe. The UK is well placed geographically to handle transhipment traffic from some countries in Western Europe, but competitive rates and quality of service are also important factors. As well as uncertainty about demand, there is uncertainty about the level of productivity achievable at existing terminals.

72. This paper does not come to a definite view on future productivity. On aggregate productivity measures UK ports compared well with their European competitors. But in terms of quality of service to shipping lines the picture appeared to be less favourable. Productivity at container terminals might be improved through better use of existing technology, such as cranes, and use of automation methods and advanced computer software to reduce labour and land for storage. But improving productivity at existing sites is much more difficult than for new sites. In addition, increasing productivity might be achieved only at the expense of poorer service, for example delays to ships at peak times.

73. On balance, and on a conservative view of future growth, it is the opinion of most of those consulted by the study that there will be pressure on capacity at UK container terminals within the next few years. Even if transhipment of cargo at UK ports, which is more sensitive to costs and quality of service than direct trade, were to stop growing, there would still be a strong underlying growth of direct trade. In other words, different levels of transhipment bring forward or put back the need for capacity requirements. The degree to which containers in direct trade use more terminal capacity than transhipped containers is uncertain. It may be the case that as existing capacity is used up and costs of expansion increase due to environmental mitigation, there will be pressure to use existing capacity more efficiently by raising productivity. Experience elsewhere suggests that it might be possible to make improvements, but there is no suitable model of terminal performance that can be reliably applied to individual sites. It was put to us that even with productivity improvements the need for new capacity is delayed rather than dismissed altogether. Given the lead times between any decisions to approve and proceed with a project, and completion of a project, and even with productivity improvements and lower transhipment, such decisions are likely to have to be made by ports and others on future capacity in the very near future.

74. Shipping lines and ports want to increase the share of inland distribution taken by rail, but they perceive difficulties due to capacity limitations either in the vicinity of the port or over the wider network. Shipping lines and ports are not optimistic about the prospects for coastal or short-sea movement of containers. The Government's 10 Year Transport Plan and Planning Policy Guidance are intended to tackle these problems.

75. There are various plans for expansion of container terminal development in the UK. Ports appear to be cautious of committing private funds for investment. This is understandable given that much of the investment is irreversible - quays and dredging cannot be used for anything else. There is considerable competition between UK and North-West Europe ports to secure direct and transhipment cargoes, but at the same time larger ships, the improvement of inland transport and the exchange of cargoes between different services encourages concentration of cargo and there are geographical barriers to entry.

76. There are various views about the merits or otherwise of dedicated facilities and common-user facilities at container terminals. From the point of view of the larger lines, dedicated facilities are seen as advantageous because of better and more controllable levels of service. However, it is not so clear that dedicated facilities benefit everyone, and efficiency and competition considerations may favour common-user facilities. There is not sufficient evidence to judge whether there is a case for government intervention favouring one format or another.

77. An important issue for government concerns the attitude to competition in the ports sector. With nearly 50% of UK container throughput being handled by one terminal operator, and the concentration of deep-sea cargo flows between a few shipping lines, competition issues might arise when considering the ownership of new port capacity.

Further Work

78. This study has noted some gaps in official statistics and also some analytical tools that might be useful in policy and planning.

79. The main gap in official port traffic statistics is the lack of a distinction between direct and transhipped cargo. The study relied on individual estimates from shipping lines and ports. There is no reason to doubt their accuracy, though there are difficulties in getting an aggregate figure for the UK and a reliable time series. A related problem is that it is impossible to identify the country of first origin or final destination of trade recorded at ports, although statistics collected under the EC Maritime Statistics Directive since 1 January 2000 do provide information on the port of loading and unloading. DTLR will consider whether data on transhipment and first origin and final destination of international trade by major port can be collected cost effectively without placing undue burdens on industry.

80. Official statistics have not recently surveyed the inland mode and origin and destination of trade through ports. The paper records estimates of mode share made by shipping lines and ports, but there may be a case for a regular official survey. This is an important area given the concern about excessive lorry traffic and constraints on transport modes near some ports. The current Engineering and Physical Sciences Research Council and DTLR funded research study being carried out by Imperial College, Modelling Decision Making Processes in UK Container Transport, will make an important contribution to the understanding of inland distribution. The final results of the study are not due for three years, but preliminary results may be available sooner.

81. The paper has discussed various measures of container terminal performance. There is not one single measure that can capture all aspects of terminal operations. DTLR will consider with ports and shipping lines and other interested parties the case for investigating appropriate key performance indicators for container terminals.

82. The coverage of this paper is limited to container traffic, and in particular deep-sea traffic. There might be a case for extending the analysis to ro-ro traffic, which is also increasing rapidly. The port capacity to deal with this traffic has rather different considerations than for container traffic, but there may well be requirements for major investment in the next 10 years.

Glossary & Abbreviations

References

Associated British Ports (2000), Dibden Terminal Environmental Statement: Non-Technical Summary (ABP, Southampton)

Baird A.J. (2001), Scapa Flow Container Transhipment Terminal (Edinburgh, Scottish Transport Studies Group)

Cargo Systems (2000), Various issues (London, IIR Publications)

Cargo Systems (2000), Measurements of Port Productivity and Container Terminal Design (London, IIR Publications)

Chen, T. (1998), Land utilisation in the container terminal, Maritime Policy & Management 25 IV (London, Taylor and Francis)

Containerisation International (1999), Yearbook (London, Emap Business Communications)

Containerisation International (2000), Various issues (London, Informa Group

Department of Environment, Transport and Regions (DETR) (2000), Maritime Statistics, (London, The Stationary Office)

DETR (2000), Focus on Ports, (London, The Stationary Office)

DETR (2000), Modern Ports: A UK Policy, (London, The Stationary Office)

DETR (2000), Transport 2010: The Ten Year Plan, (London, The Stationary Office)

DETR (2001) Planning and Policy Guidance on Transport 13 (London, DETR)

Department of Transport and British Ports Federation (1990), Transhipment of UK Deep-Sea Trade 1976-1987 (London)

Frankel, E. (1999), Intermodal integration. Lloyds Shipping Economist, 21, March (Colchester, LLP Ltd)

International Monetary Fund, (2000), World Economic Outlook (Washington, IMF)

P&O Ports, (2001) P&O Ports in the United Kingdom (www.poports.com.au)

Port of Felixstowe (2000) Harbour Revision Order: Trinity III Terminal Phase 2: Extension Environmental Statement (Peterborough, Posford Duvivier)

Royal Society for the Protection of Birds (1997), Port Development and Nature Conservation, (Sandy, UK, RSPB)

Stopford, M., (2001), A new revolution. Containerisation International, 34 I (London, Informa Group)

Turner, H.S., (2000), Evaluating seaport policy alternatives. Maritime Policy and Management, 27 III (London, Taylor and Francis)

Major North European Container Terminals