Land Transport Report - World Bank Group

WORLD BANK

Southern Mongolia Infrastructure

Land Transport Infrastructure to Support Mining Development

DRAFT BULLPIN CONSULTING LTD JULY 2008

TABLE OF CONTENTS SUMMARY

1. INTRODUCTION 14

2. TRANSPORT SECTOR IN MONGOLIA 15 2.1 INSTITUTIONAL STRUCTURE 15 2.2 ROAD SECTOR 15

2.2.1 Road legislation and finance 15 2.2.2 Infrastructure 15 2.2.3 Transport operations 16

2.3 RAIL SECTOR 17 2.3.1 1949 agreement 17 2.3.2 Rail Transport Law 17 2.3.3 Infrastructure 19 2.3.4 Transport operations 19

3. SOUTHERN GOBI DEVELOPMENT PLANS 21 3.1 INTRODUCTION 21 3.2 OVERVIEW 21

3.2.1 Coal 21 3.2.2 Other minerals 23

3.3 KEY MINING DEVELOPMENTS IN SOUTH GOBI 24 3.3.1 Overview 24 3.3.2 Timing 24

3.4 SECONDARY DEVELOPMENT 26 3.5 FUTURE TRANSPORT DEMAND 26

4. SECTOR DEVELOPMENT PLANS 29 4.1 INTRODUCTION 29 4.2 ROAD MASTER PLAN 29 4.3 RAIL INFRASTRUCTURE PLAN 31

4.3.1 Millennium Challenge 31 4.3.2 Infrastructure upgrading and new construction 32

5. ADJACENT NETWORKS 34 5.1 INTRODUCTION 34 5.2 CHINESE NETWORK 34

5.2.1 Overview of the regional rail network 34 5.2.2 Chinese railway development plans 37 5.2.3 Chinese port development plans 42 5.2.4 Handling constraints at border crossings 42 5.2.5 Operational considerations 44

5.3 RUSSIAN NETWORK 45 5.3.1 Overview of the regional rail network 45 5.3.2 Potential bottlenecks and constraints for sea-borne exports 45 5.3.3 Russian development plans 47 5.3.4 Handling constraints at border crossings and ports 47 5.3.5 Operational considerations 48

6. INFRASTRUCTURE OPTIONS 50 6.1 INTRODUCTION 50 6.2 ROAD 50 6.3 RAIL 50

6.3.1 Route options 50 6.3.2 Gauge 51 6.3.3 Intermodal options 51 6.3.4 Constraints on existing network and transhipment points 53 6.3.5 Rollingstock availability 55 6.3.6 Estimated rail capital and operating costs 55 6.3.7 Analysis of route options 56 6.3.8 Identification of preferred routes and timing 63

7. PRIVATE INVESTMENT IN MINERAL TRANSPORT 65 7.1 INTRODUCTION 65 7.2 OPTIONS 65 7.3 RAIL NETWORKS AND ACCESS 67

7.3.1 Accreditation, Licensing and Safety 67 7.3.2 Capacity Allocation and Network Management 67 7.3.3 Charging structures 68

7.4 ENCOURAGEMENT OF TPOs 68 7.5 OTHER ISSUES 69 7.6 POTENTIAL FOR PSP IN SOUTH GOBI 70

8. TRANSPORT INFRASTRUCTURE DECISION FRAMEWORK 71

WORLD BANK SOUTH MONGOLIA INFRASTRUCTURE LAND TRANSPORT INFRASTRUCTURE

DRAFT

BULLPIN

3

SUMMARY

This report describes the options, and their associated costs and benefits, for developing land transport infrastructure and services to support the development of key mines in Southern Mongolia. It discusses the potential for private-sector investment in infrastructure and develops a decision framework for infrastructure development in the region.

Future demand for transport

There are at least eight major coal and mineral projects in the South Gobi which are expected to come into production in the next five years or so; these are mostly coal mines but also including one of the largest copper deposits in the world. This will be accompanied by an increase in population and the development of associated urban centres. Together, there will be a substantial increase in the demand for transport, both to transport mineral production to markets and to provide goods and services to the local population and industry.

The principal economic driver of new rail construction will be the coal flows. Whilst the base metal mines have the capacity to earn large amounts of revenue, the volume of freight they generate is relatively limited and, even for Oyu Tolgoi, would be unable to justify significant new rail construction, whilst a regional population of less than 100,000 would likewise be much more economically served by truck and bus rather than a new railway.

The scale of the coal reserves is such that future transport volumes will largely be determined by demand, especially from northern China, rather than being constrained by availability of supply. The forecast increase in Chinese steel production will lead to continued growth in coking coal demand and Mongolia has rapidly become the most important external supplier, a position it will probably maintain into the long-term.

Exporting through either Russia or China to third countries is less certain. The competitiveness of Mongolian coal in the North Asian markets will critically depend on the cost and availability of rail transport to either Russian or Chinese ports; coking coal will probably be able to bear this cost but exports of thermal coal will be difficult, especially so through Russian ports at current rail tariffs. The most likely scenario may be for Mongolian coking coal to displace Chinese supplies from mines further towards the coast which are then freed up for local export to Japan and South Korea.

As China currently imports the equivalent of about 10 million tonnes of concentrate, Oyu Tolgoi should thus be able to readily export its entire production of copper concentrates (around 2 million tonnes p.a.) for the foreseeable future. Whilst this traffic would not justify a rail line by itself, it will almost certainly move by rail if a rail line exists.

Overall, it is likely that Mongolia will export around 25 million tonnes of minerals by 2015, of which possibly 10 million tonnes will be exported overseas, probably through a Chinese port. In the longer term, these volumes could double.

The mine inputs of supplies and fuel are generally an order of magnitude smaller than the mine output unless there is a smelter attached to the mine and, with the exception of fuel, are generally not well suited to rail transport. In comparable situations, mining supplies are generally collected from a number of suppliers and delivered as a mixed load by road as long


DRAFT

BULLPIN

4

as there is road access. These inputs therefore will rarely be a deciding factor as to whether or not a railway is constructed. In addition to the mines themselves, there will be demand for passenger and freight transport from the settlements in South Gobi. However, again this is unlikely to be significant in terms of deciding whether or not a line is built and most, if not all of this traffic is likely to be transported by road.

Infrastructure Plans in South Gobi

The objectives for the road sub-sector in the National Development Policy formulated by the GOM in 2005 include the continuing development and improvement of the corridors linking to the international highways, constructing the three Asian Highways and continuing the periodic maintenance of state highways and local roads. Achieving these objectives was addressed in more detail by a Road Masterplan prepared in 2007; this aims to provide a sealed road between Ulaanbaatar and each aimag centre as well as connecting the international network with the national roads serving the five economic regions of Mongolia.

Within South Gobi, there are plans to construct or reconstruct sealed roads between Ulaanbaatar and Dalanzadgad via Mandalgovi as well as roads linking Ovoot Tolgoi and Dalanzadgad (via Tavan Tolgoi and Oyu Tolgoi) with the Chinese border. However, the masterplan does not include any projects connecting Dalanzadgad towards the east and the Trans-Mongolian Highway

The Mongolian rail network (MTZ) is of Russian gauge and has been constructed to Russian standards. Although it is being upgraded, this is along-term project and the network is in generally poor condition with limited remaining capacity on its north-south mainline. MTZ has a dual ownership structure, with equal shares held by the Mongolian and Russian governments, and this has created considerable difficulty in terms of agreeing an investment and upgrading plan.

In 2007, the Millennium Challenge Corporation (MCC) provided a grant of US$ 188 million to GOM for rail investments. This includes funds for rollingstock which will be owned by a new leasing company which will lease them on commercial terms to MTZ and independent shippers, thus providing the start of private-operator access to the rail network. The grant does not include any infrastructure component other than new signalling equipment and this will need to be addressed if significant volumes of minerals are to be exported on any routes other than the cross-border lines to China.

A number of new lines have been proposed at various times, some of which are more likely than others; most will be built only if there are major mining developments which generate substantial export quantities of coal. Cross-border links to China from Tavan Tolgoi and Ovoot Tolgoi are likely to be the first built, followed by a link from the South Gobi to the main Trans-Mongolian Railway (TMR) when it is justified by demand.

Neighbouring rail networks

Exports of minerals from Mongolia depend critically on the capacity and charges on the neighbouring rail networks in China and Russia.

China


DRAFT

BULLPIN

5

China’s rail network is extensive but also highly congested and there is very little spare capacity on key routes, especially those crossing the mountains which run north-south west of a line between Beijing and Guangdong. Coal is about 40 per cent of the traffic carried on China’s railways and its strong growth in recent years is forecast to continue for at least the next decade, thus being the single most important factor affecting network capacity. The major coalfields are all located west of the mountains and over 700 million tonnes is carried each year, mostly by rail, from these to the eastern and southern provinces, either directly or by coastal shipping from the northern ports.

Nine railway lines, forming three groups, currently cross the mountains separating these fields from the coastal plain. Two of these lines, the Daqin line and the privately-owned and operated Shenshuohuang line, are specialist heavy-haul lines, with a combined capacity of around 600 million tonnes. Coal shipments from the main western coalbases were forecast to more than double by 2020 compared to 2006 and the railway Medium and Long-term Development Plan (MLTDP) prepared in 2004 places considerable emphasis on expanding coal transport capacity. The increase in rail capacity is complemented by expansion plans for the seven main coal ports on the east coast (including the construction of a new port at Caofeidian, a sub-port of Tangshan) which are expected to increase their capacity by 60% by 2010 compared to 2005.

In addition to increasing rail capacity across the mountains, China plans to build several other lines to carry minerals from Mongolia. A line between Hami and Linhe is currently under construction as far as Ceke; this acts as a collector line for traffic feeding in from South Gobi by road or rail. Four feeder lines are planned to carry traffic from railheads on the China-Mongolian border between the current crossing at Zamyn Uud and Urumqi in the west. A major new line is also being constructed to carry coal from Nei Monggol to Liaoning; this is planned to subsequently be extended across the border to Choibalsan in Mongolia to provide a third trunk route between northern China and Russia. These lines are constructed, at least initially, as single lines, with a capacity of about 20-25 million tonnes and could typically be built within 18 months – 2 years of being approved. Most of them are likely to be built as either joint venture lines, with the financing shared between the central Government, provinces and potential customers, or as purely private lines.

The difference in gauges between the Chinese and Mongolian systems currently requires traffic to be transhipped at the border. This is a time-consuming process as, although the physical transhipment itself is generally not a problem, it does require adequate supplies of wagons on either side of the border as well as negotiating customs requirements. The short cross-border lines are planned as standard-gauge lines, which will avoid the need for such transhipment; lines to the north and east of the South Gobi, by the same reasoning, would normally be Russian-gauge.

Although there are several non-China Railways operators currently operating, most of them do so on their own networks and there are comparatively few private wagons, and almost no private locomotives, operating on the Chinese network. The concept of a Mongolian-owned train, with Mongolian locomotive and wagons, operating through to the Chinese mainline network, is thus some way off. China Railways has actively discouraged privately-owned


DRAFT

BULLPIN

6

wagons for some years and, as yet, there are also no formal arrangements whereby a third party can pay access charges to operate over the main CR network. The normal arrangement is that traffic is carried in CR wagons; whilst they are on non-CR lines, a standard charge is made per day; whilst they are on CR lines the normal CR tariff applies.

An important consideration with the Chinese network is obtaining a reliable long-term capacity allocation. Currently, there is an annual convention at which the available capacity (which is constrained on most of the east-west routes) is allocated to users, and on which MOR’s annual transportation plan is then based. If a particular shipper fails to obtain an allocation, he can apply to the relevant railway authority for a non-plan allocation but this is not guaranteed and is also subject to change if circumstances change. For this reason, using the privately-owned Shenhua line may be an attractive long-term option for Mongolian shippers, especially as Shenhua produces very little coking coal itself and thus is not a direct competitor.

Chinese rail tariffs are controlled by national and provincial pricing authorities and are generally set to achieve cost-recovery; they thus vary between different lines. Coal traffic on the main network is typically charged around US$ .015 per net tonne-km (ntkm), with rates on other lines of up to US$ .026 per ntkm. Indicative rates from the Mongolian border to Qinhuangdao, the main coal port, are $US 25.00 from Gashuun Sukhait and $US 18.00 from Erlian.

Russia

The Russian rail network connecting to Mongolia is much simpler, essentially consisting of the Trans-Siberian Railway (TSR) linking the Mongolian network to the Far East ports. The TSR is a double-track electrified railway which is reportedly currently carrying about half its capacity. Unlike the Chinese network, therefore, line capacity is not a significant issue for Mongolian mineral exports. However, there are likely to be significant potential tariff and access barriers.

The two main coal terminals are at Vanino and Vostochnoy. Both are privately-owned, either wholly or partly by mining interests. Russian policy is that all terminals should be Russian-owned; they key issue in Russian ports is thus the ownership of terminals and the extent to which Mongolian coal would get access at times of either limited port capacity or when Mongolian coal was competing with Russian producers for the same markets.

The Russian rail network is open to third parties and shippers have been able to operate their own trains, supply their own wagons or continue to rely on the Russian railways (RZD). During this period, a number of third-party operators have emerged, some of whom are the transport arms of major industrial organisations but others of which are genuinely independent transport specialists. Any Mongolian exporter could thus use one of these operators. However, Russian access charges remain commodity-based and tariffs for transit coal are two-three times higher than those for domestic coal which is being exported; the current charges for export coal from Naushki to Vostochnoy are $85/tonne, pricing thermal coal out of the market and probably a major barrier for coking coal in the long-term.


DRAFT

BULLPIN

7

Infrastructure options in the South Gobi

The road networks that have been proposed in South Gobi are consistent with the RMP network but also include a connection between Oyu Tolgoi and Sainshand, serving Tsagaan Suvarga en route) as well as a cut-off connecting Tavan Tolgoi directly with the Dalanzadgad – Ulaanbaatar road. No proposals have yet been made for a link between Ovoot Tolgoi/Nariin Sukhait and the aimag centre at Dalanzadgad; instead Nariin Sukhait is linked to Arvaikheer as part of a north-south vertical road.

Both the roads from the mines to the Chinese border are expected to be financed by the mines, with fees then being levied for their use by third parties that are used to partially defray the cost of construction and/or maintenance.

The possible rail lines that have been presented at various times fall into three groups:

• Cross-border lines from Ovoot Tolgoi to Ceke and from Tavan Tolgoi and Oyu Tolgoi to Gashuun Sukhait

• A line from Ovoot Tolgoi through Dalanzadgad to Tavan Tolgoi which connects these two

• Connections from the Tavan Tolgoi/Oyu Tolgoi area to the main TMR; these include:

o From Tavan Tolgoi direct to Ulaanbaatar via Mandalgovi; this would provide the most direct route to the north and Russia as well as providing a reasonable service for non-mine traffic to and from the capital

o From Tavan Tolgoi to the TMR in the vicinity of Airag; this is the shortest route to the TMR but involves a longer distance for traffic heading north

o From Oyu Tolgoi to Zuunbayan and from there on the existing branch to Sainshand on the TMR. This involves even less new construction, although the Zuunbayan branch would probably have to be reconstructed.

o In addition, this analysis has considered an additional route direct from Tavan Tolgoi to Sainshand which avoids the detour via Oyu Tolgoi.

• Any of the last three routes would also fit with any longer-term development of a new route to the Russian ports via Choibalsan and the border crossing at Ereentsav.

The cost of constructing a new railway varies significantly, depending on the terrain and the standard of the track being constructed. Construction costs in China, the closest parallel to the South Gobi, vary from a minimum of under $1 million per route-km for a single-track minimum-standard railway in easy terrain to over $10 million per route-km for a double-track medium-high speed line in mountainous terrain. A line capable of carrying 25 million tonnes of minerals a year in moderate terrain such as the South Gobi, would normally be constructed as a Class I railway in China, with a ruling grade of 0.6 – 1.3%. The cost of such lines in the South Gobi is estimated at about $1.8-2.0 million per route-km, depending on the terrain and volume of earthworks.

Operating costs likewise depend on the standard of the line (in particular the axle-load) and the terrain (which affects the tonnes per train and per locomotive). The two most modern and


DRAFT

BULLPIN

8

efficient mineral lines in China, the Daqin and Shenhua heavy-haul lines, had operating costs in 2007 of just under 0.008 $US/ntkm (including depreciation) while the average cost on the CR network for all freight operations in 2006 is estimated at $US0.0093/ntkm, excluding depreciation. A figure of $US0.01/ntkm, including depreciation, has been used in this analysis, but in practice costs could range between $0.009-0.012/ntkm.

Connections to mines

When a mine is commencing production at a low level, production is normally carried by truck to a railhead, as currently happens at both Tavantolgoi and Nariin Sukhait. Where this does not adversely affect the local population or the environment, this is generally the most economical method of transport for tonnages up to around 4 million tonnes, although the precise level will depend on individual circumstances. Typical breakeven tonnages under Mongolian conditions for a mine to build its own rail connection range between 3 and 8 million tonnes; given the forecast outputs, the mine at Tavan Tolgoi justifies a rail connection under any circumstances, as would the adjacent Baruun Navan. Those at Ovoot Tolgoi and the adjacent Sumbar in the west also would probably justify a rail connection; if this is constructed to the south, then it could also serve Nariin Sukhait. Oyu Tolgoi would be unlikely to justify a rail link by itself but would be served by a line from Tavan Tolgoi to the south. However, Tsagaan Tolgoi and (even more) Tsagaan Suvarga are marginal cases at best and would probably transport by road to a suitable railhead unless other factors became significant.

Ovoot Tolgoi – Tavan Tolgoi

The case for a connection between these two lines via Dalanzadgad is less clear. Mining inputs alone would not require a connection from the north to Ovoot Tolgoi. Its justification therefore depends on whether the Ovoot Tolgoi mines wish to have an alternative outlet for their coal, either via Gashuun Sukhait and Baotou or via Russian ports.

The route to China from Ovoot Tolgoi via Tavan Tolgoi is longer by about 130 kilometres but is cheaper by about Rmb 10/tonne, excluding the cost of constructing the line. At a discount rate of 10%, this saving requires a volume of around 50 million tonnes before it becomes worthwhile and the only justification for the line would then be the option of using the Russian ports.

Connections from Tavan Tolgoi to TMR

The principal purpose in constructing one or more of the routes between Tavan Tolgoi and the TMR is likewise to provide an alternative export route to Russia or China or both. For southbound traffic, the routes via the TMR are at least 300 km further than by using the Gashuun Sukhait crossing point and the Chinese network, require a bogie exchange at the border, possible duplication of the southern part of the TMR, and also need to negotiate the heavily congested Chinese network around Datong. It is unlikely that any significant volume would use a link via the TMR in preference to the Gashuun Sukhait line unless the tariffs on the Mongolian network were exceptionally low and the principal justification for these connections is thus their usefulness for exports via Russia, always remembering that such


DRAFT

BULLPIN

9

exports are only viable as long as the fob price of coking coal at the ports remains high enough to cover the land transport cost of around $100/tonne.

For northbound traffics, the direct line to Ulaan Bataar is the shortest, being 200-400 kilometres shorter than the other options; it is also consistently the cheapest option for northbound exports, as its higher construction cost is outweighed by the need for less upgrading of the TMR and the much shorter distance with its lower operating costs.

The land transport cost to the Chinese ports is about $70/tonne cheaper than via the Russian ports and exporting via a Chinese port would therefore generally always be the first choice. The route via Gashuun Sukhait is the cheapest southern route by around $10/tonne (for an export volume of 10 million tonnes), reducing to $4/tonne if exports reach 20 million tonnes p.a. The two Sainshand options have lower costs than the Airag route as they join the TMR further south, while the Ulaan Bataar line unsurprisingly has a high cost penalty for exports to China.

The optimum choice of route to access the TMR thus depends on the direction in which the exports are expected to flow; predominantly northbound flows will favour the more northerly options whilst the converse is true for southbound exports. However, for these latter, the Gashuun Sukhait route is so superior that it is unlikely a significant volume would use the TMR unless there were exceptional circumstances.

East-west route via Choibalsan

A final option is a cross-country route, which extends the Airag route to use the existing 60-kilometre Borondor branch and then build a new line to Choibalsan. From there, traffic can use the existing line (which would need to be rebuilt) to access the Russian network at Solovievsk and subsequently join the Trans-Siberian Railway. Unfortunately, this route only saves 200 kilometres compared to using the TMR from Airag to Ulan Ude (and only 30 kilometres if the Ulaanbaatar line is constructed), as the route from Solovievsk to the TSR runs generally north-west

Because of the small distance saving, the Choibalsan route from Airag only becomes justified economically when volumes reach 30 million tonnes p.a. and, even then, the Ulaanbaatar route remains cheaper.

Summary

Whilst the cross-border routes to China are likely to be viable at around 5 million tonnes p.a., extending the network to connect to the TMR needs to wait until the export potential for Mongolian coal is clearer. It seems unlikely that any significant volume of exports to China will use the TMR via Erlian unless the line is built for other reasons but most other developments in the Tavan Tolgoi – TMR corridors are metal mines generating comparatively low volumes of product which would normally be transported by road to an existing railhead.

Whichever route is constructed, it will be better for some exports flows than others. If exports via Russia can be guaranteed and it is agreed all exports to China will use the cross-border lines, then the Ulaan Bataar route is the best but if it is required to keep options open then an alternative approach may be needed in which the route which gives the least penalty for a mix of traffics. If equal volumes of exports are assumed in the two directions, the Airag and direct


DRAFT

BULLPIN

10

Sainshand options are better than either of the extreme north and south connections (Sainshand via Oyu Tolgoi and Ulaanbaatar). However, the Airag route is the more robust as, whilst it is never the best option for either direction, it is also always within about $2/tonne of the lowest cost for either direction, a small penalty compared to the total cost of transport to the ports. It also provides the opportunity for future extension to Choibalsan should traffic volumes expand or if the network structure develops to provide a more direct link to the TSR.

Whilst this analysis is adequate at a strategic level, practical route selection will also need to take into account any engineering features of these lines, including any differences between the various options in construction costs and operating conditions and unit costs.

The distance to ports of Mongolian exports leave them very vulnerable to changes in world prices or to unilateral changes in rail tariffs in Russia and China and long-term contracts covering not only the end-user, but also port and rail access in China or Russia, would be required before any substantial new construction would be justified, or attractive to private investment.

While the cross-border lines are planned as standard-gauge lines, it is likely that most of the traffic using the connections to TMR will be to or from the north and these lines should therefore be broad-gauge. If the Airag route is selected, Oyu Tolgoi will not have a broad-gauge connection but it seems unlikely much concentrate traffic would move north and any that does could be transported by road to a railhead at Tavan Tolgoi.

Private investment in mineral railways

General considerations

The private sector can be involved in the construction and operation of railway lines in many different ways: through infrastructure construction, infrastructure operation (i.e. signalling and train control), rollingstock provision and train operation. There are five basic types of concession: • Train availability contract, where rolling-stock is provided by a third party (who also

normally also does the scheduled maintenance). These enable capital expenditure to be spread over the life of the asset, as well as guaranteeing rollingstock availability and reliability.

• Train operating concession, in which operators either pay Government (or are paid by Government) to operate specific services. Such operators pay access charges (normally a combination of a rate per train-kilometre and a rate per gross tonne-kilometre) to the infrastructure provider. In some cases, and this is the normal situation for freight, there is ‘open access’, i.e. an operator can operate trains without any financial payment to or from Government by complying with technical standards and paying access charges

• Infrastructure concession – where an organisation undertakes to build infrastructure which is then leased to Government (either directly or to a Government-owned railway) for an annual lease fee. This arrangement avoids Government having to make a large initial capital expenditure and also greatly reduces Government’s risk of project over-runs and delays.

• Infrastructure build and operate concession – where an organisation both builds the infrastructure and ‘operates’ it (i.e. provides train control and supervises train operations),


DRAFT

BULLPIN

11

similar to the arrangements used for toll-roads. One example of such an arrangement is a current proposal for a new mining development in Australia.

• Integrated concession in which a company is given the right to build and operate a railway, normally for a defined time period (historically 50 or 100 years) subject to certain conditions such as tariff control, service obligations or (on the Government side) a dividend guarantee.

There are a number of variations on these basic models; for example, a line can be financed by a mine developer, even if it is built and operated by the public sector, with the cost of the line being offset against freight revenue received from the mine.

A fundamental distinction is that between a railway built by an individual company (or companies) on which they transport their own products as part of an overall logistic chain and railways which third parties can use – either using and operating their own rollingstock or using services provided by railway operator. Many of the new railways constructed in the last fifty years have been of the first type, typically as part of an integrated mine-rail-port project. They are effectively private railways with no obligations to provide a service for third parties, although there is often some informal passenger transport.

However, in other cases, the railways have either been explicitly built to carry third-party freight or it is an obligation imposed by the Government as part of the permission to construct the railway, in order to ensure that smaller mines that cannot justify building a railway line on their own account can be provided with access at a reasonable price. A key decision for Government when granting a rail concession is therefore whether the line is to be common-user and, if so, what arrangements are proposed for this to be enforced in practice as well as in theory.

There are three broad streams of work associated with administering such a rail network with multiple users:

• Accreditation, licensing and safety • Establishing the procedures for capacity allocation and network management • Establishing the charging framework and prices

Each of these streams of work has two phases: • Defining the principles to be adopted • Developing the procedures by which they will be implemented

These are normally summarised for third party operators (TPOs) in a document which in Europe is known as the ‘network statement’.

Even though the regulatory and operational obstacles for TPOs may be overcome, they may still face difficulties in obtaining rollingstock. Most rail rollingstock has a life of over 20 years, a long period to which to commit given that comparable road vehicles may only be used for 3-5 years and then easily disposed of through the secondhand market. A key element in countries where TPOs have flourished has been the existence of third-party rollingstock which can be leased or hired for periods of one or two years. In some countries this has been through specialist leasing companies; in others from other railways. An obvious option in Mongolia is for TPOs to contract with one of the Russian railway operators for the provision


DRAFT

BULLPIN

12

of rollingstock or alternatively to develop the leasing company created as part of the Millennium Challenge project.

Other issues to be considered include whether any line that is constructed by a private company will eventually revert to the state (either at the end of the mining lease or after some more general time period) or whether ownership of the line will continue in perpetuity. The handback arrangements need to be clearly stated in the concession agreement, as well as the conditions under which the government can step in and takeover if the concessionaire is not performing as required. This in turn requires a monitoring procedure with is non-intrusive but observant.

Potential within South Gobi

The potential for PSP in the rail network in South Gobi ranges from very good to negligible. The most important thing for an investor in infrastructure is the reliability of the traffic forecasts. The lines linking the proposed mines with the industrial complexes just over the border in China will be carrying traffic that is subject to long-term contracts. Whilst such contracts cannot be relied on absolutely, they are the closest thing to certainty that the mines are going to experience, especially if the industrial complexes themselves have a stake in the mines. For many of the plants in Gansu and Nei Monggol, Mongolia is the closest and cheapest source of their inputs and this also provides some reassurance that substantial traffic volumes can be expected in the long-term.

The lines linking the mines to the existing Mongolian network unfortunately are unlikely to have any such long-term guarantees. Exporting coal through Russian orts must overcome two uncertainties: firstly, the inherent uncertainty of the international market and, secondly, the uncertainty associated with exporting through Russia, which has shown several times in recent years that it is prepared to use energy as a political weapon. Whilst it may be possible to interest some local businesses in lines connecting the mines to the existing network, it seems likely that any such interest would require some guarantees from the government. The other option is to interest some of the Russian resource companies in the South Gobi mines.

It is important that any lines that are built allow access to third parties at a reasonable rate. Whilst Tavan Tolgoi is clearly large enough to justify a line in its own right, other mines with a production capacity of, say, 2-5 million tonnes p.a. will not be and will naturally use road, at a much higher cost per tonne than if they had access to rail. At the margin, this will mean that if small potential mines adjacent to existing lines cannot obtain rail access, they will not be developed and it is in Mongolia’s national interest to ensure this does not occur.


DRAFT

BULLPIN

13

Outline decision framework

Issue Decision Comments

Road access to new mines

Privately-funded as far as public network

Determine threshold volumes above which access roads must be sealed to preserve environment. Specify construction standards

Use of public roads for hauling product

Allow up to specified volume subject to loading standards and charges

Volume limit will depend on level of other usage. Loading should be enforced at corporate level. Charges must be sufficient to cover long-run maintenance and renewal costs

Rail access to new lines Allow

Require access to third parties under specified conditions. Develop standard procedures and charging methodology. Specify handback requirements.

Connection to TMR

Allow if privately-financed. Otherwise defer until there is clear evidence of long-term demand

Undertake marketing and operations study to establish long-term potential for export to Russia. May require commercial agreement with RZhD and ports to obtain reduced long-term access and rate for export coal transitting from Mongolia. As China appears to currently have $70/tonne advantage, a better option may be to develop alternative customers in China; the ownership of the connecting border lines in China should be monitored.

Connection between Ovoot Tolgoi and Tavan Tolgoi

Allow if privately-funded.

Any Government funding should be restricted to the Tavan Tolgoi – Dalanzadgad section and only considered after a connection to TMR has been constructed.

Connection to Choibalsan Long-term option only Monitor progress of proposed Chinese

line through Choibalsan


DRAFT

BULLPIN

14

1. INTRODUCTION

The objective of the report is to identify options, and their associated costs and benefits, for developing land transport infrastructure and services to support the development of key mines in Southern Mongolia. This development is dominated in volume terms by several major coal mines which are either already producing or are planned to do so and the report has therefore concentrated on the options for the development of a railway network to distribute the mines’ production.

Following this introduction, Chapter 2 summarises the current situation in the land transport sector in Mongolia, briefly covering the institutional framework and key legislation and then describing the infrastructure and transport operations in both the road and rail sectors.

Chapter 3 reviews the mining development plans for the Southern Gobi, including the size and timing of the main mines, to provide a broad estimate of the magnitude of future transport demand.

Chapter 4 then summarises the current sector development plans, including the Road Masterplan and the various projects that have been proposed or are in progress for rail.

Because of its landlocked position, Mongolia can only sell directly to two countries, China and Russia, and for all others is dependent on the transport networks of its two neighbours to access ports. Chapter 5 describes these networks and their development plans, as well as discussing operational issues, including network congestion and rail tariffs.

Chapter 6 identifies road and rail development options in the South Gobi. The key road developments are covered by the Road Masterplan but, for rail, there are a number of options, particularly for east-west connections to the existing network. This chapter discuses these options and estimates the capital and operating costs and benefits associated with each one for the main potential export movements. The options are then compared to identify those which preferred over a range of traffic volumes.

Chapter 7 discusses the potential for private-sector investment in infrastructure, the preconditions which are conducive to the investment and the regulatory requirements which will maximize the associated public benefits.

Finally, Chapter 8 presents a decision framework to help structure the land transport infrastructure development program.


DRAFT

BULLPIN

15

2. TRANSPORT SECTOR IN MONGOLIA

2.1 INSTITUTIONAL STRUCTURE

Transport in Mongolia is administered by the Ministry of Roads, Transport and Tourism, with the day-to-day responsibility with modal departments. The Department of Roads is responsible for developing and maintaining the state road network, implementing Government road-related policies, providing road construction and maintenance companies with professional administration and operational cooperation, and supervising road, bridge construction and maintenance.

Railways have historically been the responsibility of the “Ulaanbaatar Railway”, a company jointly-owned by Mongolia and Russia; however, in 2007, a new Railway Law was passed which established the Mongolian Rail Authority (MRA), which has responsibility for planning and administering the rail sector, including any new railway companies that may be established under the Law.

2.2 ROAD SECTOR

2.2.1 Road legislation and finance

The main legislative basis controlling the road sector is the “Law on Roads”, ratified in 1998 by the Mongolian Parliament, and other related acts and regulations. Funding for the development of state roads and bridges is made partly through a road fund, financed by fuel taxes, vehicle registration fees, and the state budget but increasingly through direct allocations from the budget.

2.2.2 Infrastructure

Road Network There are around 50,000 km of roads in Mongolia, administratively classified into two categories (Table 2.1) • State roads connecting the capital to the provincial (aimag) centres, important towns

and important border crossings; about 1,500 km of these are paved and the remainder are • Local roads connecting aimag centres to district (sum) centres, where these are not

connected by State Roads. There are also Internal roads connecting Sum centres to other population centres and farms and about 460 km of roads within the Ulaan Bataar urban area.

Table 2.1 Length of road network Mongolia 2005 (km)

State Local Total State Roads 11,219 38,495 49,714 Paved 1,503 758 2,261 Gravel 1,440 498 1,938 Improved earth 1,346 499 1,845 Earth (tracks) 6,930 36,740 43,670


DRAFT

BULLPIN

16

The ‘improved earth’ roads are formed earth tracks but their low levels of maintenance in recent years have caused them to be effectively abandoned by road users, who often create new tracks across the steppe.

With only 160,000 vehicles (excluding motorcycles) in the country, the network is lightly loaded. Only a few of the State roads carry more than 1,000 vehicles per day and most are in the low hundreds.

Mongolia has an ambitious Road Master Plan, which aims to have around 10,000 km of the State network paved by 2020, whilst ensuring that 70 percent of the network as a whole is in ‘good’ condition. This plan is estimated to cost around $2 billion at an average rate of about $140 million p.a. (discussed further in Section 4.2).

2.2.3 Transport operations

Many of the current commercial road vehicles in Mongolia are of Russian origin, which are robust and easy to maintain, with ready access to spare parts. They are thus well suited to handle the poor condition of most of the road network, with generally low speeds and high vehicle wear rates are very high.

There appear to be no regulations as to permissible loads on roads and there is little control on overloading. As a result, trucks generally, and particularly those associated with the mining industry, are seriously overloaded and causing much road pavement damage. The only data readily available, from a special trial at the Chinese border crossing at Zamiin Uud in 2003, showed an average of 41 ESA per vehicle. Anecdotal data reports that the average net load carried in (typically five or six-axle) coal trucks carrying coal to China in the South Gobi is around 85 tonnes. This is many times higher than what would normally be expected (and is achieved in countries that have implemented measures to control over-loading on the road network).

It is not economic to design road pavements for such overloading and such loads also do severe damage to the ground where they are hauled over earth roads. Therefore it is crucial to implement an effective means of controlling this problem and protect the significant financial investment that has been made in the road network.

For most functional road networks around the world, management authorities set axle load limits for the road based on the design conditions for the road. Limits may take into account gross vehicle mass and axle configuration and compliance is then made using weighbridges (this planed to become standard practice in China on the arterial and tollroad network).

Road transport services (freight and passengers) are operated by the private sector, with limited regulation by MRTT which has recently begun registering transport operators on a voluntary basis. There is no requirement for formal reporting by the private road transport operators and the published data is of very limited value.

Road safety is a serious concern in Mongolia, with total fatality rates of at least 25 deaths per 10,000 registered vehicles, and personal injury accident rates at four times this figure. Currently, there are limited procedures in place aimed at the prevention of traffic accidents in Mongolia by way of safety awareness programs.


DRAFT

BULLPIN

17

2.3 RAIL SECTOR

2.3.1 1949 agreement

An agreement between the People’s Republic of Mongolia (PRM) and the Union of Soviet Socialist Republics (USSR) was in 1949 to establish a joint Mongolian-Russian shareholding company (the “Ulaanbaatar Railway”) for the construction and operation of a Russian-gauge railway line between the border near Naushki to Ulaanbaatar. Construction had begun in 1947 and operations began in 1949. The two shareholders of the company are the Mongolian Ministry of Transport and the Soviet Ministry of Roads and Communications. In 1952 the agreement was extended to cover construction of the line from Ulaan Baatar to the Chinese border and the Choibalsan lines in the east of the country were also transferred to the company in 1956. The Management Board is based in Ulaanbaatar. The Russians contributed 51% of the capital and the Mongolians 49%, partly in cash but also in kind through land, materials and credits from USSR.

The joint ownership of the line has created significant problems for Mongolia, as any upgrading needs both the approval, and a funds contribution, from the Russian government, which inherited the agreement. As a result, there has only been very limited investment in recent years and the main TMR line is currently in poor condition, leading to various proposals for Mongolia to construct its own line parallel to the existing one.

2.3.2 Rail Transport Law

The Law on Railway Transport1, adopted in July 2007, sets out the principles governing railway transport operations in Mongolia as well as providing the framework for railway safety regulation. The Law establishes a Railway Authority, to operate on a self-financing basis.

The key principles governing rail transport, as stated in the Law, are to: • ensure coordination of the railway transport under integrated regulations; • ensure regular control and supervision; • ensure access, quality and safety of railway transport services; • ensure sustainable operations in the railway sector; • create conditions for market competition in railway transport sector; • coordinate railway transport with the other transport sectors

The Law specifies that the railway strip will remain in the state ownership, and the basic structure2 will be in state or (majority-state) ownership. Other railway objects3, such as branches and railways for specific uses can have other forms ownership. The Law allows the construction and use of a basic structure by legal entities on agreed terms, given eventual transfer to state-majority ownership.

1 Source: Website www.legalinfo.mn, the Law on Railway Transport has been adopted on July 5, 2007. 2 Basic structure includes railway track and formation, engineering facilities, stations, junctions, energy and water supply,

communication, and information facilities to ensure normal operation of railway transportation as well as technical facilities such

as buildings, structures, culverts and fences. 3 Railway object means railway strip, basic structure, track, area, railway rolling stock, and other machinery and equipment and

assets related to railway transport safety.


DRAFT

BULLPIN

18

The Law also specifies the rights and responsibilities of the Parliament, Government, responsible ministry and local authorities regarding railway transport. Parliament is responsible for determining the state policy in railway transport, ensuring compliance with the Law, and ratifying international treaties and agreements. Government is responsible for the implementation of state policy in railway transportation, awarding and suspending construction licences for basic structures and determining the terms under which the railway strip is provided4.

Railway Authority The Railway Authority (RA) has 4 divisions:

• Administration and finance division;

• Policy implementation and cooperation division;

• Transport regulation division;

• Division for certification and legal affairs

It also has a section for supervision and a centre for information and registration (see organizational chart below). It reports to the Cabinet member in charge of railway transport and act within his jurisdiction.

Figure 1 Structure of the Government implementing agency, Railway Authority The RA “Strategy of Activities and Restructuring Program”, approved in August 2008, sets out to effectively plan and to implement the state management of railway transport. It has the following strategic objectives:

1. Improve efficiency and competitiveness of railway sector by developing and implementing the sector’s long, medium and short –term plans, improve the sector’s information, registration and research capacity and providing information to users; 2. Improve service quality and coverage of the railway sector transport to meet demand from the society by developing and introducing the new standards for services, endorse decisions for service quality and coverage and support implementation; 3. Ensure security of railway transport by supervising safety of railway transport, providing methodological support and compliance for entities involved in railway transportation, developing and implementing comprehensive plan to ensure railway transport safety; 4. Develop necessary legal, economic and other prerequisites for improving the state management for railway sector transport by developing necessary legal acts and regulations, ensuring the state administration leadership, improving human resource leadership; 5. Ensure implementation and compliance with international treaties, legal acts and regulations in railway transport by ensure implementation and compliance with provisions of international laws and regulations, ensure monitoring for compliance;

4 Railway strip means the land required for safely undertaking rail operations on the railway tracks.


DRAFT

BULLPIN

19

6. Improve economic and investment efficiency in railway transport and develop international cooperation by improving utilization of state assets in railway sector and improve efficiency, and developing international cooperation.

The Railway Authority (RA) is thus, amongst other things, responsible for the implementation of legislation covering both economic and safety regulation of the rail sector, as well as licensing and supervising new lines and railway operators.

2.3.3 Infrastructure

The Mongolian rail network consists of two disconnected portions: a north-south line connecting Russia and China via Ulaanbaatar, together with seven short branches mostly serving mines and a line in the extreme north-east serving Choibalsan and connecting with the TSR south of Chita. The terrain is flat to undulating; the average grade is 0.9%, although this reaches 1.8% in the section between Ulaanbaatar and Bagahangai. The total length of the network, all of which is single-track, is about 1,815 kilometres, of which 1,111 kilometres is the north-south main line. It is of Russian gauge and has been constructed to Russian standards with Russian rollingstock. The mainline is predominantly (85%) 50 kg/metre rail with wooden sleepers but is being upgraded (12%) to 60 kg/metre rail on concrete sleepers. Traffic has approximately doubled in the last decade and MTZ has constructed several new passing loops and is extending passing loops to 1250 metres; this allows a maximum train size of 6,000 gross tonnes. The capacity of the mainline is quoted as 20 pairs of freight trains per day, equivalent to 20 million tonnes of freight given the average train size of about 1700 net tonnes and allowing for a few passenger trains.

2.3.4 Transport operations

In 2007, the railway carried 14.1 million tonnes, of which 3.5 million tonnes was transit traffic, mostly timber and petroleum products from Russia to China. It also carried 4.5 million passengers (1.4 billion passenger-kilometres), almost all local. Local traffic, which has been about 50 percent of the tonnage for the past few years, largely consists of coal to the Ulaan Baatar power stations and ores to Erdenet.

Table 2.2 MTZ traffic volumes (2004-2007)

2004 2005 2006 2007 Freight Volume (000 tonnes)

Local 6857 7559 7337 7494 Import 1225 1166 1163 1332 Export 1125 1468 1814 1776 Transit 4824 5393 4466 3474 Total 14032 15586 14780 14076

Tonne-kilometres (mill) Local 1916 2160 2194 2447 Import 513 529 545 672 Export 1095 1272 1530 1398 Transit 5355 5987 4957 3856 Total 8878 9948 9226 8373

Passengers (000) International 175 186 192 207


DRAFT

BULLPIN

20

Local 4152 4042 4131 4267 Total 4326 4227 4323 4474

Passenger-km (mill) 1219 1234 1287 1401

Traffic has reduced slightly in the past two years, as transit freight has fallen, partly being diverted to direct crossing point in Manchuria. The reduction in transit freight is particularly damaging, as in 2005 it provided 73 percent of the freight revenue from 60 percent of the ton-kilometres, with a unit rate over twice that for local traffic.

At the end of 2005, the railway had 122 locomotives, 2,690 wagons and 322 passenger coaches. As in most CIS countries, the locomotive fleet is old, on average about 20 years, and in spite of some modernisation of the fleet, there is a continuing need for modern, reliable locomotives to handle the increasing traffic. This is partly addressed through the Millennium Challenge project (Section 4.3.1)

Coal tariffs are typically 6-7 tugrik/ntkm, equivalent to US$0.5 cents/ntkm. This is almost certainly a loss-making rate when asset depreciation is taken into account and an operating cost of US$ 1 cent/ntkm has been adopted in the analysis in Section 6.3.7. This is also assumed to be the cost to a private operator if it runs its own train and pays access charges to MTZ.


DRAFT

BULLPIN

21

3. SOUTHERN GOBI DEVELOPMENT PLANS

3.1 INTRODUCTION

This chapter first gives an overview of current coal and mineral production in Mongolia. It then describes the main proposed mining projects in the South Gobi, their planned production levels and start dates and the forecast transport volumes directly associated with the mines. Finally, it summarises the development associated with the mine and makes indicative forecasts of inter-regional passenger and freight movements.

3.2 OVERVIEW

Mining is the most important economic sector in Mongolia, accounting for about 27 percent of gross domestic product and nearly 80 percent5 of export revenues in 2007. It directly employs over 44,000 people. It possesses extensive mineral resources, with over 6,000 known showings/deposits of 80 different minerals. The most economically significant of these are coal, base metals (primarily copper and zinc), gold, and fluorite. It also produces limited amounts of tungsten, salt, clay, lime and aggregates. Table 3.1 summarises production and exports of the key minerals since 2002.

Table 3.1 Coal and mineral production and exports in Mongolia 2002-7 (000 tonnes p.a.)

2002 2003 2004 2005 2006 2007 Production

Coal 5544 5666 6865 7517 8074 9238 Copper concentrates 376 372 371 361 371 372 Gold concentrates 12 11 19 24 23 17 Zinc concentrates - - - 23 110 155 Fluorspar 514 488 468 508 522 638 Fluorspar concentrates 148 134 138 132

Exports Coal - 435 1635 2218 2380 3269 Copper concentrates 549 563 563 587 600 608 Zinc concentrates - - - - 106 133 Gold concentrates 12 14 19 24 15 12 Fluorspar concentrates 193 188 289 311 355 360

Source: IMF country report 2008 and Mongolian Statistical Yearbook 2007

3.2.1 Coal

Although coal production has increased sharply in Mongolia since 2004 as exports to China have increased, it is still relatively small by world standards, and particularly when compared to production in Russia and China. However, Mongolia has large deposits of relatively low-cost coal, with brown coal in the central and eastern areas and bituminous coal in the central and western areas. In total, there are 15 coal basins, with potential reserves of 150 billion tonnes and 22 billion tonnes of proved reserves. Approximately 40 coal deposits are currently

5 12 percent from precious metals and 67 percent from other mineral products


DRAFT

BULLPIN

22

being mined in Mongolia, mostly through open-pit operations, with only two underground mines. Around 96 percent of total production comes from seven large mines (Table 3.2).

Table 3.2 Coal production in Mongolia 2006 (000 tonnes p.a.)

Mine Location Capacity Production Coal type Rail accessBaganuur Baganuur city 4000 2881 Lignite Yes Shivee Ovoo Shivee Gobi, Gobisumbur aimag 2000 1200 Lignite Yes Sharyn Gol Selenge aimag 1100 710 Lignite Yes Tavan Tolgoi Omno-Govi aimag 1000 405 Bituminous No Aduunchuluun Dornod aimag 600 226 Lignite Yes Eldev Dornogovi aimag 1000 407 Bituminous Road feederNariin Sukhait Omno-Govi aimag 5000 1724 Bituminous No Others - 1300 282 Lignite Yes Total 16000 7835 Source: Various, all ultimately sourced from Ministry of Natural Resources

The production from the South Gobi (Omno-Govi) mines is exported by truck to industries in adjacent regions of China but the remainder is almost all used for electricity generation (84%), steam heating and industrial uses.

The Baganuur deposit is located 110 km east of Ulaanbaatar and has been operated since 1978. The mine, which produces from three lignite seams, is served by a branchline linked to the Transmongolian railway (TMR) at Bagakhangai.

Shivee Ovoo is located 250 kilometres south of Ulan Baatar on the TMR and was opened in 1992 with the strategic purpose of supplying the expanding domestic demand for coal in Mongolia. It has a simple geological structure and large coal reserves, estimating at more than 2 billion tonnes.

Sharyn Gol is located about 100 kilometres north of Ulan Baatar in Selenge aimag. It is connected to the TMR by a 64 kilometre branch line and supplies power plants in Erdenet and Darkhan. It is an open pit mine which commenced in 1965 but production at this mine has gradually decreased to 0.7 million tons per year, because of geological conditions and operational problems

The Tavan-Tolgoi coal deposit is unique in Mongolia for both its size, 90 km2, and the high quality of its coal. The deposit is located in the Ulannuur coal-bearing depression, 560 km south of Ulaanbaatar and 90 km east of Dalanzadgad, with 16 coal seams ranging in thickness from 2 to 72 metres (totalling 163 metres), with a maximum depth of 945 metres.

Aduunchuluun is a lignite mine supplying Choibalsan in the Dornod aimag and particularly the local power plant.

Eldev is a new mine in the Alagtogoo bituminous coalfield 300 kilometres south-east of Ulan Baatar, located 20 kilometres by road from Olon Ovoo on the TMR. It has reserves of 32 million tonnes within the Alagtogoo coalfield and has been operating since around 2001. It is now exporting 250,000 tonnes p.a. to China via the TMR.

The Nariin Sukhait coal deposit in Omnigovi aimag is in one of three major coal-bearing basins in the Gobi region, approximately 40 kilometres north of the Shivee Khuren - Ceke crossing point on the Mongolia-China border. Metallurgical coal from the Nariin Sukhait Mine is being trucked to Ceke and railed 450 km from there to a Chinese steel mill at Jiuquan


DRAFT

BULLPIN

23

in Gansu. There is a single thick seam of low-ash, high-rank, bituminous coal extending from surface to depths ranging between 120 and 150 metres, with a thickness averaging around approximately 60 metres. It continues across a major coal basin that stretches a total of 120 kilometres east and west of the mine. The mine, reported to hold 134 million tonnes of reserves, has two open pits.

3.2.2 Other minerals

Although there are over 140 registered mining projects in Mongolia, the non-coal sector is dominated by the Erdenet copper/molybdenum mine, the fluorspar mines of Mongolrostsvetmet, and the top five placer gold mining companies, accounting for 81 percent of 2001 output. Copper, molybdenum, fluorspar and placer gold account for 99 percent of output with the remainder from small tungsten and industrial mineral operations.

Table 3.3 Major mineral mines and projects in Mongolia

Deposit Minerals Location Operational Erdenet Copper, Molybdenum Erdenet city Yes Oyu Tolgoi Copper, Gold Omni-Govi aimag No Burenkhaan Phosphate Huvsgul aimag No Tumurtei Iron Selenge aimag No Tumurtiin Ovoo Zinc Suchbaatar aimag Yes Asgat Silver Bayan-Ulgii aimag No Tsagaan Suvraga Copper Dorno-Govi aimag No Bor Undur Fluorspar Khentii aimag Yes Boroo Gold Selenge aimag Yes Gatsuurt Gold Selenge aimag No Hailaast Gold Tuv aimag Yes

The Erdenet copper-molybdenum mine, 380 kilometres northwest of Ulaan Bataar, is Mongolia’s only copper mine and has been in operation since 1978, earning about half of its foreign exchange and providing almost 25 percent of government revenues. The Erdenet Mining Corporation is a Russian-Mongolian joint venture company, employing over 6,000 people. It operates a conventional open pit with proven mineable reserves estimated at about 1.5 billion tons at 0.52 percent Cu. Current grades are higher than this (0.61 percent in 2002) but are expected to decline to the average of the ore reserve. It is a high-cost produced by world standards but has been profitable in recent years and is working to improve mineral recoveries, labour productivity and down stream processing.

The Tumurtiin zinc mine in Sukbataar aimag, some 500 kilometres south-east of Ulaan Bataar, began production in 2005 and plans to produce an average of 66,000 tons of zinc concentrate over its 25-year life. It transports its concentrates by truck 360 kilometres west to the TMR at Sainshand and on by rail to smelters in China.

Mongolia ranks fifth behind China, Mexico, South Africa and Russia in world fluorite production with annual production of over 500,000 tons of ore and 180,000 tons of concentrate. Mongolia possesses twice the fluorspar reserves of China, yet only produces 6 percent of China’s annual production. Four of the six mines in Mongolia are operated by Mongolrostsvetmet accounting for over 90 percent of production; the largest is at Bor-Under, 380 kilometres south-east of Ulaan Bataar and served by a 60 kilometre branch from the


DRAFT

BULLPIN

24

TMR, which produces about two-thirds of all Mongolian production. Annual production levels are expected to remain static at around 110,000 tons of chemical-grade and 80,000 tons of metallurgical-grade concentrates in the short to medium term.

Practically all recorded gold production in Mongolia comes from placer mines operated by Mongolian and Mongolia–Russian joint venture companies, although a very small number of hard-rock operations extract limited amounts of the metal. Most placer operations are found in the Zaamar, Bayangol, Tolgoit and the North-Central goldfields. Output is declining as large readily accessible deposits with rich ore seams are depleted and smaller mines are being exploited, leading to increased unit costs and lowered ore grades. In 2007, recorded production was 12 tonnes, derived from a large number of registered gold mines widely distributed across the country. The largest 7 mines produce about half of annual gold output; over 40 percent of the production comes from Tov aimag and a further 35 percent from the Selenge, Darkhan Uul and Arkhangai aimags. Large numbers of artisan miners also extract gold from tailings. Overall gold production in Mongolia increased between 2002 and 2005, with the commissioning of new gold mines at Boroo, a medium-sized deposit 135 kilometres north of Ulaan Baatar, Bumbat, 210 kilometres west of Ulaan Baatar, in Tuv aimag, and Olon Ovoot, located 550 kilometres from Ulaanbaatar and 100 kilometres from Dalanzadgad in the South Gobi, but has since declined to the 2002 level.

3.3 KEY MINING DEVELOPMENTS IN SOUTH GOBI

3.3.1 Overview

Table 3.4 shows the major coal and mineral projects currently planned for the South Gobi. It includes estimated lives and assumed annual production and start dates. These can be expected to change as the projects are fully developed; in addition, as the coal deposits, in particular, are only partially explored, further developments can be expected in the medium-term.

Table 3.4 Coal and mineral projects in South Gobi

Mine Life (years) Production (000 t p.a)

Mineral Start date

Tavan Tolgoi 200+ 15000 Coal 2012 full Nariin Sukhait 40 2000 Coal 2003? Ovoot Tolgoi 50 5000 Coal 2008 Sumbar 50 5000 Coal 2015? Baruun Naraiin 20 6000 Coal 2012 Tsagaan Tolgoi 20 2000 Coal 2012? Oyu Tolgoi(1) 15-50 650 Copper 2012 Tsagaan Suvraga(1) 30 150 Copper 2015?

(1) Also small quantities of other metals; copper assumed to be transported as concentrate (30% copper)

Source: Company websites and various

3.3.2 Timing

Tavan Tolgoi is a major coal deposit, including both coking and thermal coal, located about 150 kilometres north of the Chinese border and the same distance south east of Dalanzadgad. It contains over 4.5 billion tonnes of established reserves, of which 1.9 billion tonnes are coking coal.


DRAFT

BULLPIN

25

Nariin Sukhait is one of approximately 40 coal deposits presently being mined in Mongolia. The Nariin Sukhait coal deposit is within one of three major coal-bearing basins located almost entirely on lands controlled by Ivanhoe in the Gobi region. Metallurgical coal from the Nariin Sukhait Mine is being shipped to a Chinese steel mill. The coal seam, one of five conformable seams identified to date at Nariin Sukhait, has been mapped in outcrop and sub-crop throughout a major coal basin that stretches a total of 120 kilometres, east and west of the mine, on ground controlled by Ivanhoe. There is a single thick seam of low-ash, high-rank, bituminous coal extending from surface to depths ranging between 120 and 150 metres, with thickness ranging from 32 to 90 metres, with an average of approximately 60 metres.

Ovoot Tolgoi is an extension of the Nariin Sukhait deposit, located about 45 kilometres north of the Chinese border at Ceke. The coal has the same broad characteristics as the Nariin Sukhait coal and production, which commenced in April 2008 will be delivered to similar customers, power plants and steel plants in northern China, particularly in Gansu some 450 km to the south. Coal will initially be trucked to the border at Ceke, where a coal loaded handling the Nariin Sukhait production is already in operation. Measured and indicated reserves are estimated at 250 million tonnes, in two main fields; longer-term plans include the development of an underground mine. Planned annual production is not known but has been assumed at 5 million tonnes.

The Sumbar deposit lies 16 kilometres east of Ovoot Tolgoi. The deposit is currently being delineated but initial results show coal of similar quality to Ovoot Tolgoi, although the seam is not quite as thick.

Baruun Navan is located about 20 kilometres west of Tavan Tolgoi with a reported potential production of about 120 million tonnes of coal, of which 70 million is coking coal and 50 million tonnes is thermal coal. It is expected to take 2 years to construct the mine, with production commencing in 2011 or 2012. The annual production is planned at 6 million tonnes, giving a mine life of 20 years. The potential customers are power plants and steel producers in northern and north-eastern China, who would be served by an extension of the Tavan Tolgoi rail line.

The Oyu Tolgoi gold, copper and molybdenum prospect is a world-class deposit approximately 560 kilometres due south of Ulaan Baatar and 200 kilometres south-east of Dalanzadgad. The deposit has inferred resources of some 489 million tons of ore, grading 1.08 percent copper and 0.07 g gold per tonne, i.e. about 5 million tonnes of copper and 30 tonnes of gold but major extensions of the ore body have been identified which will significantly increase these figures. Copper concentrates from the mine will probably be sold to Chinese smelters.

Tsagaan Tolgoi is a small (40 million tonnes of reserves) coal deposit about 100 kilometres west of Oyu Tolgoi which has been suggested might be a source for the power plant supplying electricity to the Oyu Tolgoi development.

The region surrounding Tsagaan Suvarga, approximately 150 km northeast of Oyu Tolgoi in Dornogovi aimag, contains seven significant copper-molybdenum deposits, of which the most significant is at Surven-Sukhait. This deposit is 164 kilometres southwest of the Zuun-Bayan station, at the end of a 50 kilometre branch from the TMR. Commercial ore reserves of the


DRAFT

BULLPIN

26

deposit have been estimated at well over 220 million tons at 0.54 percent copper and 0.019 percent molybdenum, similar grades to those at Erdenet, although the reserves are only 15% as large. As yet, there are no firm plans for its development.

3.4 SECONDARY DEVELOPMENT

The forecast direct employment when the projects are in production is summarised in Table 3.5. These will live on adjacent to the mine site; practice in many countries is to work on a week-on, week-off basis (or similar) and this would mean that approximately half the staff would be living at any one time at the minesite and the others at a local centre such as Dalanzadgad.

Table 3.5 Forecast direct employment in South Gobi projects (post-construction)

Mine Production (000 t p.a)

Employment

Tavan Tolgoi 15000 1200 Nariin Sukhait 2000 150 Ovoot Tolgoi 5000 400 Sumbar 5000 400 Baruun Naraiin 6000 500 Tsagaan Tolgoi 2000 150 Oyu Tolgoi(1) 2000 3000 Tsagaan Suvraga(1) 500(2) 1000 Total 6800

(1) Production figure is for concentrate (30% copper)

(2) Assumed for planning purposes

Source: Company websites and various

There will also be indirect employment generated by the projects. The ratio of total employment created to direct employment in the projects is known as the employment multiplier. Estimates of the value of this vary widely from under 1.5 to 4 or more. The lowest estimates are associated with ‘fly-in fly-out’ mines with long-distance commuting and most mine services sourced from major centres. There is considerable doubt on the validity of the higher estimates and the current consensus appears to be about 1.7-2.2; a value of 2 has been adopted for the purposes of these forecasts (i.e. one non-mining job created for each project job). Assuming a workforce participation rate of 60% gives a total population increase attributable to the projects of about 25,000 and a total population in South Gobi of about 75,000.

3.5 FUTURE TRANSPORT DEMAND

The outputs from the various mines summarised in Table 3.4 clearly show that the principal economic driver of new rail construction will be the coal flows. Whilst the base metal mines have the capacity to earn large amounts of revenue, the volume of freight they generate is relatively limited and, even for Oyu Tolgoi, would be unable to justify significant new construction, whilst a regional population of less than 100,000 would likewise be much more economically served by truck and bus rather than a new railway.


DRAFT

BULLPIN

27

The scale of the coal reserves is such that future transport volumes will largely be determined by demand, especially from northern China, rather than being constrained by availability of supply. Demand for coking coal, in particular, is likely to be an important factor.

Historically, China produced all the coking coal it required, with very limited imports. With the rapid increase in steel production, imports have increased but these still remain a very small proportion of total demand. In 2007, estimated production of metallurgical coal was 365 million tonnes with exports of 6 million tonnes and imports of 3 million tonnes. The production of steel is expected to continue to increase at about 8 percent p.a. for at least the medium term and this is expected6 to lead to continued growth in imports, to around 20 million tonnes by 2013. Mongolia is well-placed to capture the majority of this import demand. Whilst a significant proportion of China’s imports in 2005 and 2006 came from Australia, the world’s largest coking coal exporter with an estimated 60 percent of the world market, Mongolia has rapidly become the most important supplier and so far in 2008 has supplied 50 percent of coking coal imports, and over 75 percent in June and July.

Given its geographical proximity, it is reasonable to assume that Mongolia will provide all China’s coking coal requirements, other than a few specialised shipments. Whether it can supply more, and squeeze out domestic production, is a moot question. The largest coking coal production in China is from southern Shanxi and, in eastern China, Huaibei in Anhui.

Any additional coal that is imported into China from Mongolia will thus either be exported direct or displace other coal which will then be in turn exported. Its main potential markets will be the same as Chinese exports: Japan, Taiwan and South Korea. Its main competitors in these markets, for both coking and thermal coal, will be Australia and Indonesia; currently Chinese coal enjoys a $10/tonne freight advantage to Japan over Australian coal and Mongolian coal will thus have to be mined and transported to the coast with an FOB price no more than $10/tonne above that for Australian coal. Although Australian FOB prices in 2008 ranged from about $80/tonne for thermal coal to $150/tonne for coking coal, the long-term level is likely to be much lower and the FOB cost is no more than $50/tonne. The competitiveness of Mongolian coal in the North Asian markets will thus critically depend on the cost of rail transport to either Russian or Chinese ports.

China is not self-sufficient in copper, reportedly producing only about 30% of its concentrate requirements. It currently imports about 4-5 million tonnes of concentrate7 and about 1.5 million tonnes of refined copper, equivalent to around 3 million tonnes of copper, or 10 million tonnes of concentrates. Oyu Tolgoi should thus be able to readily export its entire production of copper concentrates (around 2 million tonnes p.a.) for the foreseeable future. Unless volumes are very high, these are generally transported by road in other countries if there are no existing rail lines. However, where a rail line exists, rail will almost always carry the product.

For planning purposes, the forecast volumes given in Table 4.6 have been assumed for export coal from South Gobi.

6 See, for example, Australian Commodities , March 2008, Australian Bureau of Agricultural and resource economics (ABARE) 7 Assuming copper concentrates are 30% copper, this is equivalent to around 1.5 million tonnes of refined copper


DRAFT

BULLPIN

28

Table 3.6 Assumed mineral volumes ex South Gobi 2005-2025 (million tonnes p. a.)

Medium-term (2015)

Long-term (2025)

Coal China via Ceke 5 7 China via other 10 20 Overseas 10 20

Copper concentrates ex Oyu Tolgoi 2 2 Total 27 49

The mines will also require inputs of mining supplies and fuel. However, these are generally an order of magnitude smaller (typically 5-10%) than the mine output unless there is a smelter attached to the mine and, with the exception of fuel, are generally not well suited to rail transport. In most other comparable situations (e.g. South Africa, Zambia and Australia), mining supplies are generally collected from a number of suppliers and delivered as a mixed load by road as long as there is road access. These inputs therefore will rarely be a deciding factor as to whether or not a railway is constructed.

In addition to the mines themselves, there will be demand for passenger and freight transport from the settlements in South Gobi. However, again this is unlikely to be significant in terms of deciding whether or not a line is built. Trip rates between the region and Ulaanbaatar are unlikely to exceed 0.5 trips per capita p.a., say 50,000 p.a. Allowing for trips by car and plane, this can be easily handled by 3 or 4 buses daily. In any event, unless a direct rail line is built between the region and Ulaanbaatar, a rail trip will be around 1000 kilometres compared to around 500 km by road. Non-mine freight to and from the region is likely to be around 0.5-1 tonne per capita p.a., equivalent to around 50,000 tonnes p.a.; this also is too low to support a rail link and most, if not all of this traffic is likely to be transported by road.


DRAFT

BULLPIN

29

4. SECTOR DEVELOPMENT PLANS

4.1 INTRODUCTION

This chapter summarises the current network development plans of both the road and railway sectors, with special reference to the South Gobi, as background to the discussion of specific projects in that region to support mining development.

4.2 ROAD MASTER PLAN

In 1993-4 a road master plan (RMP) was prepared with assistance from the ADB which has since been largely implemented. The plan formulated a policy for both the road sector and transport in general and developed a program for improving 5362 kilometres of road at a cost of $427 million. At that time (1993) the total improved road network of the country was only 3097 kilometres, of which 1018 kilometres was paved and 2079 kilometres was gravel. By 2006, 2027 kilometres of the 5362 kilometres in the plan had either been improved or were committed to be improved and the target for 2010 is to complete 3479 kilometres.

In [2005] GOM formulated a National Development Policy (NDP), covering all economic sectors, between the years 2006 to 2021. Its objectives for the road sub-sector include the continuing development and improvement of the corridors linking to the international highways, constructing the three Asian Highways (AH3/AH4/AH32) (collectively known as the ‘Millennium Road’ project) and continuing the periodic maintenance of state highways and local roads. These objectives were translated into a set of staged development targets specifying the length of road to be improved and the condition that is maintained at by 2021; the proportion of the state road network (SRN) that is to be maintained at, or better than, ‘Good’ condition is to increase from 30 percent in 2006 to 70 percent by 2021, with 20 percent in ‘average’ condition from 2016. To achieve this, GOM policy to improve all unpaved roads directly to asphalt, rather than use the intermediate option of gravelling.

The achievement of these objectives was addressed in more detail by an RMP prepared with World Bank assistance in 2007. The key feature of the RMP is provide a sealed road between Ulaanbaatar and each aimag centre and provide full connectivity between the international network and the national roads serving the five economic regions of Mongolia. The plan has three stages (in addition to the currently committed roads program):

• Immediate (2008-2010): this includes periodic and rehabilitation works to protect the sections of the existing paved road network which are currently in ‘good’ condition from further deterioration and widening a number of roads to improve safety and increase design speeds. It also includes preparation work for the subsequent stages by adopting new lower-cost pavement technologies, expanding the strategic planning capacity of the MRTT and the further development of road and bridge information management systems linked with weigh-in-motion traffic sensors and the installation of heavy vehicle weigh bridges.


DRAFT

BULLPIN

30

• Short-term (2011-2015): this includes the periodic maintenance and rehabilitation of roads currently in ‘fair’ to ‘poor’ condition because of the maintenance backlog and continuing the 1993-1994 plan for asphalt construction. It also includes the replacement of a number of existing wooden bridges by concrete bridges to maintain design speeds and ensure road user safety.

• Medium-term (2016-2020): this focuses on increasing road connectivity between the eastern and western regions of the country and creating a fully-developed all-weather national road network.

Table 4.1 summarises the planned output and cost of the RMP.

Table 4.1 Currently committed and RMP 2007 programs and cost 2008-2020 (2006 prices)

Unit Cost ($US million)

Unit cost ($ 000/unit)

Currently committed (km) 2455 Recurrent maintenance - 237 - Capacity expansion (km) 352 47 134 New construction (km) 5770 1438 249 Bridges (number) 190 58 305 Other - 3 - Total 1783 -

Recurrent expenditure covers routine maintenance, periodic maintenance (assumed to be done every 8 years at a cost of $50-60,000/km for a sealed road) and $42 million to address deferred maintenance.

The main component of the RMP is the upgrading of earth roads to fully engineered 7 metre /10 metre asphalt concrete standard, providing all-weather connectivity between Ulaanbaatar and twenty one aimag centres and access for each of the five economic regions to an international border post.

The total cost of the program was estimated at US$1.8 billion. This cost represents approximately 3.3 percent of the country’s forecast GDP over the period (assuming 6 per cent annual growth), which is high, but not enormously so by world standards, and will be partly funded from specific users such as the mining sector. The RMP states that, where possible, GOM should encourage resource extraction investment projects to upgrade or construct roads used by a mine to transport its output to the processing site. These roads can then be handed over back to the GOM who would then operate the road on a toll basis to at least cover the road’s recurrent maintenance cost.

Table 4.2 summarises the projects within the South Gobi region included in the RMP.


DRAFT

BULLPIN

31

Table 4.2 RMP projects affecting South Gobi (2008 prices)

Date Distance (km) Cost ($US million)

Ulaanbaatar – Mandalgovi 2008-10 230 BOT? Dalanzadgad – Gashuun Suhait 2008-09 329 99 Ovoot Tolgoi – Chinese border 2009 [40?] 12 Mandalgovi – Dalanzadgad 2011-15 293 88 Total 892

It does not include any projects connecting Dalanzadgad towards the east and a junction with the trans-Mongolian Highway north of the Chinese border at Zamyn-Uud. The costs of these roads are estimated from the 2006 unit costs in Table 4.1, increased by 20% to update to 2008 price levels (i.e. an average cost of $300,000 per kilometre).

Figure 4.1 RMP projects 2008-20

4.3 RAIL INFRASTRUCTURE PLAN

4.3.1 Millennium Challenge

The dual ownership of MTZ has created considerable difficulties for Mongolia, not least in terms of agreeing an investment and upgrading plan, as the Russian interest would be liable for 50% of any investment cost, given that the railway is not capable of financing any significant expenditure from retained earnings. In 2007, the US Government, acting through the Millennium Challenge Corporation (MCC), provided a grant of US$ 285 million to GOM, of which US$188 million was railway-related. The rail component includes:

• the acquisition of around 30 freight locomotives, up to 75 new open top freight wagons, up to 75 new specialized freight wagons, track maintenance equipment, and a modernized signalling and communications system for the mainline;

• supplying these assets to a new, initially government-owned, contractor-operated leasing company (‘LeaseCo’) which will be established to lease them on commercial


DRAFT

BULLPIN

32

terms to MTZ and independent shippers. The project will support a review to determine the best method for the creation and operation of LeaseCo and assist in forming its board of directors, staffing its management unit, and contributing towards its start-up costs. It will also assist in contracting a private sector firm to manage and operate LeaseCo (‘OpCo’) and organising oversight of OpCo by an outside project management firm together with GOM.

• operational training and financial management technical assistance to MTZ

• technical assistance to the Mongolia Railway Authority (‘MRA’) to strengthen its technical capacities in areas such as rail safety, pricing, and track access licensing;

• technical assistance to the Customs General Administration to strengthen its capacity relating to the rail sector and the international transport of natural resources; and

The covenants to the grant include MTZ upgrading approximately 35 km of track to 65 kg/metre rails for each of 2007 and 2008 and approximately 50 km of track annually thereafter. MTZ is also committing to lease the rail equipment acquired under the grant from LeaseCo at fair market rates and allowing LeaseCo to lease such equipment to other shippers and to allow such equipment to operate on MTZ's tracks.

4.3.2 Infrastructure upgrading and new construction

Millennium Challenge quite specifically does no include any component related to duplication and/or upgrading of the main line, other than the signalling equipment, not to any other new construction. Whilst upgrading of the mainline is being addressed, albeit rather slowly, through the upgrade to 65kg/metre rail, this will take about 15 years at the rate of progress included in the MCC agreement.

Duplication seems to be even further away unless a major source of third-party funding can be identified. Whilst upgrading to 65kg/metre rail is costing around $300,000 per track-km, duplication would probably cost between $1-3 million per track-km, depending on how much realignment and reconstruction was carried out simultaneously. The need for any such duplication depends on the rate of growth of traffic; local traffic is mostly coal to and from power plants and is reasonably predictable. Transit traffic is less so; there is an alternative route through Manchuria which is already used by the majority of traffic ex China and the two main commodities ex Russia are timber, which is subject to export quotas, and oil, for which an alternative pipeline is under construction. As current traffic levels are only about 60% of the TMR capacity, particularly with the improved signalling, there is ample room for traffic growth unless large additional quantities (say 5-10 million tonnes p.a.) of export minerals are transported on TMR. If these are from southern Mongolia to China, the bottom 250 kilometres from around Sainshand will need to be duplicated; if they are exported to Russia either the top 870 kilometres will need to be duplicated or a new route via Choibalsan will need to be constructed.

The difficulties in investing in infrastructure under the current ownership of MTZ have led to suggestions that a second independent track could be constructed parallel to the existing one, but owned and possibly operated by a 100% Mongolian entity. This solution appears fraught with many potential problems, particularly as Mongolia relies heavily on feeding traffic to


DRAFT

BULLPIN

33

and from Russia; a more practical solution if the Russian partner in MTZ is reluctant to invest might be for Mongolia to invest by itself and for MTZ to then lease the second track and pay lease charges through revenue from access charges.

A number of new lines have been proposed at various times, some of which are more likely than others. Figure 3.2 shows the network presented in the National transport Strategy prepared in 2007. Most of the lines shown only likely to be built if there are major mining developments which generate substantial export quantities of product; in practice this will mean coal. The map shows connections between Tavan Tolgoi and Oyu Tolgoi and the Chinese border but also lines linking to Sainshand and Mandalgovi, where there is a junction with an east-west line traversing the whole breadth of Mongolia.

This network shown in Figure 3.2 must be regarded as a very long-term plan, of which the lines needed to develop the South Gobi in the short and medium term are only a small subset. Proposals for short and medium-term development which have been presented to various seminars are discussed in more detail in Chapter 5.

Source: National Transport Strategy for Mongolia ADB 2007 Figure 4.2 Possible new rail construction


DRAFT

BULLPIN

34

5. ADJACENT NETWORKS

5.1 INTRODUCTION

This chapter describes the Chinese and Russian rail networks connecting to possible rail infrastructure in Mongolia, the regional port infrastructure and capacity and the estimated costs for exporting Mongolian minerals through Chinese or Russian ports.

5.2 CHINESE NETWORK

5.2.1 Overview of the regional rail network

China’s rail network is extensive but also highly congested and there is very little spare capacity on key routes, especially those crossing the mountains just to the west of a line running north-south between Beijing and Guangdong. Coal is about 40 per cent of the traffic carried on China’s railways and, with its strong growth in recent years, which is forecast to continue for at least the next decade, is the single most important factor affecting network capacity.

The influence of coal on China railways is exacerbated by the location of the coalfields. Although coal is widely distributed throughout most of China, and most countries have a significant production, the major coalfields are all located west of the mountains. In the Eleventh Five-year Plan (11FYP) the provinces have been classified into three groups:

• Those which need to import coal from other parts of China; this is virtually everywhere east of the mountains

• Those which are self-sufficient; this includes the south-west and also western China (Gansu, Xinjiang, Qinghai)

• Those which are net coal exporters; this covers Shanxi, Shaanxi, Nei Monggol, Ningxia and, by extension, Mongolia

Figure 5.1 shows the major coal consumers within northern China, together with the main railway routes of the coal network as in 2006. Several power stations are located on the Shaanxi and Hebei coalfields, together with western Shandong, with others near the major population and industrial centres of Beijing, Tianjin and Liaoning. The coking plants are located either at major steel producing centres, such as Baotou in Nei Monggol and Anshan in Liaoning, or on the main coking coalfields in Taiyuan and southern Shanxi.

The major coal ports are located along the east China coast from Jinzhou in the north to Lianyungang in the south. Although China typically exports 50-70 million tonnes of coal annually, mostly to Japan and South Korea, the main coal traffic through these ports is domestic coal carried by coastal shipping to Jiangsu, Shanghai, Zhejiang, Fujian and Guangzhou.

The majority of this coal is carried from the main inland fields in Shaanxi, Shanxi, Ningxia and Nei Monggol, mostly by rail but with a significant volume by road where the railway is at


DRAFT

BULLPIN

35

capacity. Nine railway lines, forming three groups, currently cross the mountains separating these fields from the coastal plain:

• The northern route which carries coal from northern Shanxi, northern Shaanxi, Ningxia and Nei Monggol to Qinhuangdao and Huanghua. This group includes the two most heavily used lines: the Daqin line and the privately-owned and operated Shenshuohuang line

• The central route from central Shanxi to the Shandong ports

• The southern route from southern Shanxi and southern Shaanxi to Rizhao and Lianyungang.

Of these, the northern corridor is by far the most heavily used: in 2005 the northern corridor carried over 500 million tonnes, compared to about 75 million tonnes on the central corridor and 110 million tonnes on the southern corridor. These volumes increased significantly in 2006 and 2007 and the Daqin line is now carrying around 400 million tonnes by itself.


DRAFT

BULLPIN

36

Figure 5.1 Major centres of coal demand in northern China 2006 Source: Based on Barlow Jonker unpublished report 2006

Projections presented at an NDRC seminar in 2006 forecast that shipments from the main western coalbases (Shaanxi, Shanxi, Nei Monggol and Ningxia) would increase from around 540 million tonnes in 2005 to 1280 million tonnes by 2020; shipments through the northern ports would increase from 390 million tonnes to over 690 million tonnes. This requires


DRAFT

BULLPIN

37

capacity over double that which existed at that time and it is not surprising that the railway Medium and Long-term Development Plan (MLTDP), which was prepared in 2004 and covers the period to 2020, places considerable emphasis on expanding coal transport capacity.

Table 5.1 gives the seven main coal ports on the east coast, together with indicative estimates of the port capacity expected in 2010. The bulk of the 2005 throughput was domestic transhipment to eastern and southern China; exports totalled only about 70 million tonnes, about 20 percent of the total port coal throughput.

Table 5.1 Coal shipment ports – East China

Coal shipment 2005 (tonnes

million)

Capacity (tonnes million

2005)

Capacity (tonnes

million) 2010 Qinhuangdao 145 137 187 Tangshan 14 30 110 Tianjin 80 73 100 Huanghua 67 75 110 Qingdao 9 23 23 Rizhao 20 25 35 Lianyungang 14 22 40 Total 350 385 605

5.2.2 Chinese railway development plans

Plans to expand rail capacity through the mountains were set out in some detail in the MLTDP. Figure 5.2 shows the routes, with the red line showing the screenline marking where the mountains are crossed.


DRAFT

BULLPIN

38

Figure 5.2 Coal transport routes between Shaanxi and eastern China 2008

The corridors from the Datong region, covering Datong, Shuozhou, Xinzhou and the west of Inner Mongolia (excluding Dongsheng) include:

• Datong-Qinhuangdao (653 kilometres), an electrified double-track line equipped with automatic blocking signalling (ABS). Its 2008 freight transport capacity is 400 million tonnes and it is effectively at capacity

• .engtai-Shacheng-Datong (Fengshada) (354 kilometres), an electrified double-track line with ABS. Its freight capacity is 65 million tonnes (allowing for 24 pairs of passenger trains) and it is at capacity. Coal represents about 45 percent of its traffic. A relief line is being constructed from Jining to Zhangjiakou and the remainder upgraded to give a freight capacity of 85 million tons.

• Jining-Tongliao (Jitong) (942 kilometres) is a single non-electrified line equipped with semi-automatic blocking (SABS). Its freight capacity is 10 million tonnes and it will be upgraded to increase its capacity to about 15 million tonnes.

The Shenfu region mostly consists of mines operated by the Shenhua Corporation. Output is expected to reach 180 million tonnes in 2020, of which 160 million tonnes will be carried by rail to Huanghuagang, south of Tianjin. This line is a privately-constructed and operated electrified double-track railway, equipped with ABS. It is currently carrying around 100 million tonnes on its mainline and the line will be upgraded to increase its capacity to at least 200 million tons. In addition to the mainline, there are two feeder lines (one connected to


DRAFT

BULLPIN

39

Daqin) and, importantly for Mongolian coal, the Baoshen line from Baotou to near Shenmu (Figure 5.3).

Figure 5.3 Shenhua network and mines 2008 The Taiyuan region covers mines in Xishan, Fenxi, Huozhou, Yangquan, Taiyuan, Jinzhong and Lüliang and is the main source of coking coal in China. The existing and proposed coal transport corridors are:

• Shijiazhuang-Taiyuan (231 kilometres), an electrified double track line equipped with ABS. The line has a freight capacity of 75 million tons is operating at capacity. A new 200 km/hour mixed-traffic line is being constructed from Shijiazhuang to Taiyuan which will more than double capacity. The connecting lines eastward from Shijiazhuang-Dezhou and Qingdao-Jinan will also be electrified.

• Houma-Yueshan (252 kilometres), an electrified double-track line equipped with ABS. The line has a freight capacity of 80 million tonnes, and carries about 50 million tonnes of coal. It is being upgraded to 100 million tonnes capacity, with 70 percent of this being allocated for coal.

• Handan-Changzhi (220 kilometres), a non-electrified single line with SAB and a capacity of 15 million tons. It carries about 5 million tonnes of coal. Both this line and its extension between Handan and Jinan are planned to be duplicated, with the freight capacity increased to about 80 million tons, 50 percent of which will be allocated for coal transport.

• Beijing-Taiyuan (Jingyuan) (437 kilometres), a non-electrified single line with SAB and a capacity of 18 million tonnes of freight. It is currently operating at capacity, carrying 10-15 million tonnes of coal. It is planned to be upgraded to increase its capacity to 20 million tons.


DRAFT

BULLPIN

40

• A completely new dedicated coal line has also been proposed between Liuliang and either Rizhao or Shandong. This will be majority owned by the three provinces of Shanxi, Henan and Shandong, with MOR owning about 30-35%. The line will be a high-quality double-track electrified line similar to the Shenhua and Daqin lines, with a potential eventual capacity of 200 million tonnes or above.

East and south Shanxi, which covers mines in Jincheng, Luan and surrounding areas, is a major source for coal used in the chemical industry. Output is expected to double to about 200 million tonnes by 2020, of which 115 million tonnes will be transported by rail along two corridors:

• Handan-Changzhi (see above)

• Taiyuan-Jiaozuo (434 kilometres), a non-electrified single-line with SAB as far as North Changzhi and an electrified double-track line with SAB from there to Yueshan. The freight capacity for the southern section is 50 million tonnes it is operating close to capacity carrying over 40 million tonnes of coal. Following electrification of the connecting Xinxiang-Yueshan section, the capacity of the Taiyuan-Jiaozuo line will be increased to about 90 million tonnes.

The Shaanxi region covers the Hancheng, Huangling, Tongchuan and (to be developed) Yulin areas. Coal output is expected to reach 200 million tonnes by 2020, 90 percent of which will be carried by rail on three corridors:

• Zhengzhou-Xian section (511 kilometres) of Liangyungang-Lanzhou (Longhai) line, an electrified double-track line with ABS. The line carries 56 pairs of passenger trains, leaving only about 50 million tonnes of freight capacity and is currently saturated. Construction of a dedicated passenger line linking Zhengzhou-Xian is underway, together with duplication of Baotou – Xian and the existing Zhengzhou-Xian section can then be used for freight transport with a capacity of about 100 million tonnes.

• Houma-Xian (288 kilometres), a non-electrified single-track line with SAB and a capacity of 15 million tonnes, of which about 40 percent is utilized. It is planned to upgrade its capacity 20 million tonnes.

• Xian-Nanjing (1,086 kilometres total length), a recently-constructed single line, of which Xian-Nanyang is electrified. The line has a freight capacity of 20 million tonnes but duplication is planned to increase its capacity to over 50 million tonnes or more.

• Xian-Ankang, an electrified single line equipped with SAB. It has a freight capacity of 14 million tons and is at capacity. It is planned to duplicate the line and increase its capacity to 50 million tonnes.


DRAFT

BULLPIN

41

Table 5.2 Capacity of Coal Transport Corridors by 2020

Coal Transport Capacity (million tonne)

Railway Lines Mileage (km)

Current Status

Upgrading Measures

2002 2020 Increase 1 Beijing-Shacheng-Datong 354 Double line Upgrading 65 100 35 2 Datong-Qinhuangdao 653 Double line Improving 100 200 100 3 Jining-Tongliao 942 Single line Upgrading 10 20 10 4 Shenmu-Suoxian-

Huanghuagang 802 Single &

Double Double tracking

80 200 120

5 Shijiazhuang-Taiyuan 231 Double line Separation of passenger & freight lines

75 100 25

6 Houma-Yueshan 252 Double line Improving 80 100 20 7 Handan-Changzhi 222 Double line Double

tracking 15 100 85

8 Beijing-Taiyuan 437 Single line Electrification

18 20 2

9 Taiyuan-Jiaozuo 434 Double line Improving 50 100 50 10 Lianyungang-Lanzhou 511 Double line Separation of

passenger & freight lines

49 100 51

11 Houma-Xian 288 Single line Improving 15 20 5 12 Nanjing-Xian 1,086 Double

tracking 70 70

13 Xian-Ankang 260 Single line Double tracking

14 50 36

In addition to increasing the capacity across the mountains, China plans to build several other lines that could carry minerals from Mongolia: • The Hami – Linhe line is currently under construction as far as Ceke; this provides a

cut-off for transport between northern China and Xinjiang but also acts as a collector line for traffic feeding in from Mongolia by road or rail, being around 400 kilometres north of the current line from Wuwei to Urumqi

• A number of feeder lines are either under construction or are planned to carry traffic from railheads on the China-Mongolian border. These include (from the west):

o Urumqi – Altai Khan o Ceke – Qingshui (already constructed with a coal loading terminal at Ceke) o Gashuun Sukhait - Linhe (planned to carry traffic from Oyu Tolgoi and

Tavan Tolgoi o Mandula - Baotou o a major new line to carry coal from Bayan Ul in Nei Monggol to Fuxin in

Liaoning. This is planned as Phase 1 of a new line which will be extended to Zhuengadabuqi on the Mongolian border (Phase 2) and then to Choibalsan in Mongolia to provide a third trunk route between northern China and Russia.

These lines are planned, at least initially, as single lines, with a capacity of about 20-25 million tonnes and could typically be built within 18 months – 2 years of being approved. Most of them are likely to be built as either joint venture lines, with the financing shared between the central Government, provinces and potential customers, or as purely private lines.


DRAFT

BULLPIN

42

5.2.3 Chinese port development plans

Table 5.1 shows the 2005 and planned 2010 port capacities at the main Chinese coal ports. Capacity is expected to increase by 200 million tonnes during this period. The bulk of the new capacity will be added at Qinhuangdao (50 million tonnes), Tangshan (80 million tonnes through the construction of a new port at Caofeidian), Tianjin (37 million tonnes) and Huanghua (35 million tonnes, to be provided by Shenhua).

5.2.4 Handling constraints at border crossings

The difference in gauges between the Chinese and Mongolian systems currently requires traffic to be transhipped at the border. This is a time-consuming process, exacerbated by the bureaucratic procedures. The transhipment itself is generally not a problem but it does require adequate supplies of wagons on either sides of the border and this is often reported (particularly on the Chinese side) to add considerably to the delay.

Where the traffic only travels a short distance within one of the countries, a system of dual gauge operation was often used historically. However, when the gauges are as close together as the Chinese and Mongolian, with under 100 mm between the tracks, dual gauge is really only suitable for short distances at relatively slow speeds8. Minerals can be transhipped relatively economically and straightforwardly, either by direct transfer (e.g. one gauge passing over another, with the wagon above discharging directly into the wagon below) or by unloading into a silo and then reloading in a separate operation; this method is used at Dostyk, where around 5 million tonnes of ores are transhipped from the Kazakhstan system to the Chinese system (see Figure 5.3).

Where a line has been dedicated to a particular traffic, however, it has often been built to the same gauge throughout, even though this is different to the rest of the network in one of the countries. This approach was taken in both Poland and Slovakia, where steel-making inputs were imported from Ukraine some hundreds of kilometres on lines which were constructed to Russian gauge rather than the European standard gauge.

If the Chinese lines currently planned and being constructed to the Mongolian border are continued to mines inside Mongolia it seems clear from a railway operations viewpoint they should be built to Chinese gauge and avoid any transhipment. However, traffic which moves into China via Zamyn-Uud and Erlian will almost certainly be on broad gauge and this will require the construction of a substantial facility there for transhipping the coal and minerals.

8 Another alternative is what is known as ‘gauntlet’ track where four lines, two from each gauge, are placed on a single formation

with enough room between them to ensure no objects can get lodged between the pairs of rails.


DRAFT

BULLPIN

43

Kazakh wagon being unloaded on top level into silo

Chinese wagons at ground level being loaded Figure 5.3 Mineral transhipment facility at Druzhba (China – Kazakhstan border)


DRAFT

BULLPIN

44

5.2.5 Operational considerations

Although there are several private (or at least non-China Railways) operators currently operating, most of them do so on their own networks and there are comparatively few private wagons, and almost no private locomotives, operating on the Chinese network. The concept of a Mongolian-owned train, with Mongolian locomotive and wagons, operating on the Chinese mainline network, is thus some way off. China Railways has actively discouraged privately-owned wagons for some years, principally because it wishes to maximise wagon capacity utilisation by, for example, backloading general freight of some sort in coal wagons that arrive at destinations in eastern China.

As yet, there are also no formal arrangements whereby a third party can pay access charges to operate over the main CR network. The normal arrangement is that traffic is carried in CR wagons; whilst they are on non-CR lines, a standard charge is made per day; whilst they are on CR lines the normal CR tariff applies. The customer thus pays the CR tariff for the CR portion of the trip and the non-CR tariff for the non-CR portion – much like the arrangements for international transport.

An important consideration with the Chinese network is obtaining a reliable long-term capacity allocation. Currently, there is an annual convention at which the available capacity (which is constrained on most of the east-west routes) is allocated to users, and on which MOR’s annual transportation plan is then based. If a particular shipper fails to obtain an allocation, he can try for a non-plan allocation by applying to the relevant railway authority but this is not guaranteed and is also subject to change if circumstances change.

Chinese rail tariffs have two elements, a flagfall and a rate per kilometre, with commodities classified into six groups. Most coal falls into Class 4, for which the tariff is Rmb 9.30+.0434*kilometres. A 2000-km trip would thus be charged Rmb 96.10 (about USD 14). Washed coal falls into Class 5: Rmb 10.20+.0491*distance - so a 2000 km trip would be Rmb 108 or USD 16/tonne. There are in addition two surcharges. One is a 'Railway Construction Fund surcharge' which is levied on most traffic, including coal. This is currently .033 per ntkm (Rmb 66 for 2000 km). The other is an 'electrification surcharge' of Rmb .012 per ntkm which is charged if traffic is moving over electrified lines.

On joint venture and ‘local’ lines, tariffs are generally set locally and approved by the provincial pricing authority. These are generally set on a cost-recovery basis, so may easily be two or three times those on the national network (although there is no RCF surcharge on these lines). Thus the tariffs on the Shenhua lines are set at maxima ranging from 12 fen/ntkm to 18 fen/ntkm.

In addition, there is a special tariff on the Daqin network, which was originally set to recover the capital cost of construction. In order to avoid distorting traffic flows, the same tariff was also applied to the two main routes competing with Daqin. In 2007, the average tariff on the Daqin lines was Rmb 0.12/ntkm.

The actual tariff to be paid for transport between the Mongolian border and a shipment port will thus depend on the precise route and the extent to which (on the non-MOR lines) the maximum rate is charged or a discount applied. Indicative rates, assuming CR wagons, are Rmb 161/tonne ($US 23.50) from Gauhuun Sukhait to Huanghua via Baotou and the Shenhua


DRAFT

BULLPIN

45

network, Rmb 170/tonne ($US 24.80) from Gashuun Sukhait to Qinhuangdao and Rmb 123/tonne ($US 18) from Erlian to Qinhuangdao via Jining and the Daqin line. Tariffs in private wagons would need to be negotiated but would probably attract a discount of about 15% for the round trip (including a separate charge for the haulage of the empty wagons on the return leg).

5.3 RUSSIAN NETWORK

5.3.1 Overview of the regional rail network

The Russian rail network connecting to Mongolia is relatively simple, consisting of two parallel east-west rail routes, the Trans-Siberian (TransSib) and the more northerly Baikal-Amur (BAM), together with a number of connecting lines. To the west, the lines serve Irkutsk and the Siberian mining and industrial complexes and, to the east, the ports of Vanino, opposite Sakhalin, and, further south on the Gulf of Amur, the ports of Vladivostock, Nakhodka and in the Posyet area towards the Korean/Chinese border.

The Trans-Sib runs 9,300 km from Moscow to Vladivostock, with the Bam diverging at Taishet, almost halfway along its route. It is double-track apart from the bridge over the Amur at Khabarovsk (currently being upgraded) and is electrified along its entire length. According to RZD, its capacity is about 100 million tonnes; its current traffic is not publicly available but is understood to be about 50 million tonnes. The line west of Ulan-Ude is planned to increase to 110 million tonnes p.a. by 2010 and to 125 million tonnes p.a. by 2025.

The BAM runs 4,300 km from Taishet in the west, north of Lake Baikal, to Sovgavan in the east, paralleling the TransSib around 500 km to the north through difficult terrain and with about 1,000 km of the route through permafrost. There is a line linking the two routes between Tynda and Bamovskaya. The western section of the BAM from Taishet to Taksimo (1,429 km) is electrified, with the first 704 km as far as Lena also being double-track. It is understood duplication throughout is planned for the near-medium term. The capacity of the westernmost section (Taishet – Ust Kut) is 50 million tonnes but that of the remaining single-track section is about 15 million tonnes. The capacity between Ust-Kut and Tinda is planned to increase to 30 million tonnes by 2010 and 50 million tonnes by 2025. Traffic volumes are generally low: about 8 million tonnes p.a. on the western section and 5.5 million tonnes p.a. on the east

5.3.2 Potential bottlenecks and constraints for sea-borne exports

In 2007 Russia exported about 97 million tonnes of coal p.a. out of a total annual production of 314 million tonnes p.a, two-thirds of which is opencast. Sixty per cent of production is from the Kuzbass, in western Siberia, with smaller amounts from Pechora, Donetsk, Kuznetsk, Kansk Achinsk, Neryungry and Buryatiya in the Far East and the northeast.


DRAFT

BULLPIN

46

Table 5.3 Coal production in Russia 2005 and 2006

2005 (MT) 2006 (MT) Coking coal 70 70 Steam coal 230 239 Highly volatile 96 103 Low volatility 50 52 Anthracite 9 9 Lignite 75 75 Total 300 309

At the start of 2007, Gazprom acquired an interest in the biggest producer SUEK, a strategic move focussed on cooperation in power generation. The extent to which this deal has been approved and finalized under company law is not known. Of the output, some 90 Mt or 29 % goes into exports, and 219 Mt is consumed by Russia itself. The six largest coal producers mine 55 % of Russia's coal. Exports of steam coal are mostly from opencast pits in Kemorovo with very low production costs, shipped through both Pacific and western ports. Freight rates for Russian producers in 2007 were typically about USD 25/t for hauls of about 4000 km.

Table 5.4 Major coal producers in Russia 2006

Tonnes (million) 2006

% total Russia

SUEK 89.4 Kuzbassrazrezugol 41.4 Yuzhkuzbassugol 16.1 Yakutugol 9.5 Vorkutaugol 6.8 LuTEK 5.5 Subtotal 168.7 Total 309

The six largest producers represent 55 percent of total output (Table 5.4). Total coal exports in 2006 were 89.9 Mt (14 MT of coking coal and 76 MT of steam coal and anthracite), of which 6.7 Mt were land-borne to other CIS countries and 6.4 Mt landborne to other countries, leaving 76.8 MT seaborne, up from 60 million tonnes in 2004. In the Far East, 18.3 Mt of coal was shipped, including approximately 5 Mt of coking coal. Current capacities in the Far East are: Vostochnoy – over 15 mtpa, Posyet – over 1.9 mtpa, Nakhodka 0.4 mtpa, Vanino 0.6 mtpa and 2006 shipments were thus close to capacity.


DRAFT

BULLPIN

47

Table 5.5 Sea-borne exports Russia 2004-6 (million tonnes)

2004 2005 2006 Baltic and Northern Russia 31.2 37.9 43.3 Black Sea 13.8 13.8 15.2 Far East

Vostochnoy 14.0 Vanino 0.4 Posyet 1.6 Nakhodka 0.1 Subtotal 15.2 16.5 18.3

Total 60.2 68.2 76.8 Of which

Steam coal 53 60 68 Coking coal 7 8 9

5.3.3 Russian development plans

Russia is planning to expand its capacity to export coal through its own ports from 44 Mt in 2006 to 155 Mt in 2020, with production expected to increase to around 450 Mt by 2020, of which about 80 Mt will be coking coal. The capacity of the capesize port of Vostochnoy is planned to be extended from its current 16 Mt to 25 Mt while, in the northern Sea of Japan, a new coal terminal is being constructed at Vanino by SUEK with a planned throughput of 6 million tonnes p.a in 2009, increasing to its 12 million tonnes capacity in 2010, assuming the Kuznetovsky tunnel on the branch line is replaced by one providing greater capacity. It is understood traffic to Vanino will travel on the BAM (which will also undergo some upgrades) from Siberia as far as the junction with the Vanino branch at Komsomolsk.

The Mechel group, which includes the Mechel Steel plant in central Russia, the Yuzhny Kusbass coal mines and the Korshunovsky iron-ore mine in southeastern Siberia, is set to take control of the Russian far eastern port of Posyet from MDM, a Moscow-based banking and metals group and there are plans to increase its annual capacity to 10 million tonnes. Plans have also been floated by Sibuglemet to develop a new port with a capacity of 8 million tonnes to export coking coal. 5.3.4 Handling constraints at border crossings and ports

With a common railway gauge, any delays at borders will be due to bureaucratic procedures rather than physical barriers. However, in the ports, the principal problem is likely to be gaining access to terminals, which are largely controlled by Russian coal companies or their marketing arms.

This is illustrated by an attempt by China in 2004 to export products, particularly coal, from the Tunangan region of Jilin through the adjacent Russian ports of Posyet and Zarubino where they proposed to construct terminals. The Russian government stated they ‘advocated the use of existing port facilities by Russian companies and doesn't see the need to rent out port terminals to foreign companies, when Russian companies can provide the transportation services and develop port facilities’. In addition, at the time the majority owner of the port of Posyet was Siberian Coal Energy Co (SUEK), which provided 64 percent of the throughput of the port and stated that Chinese coal would be competing for the same export markets as


DRAFT

BULLPIN

48

Russian coal produced by SUEK and that ‘It doesn't make sense to assist natural competitors9’.

They key issue in Russian ports is thus the ownership of terminals and the extent to which Mongolian coal would get access at times of either limited port capacity or when Mongolian coal was competing with Russian producers for the same markets.

5.3.5 Operational considerations

Since the creation of RZD in 2003, the Russian rail network has in theory been open to third parties and shippers have been able to operate their own trains, supply their own wagons or continue to rely on RZD. During this period, a number of third-party operators have emerged, some of whom (such as Severstaltrans) are the transport arms of major industrial organisations but others of which are genuinely independent transport specialists. Most of these operators own their own wagons but they generally continue to rely on RZD for traction, except where they have a large-volume movement over a relatively short distance. The tariff structure supporting this is unlike most other access charges in that the tariffs remain commodity-based; the charge for traction is a fixed percentage of what is being hauled, even though the cost is independent of the particular commodity being hauled.

The third-party operators are not restricted to Russia alone and can in theory operate internationally across borders but in practice only Kazakhstan has a similar tariff structure and movements are very limited at this stage. However, there is no physical reason why Mongolia could not adopt a similar system and allow third-party wagons to be hauled from Mongolian mines to Russian ports.

Russian tariffs for domestically-sourced coal are relatively low, with a strong distance taper. Most Russian mines have historically paid around $25 per tonne for their export coal to be hauled from Siberian to the Far East ports. These rates have increased significantly in 2008 although coal hauled in private wagons remains at around 2006 levels. However, the Russian tariffs are sharply different for non-Russian coal transitting the network. Table 5.6 gives the comparable rates from the two border crossing points, north of Ulaan Baatar and north of Choibalsan. It shows two rates for each movement: one is for carriage in RZhD wagons and the other is for carriage in private wagons. If private wagons are used, there is also a charge for empty wagon haulage for the return leg; when this is taken into account the current rates for transit coal are around two - three times higher than for domestically-produced export coal, even before the discounts which large-volume producers are often able to negotiate.

9 SUEK had also been a vocal opponent in the past of proposals from China to build rail access for their coal exports to the trans-Siberian rail network and thence to the larger Russian far eastern ports of Vladivostok and Vostochnoy.


DRAFT

BULLPIN

49

Table 5.6 Rail tariffs Russia (September 2008)

Origin Destination Distance (km)

Wagon owner Tariff Conversion

factor Tariff ($US)

Transit coal for export (CHF) Naushki Nakhodka-Vostochnaya 4047 RZhD 94.62 1.11 85.24 Private 80.43 1.11 72.45 Naushki Vanino 3872 RZhD 92.02 1.11 82.90 Private 78.22 1.11 70.47 Solovievsk Nakhodka-Vostochnaya 3483 RZhD 85.13 1.11 76.69 Private 72.36 1.11 65.19 Solovievsk Vanino 3308 RZhD 82.58 1.11 74.40 Private 70.19 1.11 63.23 Domestic coal for export Rouble Naushki Nakhodka-Vostochnaya 4041 RZhD 947.78 25 37.91 Private 510.30 25 20.41 Naushki Vanino 3866 RZhD 938.18 25 37.53 Private 510.12 25 20.40 Solovievsk Nakhodka-Vostochnaya 3481 RZhD 919.68 25 36.79 Private 508.92 25 20.35 Solovievsk Vanino 3306 RZhD 906.97 25 36.28 Private 503.32 25 20.13 Tariff for a block of 20 open wagons at 60 tonnes/wagon. Whilst coking coal can probably bear these tariffs at current world prices, it is doubtful whether this could be sustained in the long-run and Mongolian thermal coal using Russian routes is clearly priced out of the market.


DRAFT

BULLPIN

50

6. INFRASTRUCTURE OPTIONS

6.1 INTRODUCTION

This chapter identifies the various road and rail infrastructure options for servicing the mines and transporting production. The sector plans discussed in Chapter 3 provide an overall framework but, for rail in particular, little guidance on the short and medium-term options. This chapter presents options that have been presented by various groups within GOM to recent seminars on mining development in the South Gobi and assesses them in terms of their costs and benefits.

6.2 ROAD

The road networks that have been proposed in South Gobi are consistent with the RMP network (Table 4.2) but also include a connection between Oyu Tolgoi and Sainshand, serving Tsagaan Suvarga en route) as well as a cut-off connecting Tavan Tolgoi directly with the Dalanzadgad – Ulaanbaatar road.

Curiously, no proposals have been made for a link between Ovoot Tolgoi/Nariin Sukhait and the aimag centre at Dalanzadgad. Instead Nariin Sukhait is linked to Arvaikheer as part of one of the north-south vertical roads.

Both the roads from the mines to the Chinese border are expected to be financed by the mines, with fees then being levied for their use by third parties that are used to partially defray the cost of construction and/or maintenance. This will probably require construction of the two roads on a BOT basis, with tolls being charged on users for a specified time before the roads revert to the state.

6.3 RAIL

6.3.1 Route options

Two possible rail networks have been presented in different seminars; although they have many links in common they also have some differences.

The presentations have included the following links:

• Cross-border lines from Ovoot Tolgoi to Ceke and Tavan Tolgoi and Oyu Tolgoi to Gashuun Sukhait

• A 300 kilometre line from Ovoot Tolgoi through Sumbar to Dalanzadgad and Tavan Tolgoi

• A connection from this line to the main TMR; three alignments have been proposed: o From Tavan Tolgoi direct to Ulaanbaatar via Mandalgovi; this would provide

the most direct route to the north and Russia as well as providing a reasonable service for non-mine traffic

o From Tavan Tolgoi to the TMR in the vicinity of Airag; this is the shortest route to the TMR but involves a longer distance for traffic heading north


DRAFT

BULLPIN

51

o From Oyu Tolgoi to Zunnbayan and from there on the existing branch to Sainshand on the TMR. This involves even less construction, although the Sainshand branch would probably have to be reconstructed. A variant of this option is a direct line from Tsagaan Suvarga to the line from Oyu Tolgoi to the Chinese border

o In addition, this analysis has considered an additional route direct from Tavan Tolgoi to Sainshand which avoids the detour via Oyu Tolgoi.

Either of the last two routes would also fit with any longer-term plan for a new route to the Russian ports via Choibalsan, saving about 200 kilometres compared to the route via the existing TMR to Ulan Ude and subsequent travel on the TSR10. This new route would also serve Baruun-Urt and Tumurtin en route to Choibalsan; however, these mines, as well as Choibalsan itself, also have the option of connecting directly to Chinese branches which approach the Mongolian border at Dular and Huolinhe.

These alternatives are discussed in detail in Section 6.3.7; the sections immediately following support the analysis in that section and discuss issues which are common to all rail options.

6.3.2 Gauge

With the Chinese network being standard-gauge, any traffic to China from the existing MTZ network must either be transhipped or bogie-exchanged at some point. In practice, only the passenger traffic and a very limited amount of inwards freight traffic is bogie-exchanged by the Chinese at any of their borders with all other traffic being transhipped. There is generally only one bogie exchange facility at any given crossing but transhipment is normally carried out by the importing railway; thus most imports to China are transhipped at the Chinese border point (Erlian in the case of TMR) whilst exports are transhipped at the non-Chinese border point (Zamyn-Uud in the case of TMR).

Historically there has been comparatively little bulk mineral traffic crossing the border. In 2006 Chinese statistics show 910,000 tonnes of metallic ores and 160,000 tonnes of coal were imported at Zamyn-Uud.

The cost of transhipment depends on the volume being transhipped, the type of cargo and the facilities. For the purposes of this report, a cost of $1/tonne has been assumed for the transhipment itself but in practice, indirect costs are often incurred through reduced utilisation of rollingstock unless there the transhipment is efficiently planned and performed and there is adequate storage capacity at the transhipment point.

6.3.3 Intermodal options

Currently, intermodal transport is being used to transport production at both Tavantolgoi and Nariin Sukhait, with trucks being used to transport coal to a railhead near the border where it is transhipped for onward transport; similar arrangements exist for many small-medium mines in China itself which do not have direct rail connections.

10 However, there would be almost no saving compared to a direct line to Ulaan Bataar and then via Ulan Ude.


DRAFT

BULLPIN

52

0

2

4

6

8

10

12

14

16

1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Cost difference (c/ntkm)

Volu

me

(mtp

a)

5% 10% 15%

Range of typical cost differences

Assumptions: • a rail connection could be built for $2 million/km, while a comparable road connection would

cost $500,000 • the cost of road and rail transport (excluding infrastructure) will typically be in the range $US

0.04-0.06 per ntkm for truck and $US 0.07-0.10 for rail • a 20-year project life

Figure 6.1 Breakeven tonnages for rail line construction

Where road transport does not adversely affect the local population or the environment, this is generally the most economical method of transport for tonnages up to around 2-4 million tonnes. The precise point at which a direct connection is justified will depend on individual circumstances; Figure 6.1 shows typical breakeven tonnages for Mongolian conditions for a range of discount rates (a commercial operation would typically use 15% or greater). The minimum annual tonnage required to justify a direct rail connection, in the absence of other considerations, is in the range 3-8 million tonnes; thus a mine producing 2 million tonnes of output will rarely justify a rail connection on its own account, whilst one of 10 million tonnes or above will do so almost always.

Table 6.1 South Gobi developments and rail connections

Mine Output (mtpa) Rail connection Tavan Tolgoi 15000 Yes Nariin Sukhait 2000 Unlikely(1) Ovoot Tolgoi 5000 Probable Sumbar 5000 Probable Baruun Naraiin 6000 Probable Tsagaan Tolgoi 2000 Unlikely Oyu Tolgoi 2000 Unlikely(1) Tsagaan Suvraga 500 No

(1) But rail connection will be justified in combination with adjacent developments

Table 6.1 summarises the implications of this analysis for the proposed mining developments in South Gobi. The mine at Tavan Tolgoi in the east justifies a rail connection under any


DRAFT

BULLPIN

53

circumstances, as would the adjacent Baruun Navan. Those at Ovoot Tolgoi and the adjacent Sumbar in the west also would probably justify a rail connection; if this is constructed to the south, then it could also serve Nariin Sukhait. Oyu Tolgoi would be unlikely to justify a rail link by itself but would be served by a line from Tavan Tolgoi to the south. However, Tsagaan Tolgoi and (even more) Tsagaan Suvarga are marginal cases at best and would probably transport by road to a suitable railhead unless other factors became significant.

6.3.4 Constraints on existing network and transhipment points

There are likely to be few capacity constraints on road haulage of minerals in the South Gobi. An annual tonnage of 3 million tonnes p.a. (the likely range before rail is considered) can be carried by between 100 and 200 loaded trucks per day, depending on truck size, well within the capacity of a standard two-lane road. Special arrangements are generally made in most countries in which minerals are hauled over well-defined routes in isolated areas such as the South Gobi. Vehicles are generally truck-trailer combinations, with either two or three trailers, with each trailer carrying from 20 to as much as 35 tonnes of payload11; however, axleloads normally comply with local standards, normally 8-10 tonnes per axle, and overloading can be easily controlled with such vehicles.

Figure 6.2 Mineral haulage – 147 tonnes gross, 105 tonnes net, axleload 9 tonnes

Rail, however, has several capacity constraints. The most important is the capacity of the existing single-track TMR, which is a 23-tonne axle-load line in poor condition (Figure 6.3). Whilst the axle-load itself is not a fatal constraint (as any export movements would be

11 Figure 6.2 shows a three-trailer combination operating in South Africa over mine roads which have public access. Each trailer

carries 35 tonnes and the gross combination mass is 147 tonnes.


DRAFT

BULLPIN

54

constrained by the permissible axle-load on either the Russian or Chinese systems), the condition of the track and single line are.

The Russian network is generally built to 23 tonne axle-load and, while RZD is considering increasing this on selected routes, the upgrading of the entire trans-Siberian network will take many years. The Chinese network is generally built to a 20 tonne axle-load but since 2005 new wagons and lines are being designed for a 25-tonne axle-load. The Daqin line has already been upgraded to 25 tonnes but the Shenhua line remains at 23 tonnes. There is clearly little purpose in designing any new coal lines in the South Gobi to more than 25 tonnes, and even that is only for traffic to eastern China.

Figure 6.3 Current TMR track – 50 kg/m rail, wooden sleepers, poor ballast

The theoretical capacity of the TMR is currently around 20 pairs of trains each day. If additional crossing points are introduced, the signalling upgraded and the track improved so that running speeds could increase, the capacity could probably be increased to around 35 pairs per day. Allowing for one or two pairs of passenger trains, this is equivalent to about 25-30 million tonnes p.a. of freight. This would allow the export of possibly 10 million tonnes p.a. of minerals but anything greater than this would require duplication.

The TSR is reported to have ample spare capacity but on the Chinese network much of the national network is at or near capacity and the most promising route is likely to be the Shenhua lines from Baotou to Shenmu and on to Huanghua.

The line from Erlian to Jining is single-track and will thus need duplicating if and when the TMR is duplicated. However, from Jining onwards the South Gobi coal will be competing with local production for the limited capacity available on Daqin and its links and will inevitably be given lower priority than the domestic traffic.


DRAFT

BULLPIN

55

6.3.5 Rollingstock availability

Rollingstock is unlikely to prove a problem as long as MTZ permits non-MTZ/CR/RZD wagons to operate over its network, as is envisaged under the Millennium Challenge grant. Exports to Russia can be handled through operators such as Globaltrans who specialise in providing wagons for private customers and, for major movements which can support dedicated motive power, locomotives.

Exports to China will best be handled in Chinese wagons; although no standard-gauge wagons Chinese wagons currently leave China, metre-gauge wagons work through between Kunming and Vietnam and there seems no reason why they could not do so for the South Gobi mines. The only constraint privately-owned wagons would face is the need for an agreed procedure to minimise customs-related delays at the border. Currently there are no wagon leasing companies in China but there is an ample number of wagon builders (ten manufacturers produced the standard C70 and C8012 coal wagons in 2007). Similarly a standard DF diesel locomotive can be obtained from four manufacturers.

6.3.6 Estimated rail capital and operating costs

The cost of constructing a new railway varies significantly, depending on the terrain (which affects the quantity of earthworks and the number of bridges and tunnels and also the standard of the track being constructed. Construction costs in China, the closest parallel to the South Gobi, vary from a minimum of under $1 million per route-km for a single-track minimum-standard railway in easy terrain to over $10 million per route-km for a double-track medium-high speed line in mountainous terrain.

A line capable of carrying 25 million tonnes of minerals a year in moderate terrain such as the South Gobi, would normally be constructed as a Class II railway in China, with a ruling grade of 0.6 %. The cost of such lines is estimated at $1.3-2.0 million per route-km, depending on the terrain and volume of earthworks. Table 6.2 gives an indicative breakdown for light to moderate terrain.

12 The number refers to the net tonnes carried by each wagon; the C80 currently only runs on the Daqin line.


DRAFT

BULLPIN

56

Table 6.2 Indicative construction cost ($US 000 per route-km) (mid-2008 prices)

Item Cost % Site preparation 100 8% Earthworks, grading etc 370 29% Culverts, drains, bridges, fencing 120 9% Trackwork

Rails 156 13% Sleepers and fastenings 125 10% Ballast 87 7% Installation 30 2% Subtotal 399 31%

Loops, signalling, communications 171 15% Other (buildings, temporary works etc) 59 5% Stations and yards 54 4% Total 1295 100% Contingency 259 20% Engineering, project management 129 10% Total 1683 130%

For the purposes of route comparison and evaluation a unit construction cost of $1.8 million/km has been used.

Operating costs likewise depend on the standard of the line (in particular the axle-load) and the terrain (which affects the tonnes per train and per locomotive). The two most modern and efficient mineral lines in China are the Daqin and Shenhua heavy-haul lines. In 2007, the average operating costs were almost identical at just under 0.008 $US/ntkm (Table 6.3). The table also includes the average cost on the CR network for all freight operations, estimated at $US0.0093/ntkm, excluding depreciation. For the purposes of the following analysis, a figure of $US0.01/ntkm, including depreciation, has been used as representative of a range between $0.009-0.012/ntkm.

Table 6.3 Indicative operating costs 2007 ($US c/ntkm)

Item Daqin Shenhua CR (dsl/elec)

Labour 0.12 0.12 0.21 Materials 0.07 0.10 Fuel and power 0.13

0.18 0.15 Repairs and maintenance 0.13 0.12 Other 0.17 0.13

0.33

Subtotal 0.68 0.59 0.79 Depreciation 0.10 0.17 n.a. Total 0.78 0.76 n.a.

Source: Shenhua and Daqin annual reports, IBRD estimate

6.3.7 Analysis of route options

Cross-border lines to Tavan Tolgoi and Ovoot Tolgoi

The comparison of road and rail transport in Section 6.3.3has shown that the line linking Tavan Tolgoi to Gashuun Sukhait should be constructed as soon as the coal mines production increases to any marked extent. Similarly the line from Ovoot Tolgoi south is also likely to be


DRAFT

BULLPIN

57

justified as soon as production reaches 4-5 million tonnes p.a. The analysis therefore assumes that both these lines are constructed as Chinese-gauge lines. The same reasoning would apply to other proposed new mines near the Chinese border as they developed.

Ovoot Tolgoi – Tavan Tolgoi

The case for a connection between these two lines via Dalanzadgad is less clear. Mining inputs alone would not require a connection from the north to Ovoot Tolgoi. Its justification therefore depends on whether the Ovoot Tolgoi mines wish to have an alternative outlet for their coal, either via Gashuun Sukhait and Baotou or via Russian ports.

Table 6.4 Comparison of routes between Ovoot Tolgoi and Baotou

From To Distance Cost/ntkm (Rmb) Cost/tonne (Rmb) Via Tavan Tolgoi Ovoot Tolgoi Tavan Tolgoi 380 0.10 38.00 Tavan Tolgoi Wuyuan 378 0.10 37.80 Wuyuan Baotou Xi 177 0.13 22.82 Total 935 98.62 Via Ceke Ovoot Tolgoi Ceke 20 0.10 2.00 Ceke Linhe 550 0.15 84.15 Linhe Baotou Xi 234 0.10 23.10 Total 804 109.25

Table 6.4 compares the distances and operating costs between Ovoot Tolgoi and Baotou for a route via Tavan Tolgoi and the route which will shortly be available using the new line from Ceke to Linhe. The route via Tavan Tolgoi is longer by about 130 kilometres but is more expensive by about Rmb 10/tonne, excluding the cost of constructing the line. At a discount rate of 10%, this saving requires a volume of around 50 million tonnes before it becomes worthwhile and the only justification for the line would then be the option of using the Russian ports. If this is the case, using a discount rate of 10%, the line would need to generate a saving of about Rmb 50/tonne (US$7/tonne) in the net return to the mine owners to be worth constructing. The possibilities of this are discussed below in conjunction with the Tavan Tolgoi traffic.

Connections between Tavan Tolgoi and TMR

Four routes (Options 1-4 in Table 6.5) have been identified for this connection. The principal purpose in constructing one or more of these routes is to provide an alternative export route to Russia or China or both and the table gives the distance by each option to Vostochnoy (for exports via a Russian port), Qinhuangdao and Huanghua13 (assuming capacity is available on the rail network for exports via a Chinese port).

For southbound traffic, the routes via the TMR are at least 300 km further than by using the Gashuun Sukhait crossing point and the Chinese network. In addition, all these routes require a bogie exchange at the border (as well as possibly increasing the capacity of the TMR by duplication) and also need to negotiate the heavily congested network around Datong to gain

13 Although Huanghua is not one of the three Chinese ports open to Mongolia under the current transit agreement, it is assumed

this port could be readily added if required.


DRAFT

BULLPIN

58

access to either the Daqin railway or, further south at Shouzhou, the line to Huanghua. It is therefore unlikely that any significant exports would use a link via the TMR in preference to the Gashuun Sukhait line and a major consideration determining access to the TMR should thus be its usefulness for exports via Russia.

Table 6.5 Alternative rail routes (distance to ports)

Route Base 1 2 3 4

Route option Gashuun Sukhait Sainshand Sainshand

direct Airag Ulaanbaatar

Northbound traffic Tavan Tolgoi – TMR 507 400 400 530 TMR – Russia border (Naushki) 876 876 752 402 Russia border – Vostochnoy 4047 4047 4047 4047 Total 5430 5323 5199 4979

Southbound traffic

Border crossing point Gashuun Suhait Erlian Erlian Erlian Erlian

Tavan Tolgoi – TMR 507 400 400 530 Tavan Tolgoi – Chinese border 192 TMR – Chinese border 237 237 361 711 Chinese border – Baotou 367 Baotou – Huanghua 1003 Chinese border – Datong 634 461 461 461 461 Datong - Qinhuangdao 653 653 653 653 653 Datong – Shouzhou 129 129 129 129 Shouzhou – Qinhuangdao 639 639 639 639 Total (Qinhuangdao) 1479 1858 1751 1875 2355 Total (Huanghua) 1562 1973 1866 1990 2468

For northbound traffics, the direct line to Ulaan Bataar is clearly the shortest, being 200-400 kilometres shorter than the other options; this needs to be balanced against the increased cost of construction compared to the other routes.

The cost of transport by the alternative routes has been calculated as the sum of operating cost (excluding infrastructure capital costs) on the new South Gobi lines and rail tariffs on the other networks, using the tariffs documented in Sections Error! Reference source not found., 5.2.5 and 5.3.5.


DRAFT

BULLPIN

59

Table 6.6 Alternative rail routes (operating cost) ($US/tonne)

Route Base 1 2 3 4

Gashuun Sukhait Sainshand SS direct Airag Ulaanbaatar

Northbound traffic Mongolia 13.83 12.76 11.52 9.32 Russia 85.24 85.24 85.24 85.24 Total 99.07 98.00 96.76 94.56

Southbound traffic Mongolia 1.92 7.44 6.37 7.61 12.41 Transhipment 1.00 1.00 1.00 1.00 China (Qinhuangdao) 24.76 17.94 17.94 17.94 17.94 China (Huanghua) 28.57 19.13 19.13 19.13 19.13 Total (Qinhuangdao) 26.68 26.38 25.31 26.55 31.35 Total (Huanghua) 30.49 27.57 26.50 27.74 32.54

For northbound traffic, the cost of transport is dominated by the Russian tariff of $85/tonne (or possibly a bit less if private wagons are used). This route is therefore only viable as long as the fob price of coking coal at Vostochnoy remains high enough to cover the land transport cost of around $100/tonne.

Southbound traffic has slightly lower operating costs on the options via the TMR, as the Chinese routes include a significant component representing infrastructure capital costs. Overall, the land transport cost is about $70/tonne cheaper than via the Russian ports and exporting via a Chinese port would therefore generally always be the first choice.

These differentials in operating cost need to be balanced against the corresponding differentials in capital and infrastructure maintenance cost (Table 6.7). As any significant additional tonnage on the TMR will require additional capacity, an allowance of $0.5 million has been included in the analysis once export volumes exceed 5 million tonnes p.a.14

14 In practice, if demand on TMR is increasing quickly, the export coal will merely bring forward what would otherwise be spent a few years later. However, a significant export volume will require duplication of the existing track at a cost rather larger than that allowed in the table.


DRAFT

BULLPIN

60

Table 6.7 Alternative rail routes (infrastructure capital and maintenance cost) ($US)

Route Base 1 2 3 4

Gashuun Sukhait Sainshand SS direct Airag Ulaanbaatar

New construction (km) 192 343 400 400 530 Cost/km ($US mill) 1.8 1.8 1.8 1.8 Reconstruction (km) 50 Cost/km ($US mill) 0.5 Additional capacity on TMR(1)

Northbound (km) 876 876 752 402 Southbound (km) 237 237 361 711 Cost/km 0.5 0.5 0.5 0.5

Capex ($US mill) (excl new capacity) Total 642 720 720 954 Depreciation p.a. 26 29 29 38 ROI p.a. 45 51 51 67

Maintenance cost ($US) $4000/route-km p.a. + $0.002/ntkm (1) Assumed to be required once export volumes reach 5 million tonnes.

The combined capital and operating costs of the various options for a range of traffic volumes going either north or south are shown in Figures 6.1 and 6.2. The costs for the southbound traffics are to Datong (the common point for the traffic to Qinhuangdao) while for the northern traffics they are to the Russian border at Naushki.

The route via Gashuun Sukhait is the cheapest southern export route by around $10/tonne, for an export volume of 10 million tonnes, reducing to $4/tonne if exports reach 20 million tonnes p.a. By 30 million tonnes p.a. (probably the maximum for this type of single-track line), the gap is down to $2.50/tonne. This analysis assumes that the Gashuun Sukhait line is already constructed on the Mongolian side and there is thus no capital cost associated with using it. The two Sainshand options (Options 1 and 2) have lower costs than the Airag route (Option 3) as they join the TMR further south, while the Ulaan Bataar line (Option 4) unsurprisingly has a high cost penalty for exports to China.

However, Ulaan Bataar is consistently the cheapest option for northbound exports, as its higher construction cost is outweighed by the need for less upgrading of the TMR and the much shorter distance with its lower operating costs.


DRAFT

BULLPIN

61

0

5

10

15

20

25

30

35

40

45

50

5 7 9 11 13 15 17 19 21 23 25 27 29

Traffic volume (tonnes million p.a.)

Cos

t/ton

ne ($

US)

Base Option 1 Option 2 Option 3 Option 4

Figure 6.1 Total cost – southern exports to Datong (US$/tonne)

0

5

10

15

20

25

30

35

40

45

5 7 9 11 13 15 17 19 21 23 25 27 29


Cos

t/ton

ne ($

US)

Option 1 Option 2 Option 3 Option 4

Figure 6.2 Total cost – northern exports to Naushki (US$/tonne)

The optimum choice of route to access the TMR thus depends on the direction in which the exports are expected to flow; predominantly northbound flows will favour the more northerly options whilst the converse is true for southbound exports. However, for these latter, the Gashuun Sukhait route is so superior that it is unlikely a significant volume would use the TMR unless there were exceptional circumstances.


DRAFT

BULLPIN

62

East-west route via Choibalsan

A final option is a cross-country route, which extends the Airag route to use the existing 60-kilometre Borondor branch and then build a new line to Choibalsan. From there, traffic can use the existing line (which would need to be rebuilt) to access the Russian network at Solovievsk and subsequently join the Trans-Siberian Railway. Unfortunately, although this route saves some distance compared to using the TMR to Ulan Ude, the saving is not as large as might be expected as the route from Solovievsk to the TSR runs generally north-west (Table 6.8)

Table 6.8 Alternative rail routes for northbound traffic

Route 3 Airag

4 Ulaanbaatar

5 Choibalsan

Distance to port (km) Tavan Tolgoi – Russian border 1152 932 1465 Russia border – Vostochnoy 4047 4047 3483 Total 5199 4979 4948

Operating cost ($US) Mongolia 11.52 9.32 14.65 Russia 85.24 85.24 76.69 Total 96.76 94.56 91.34

New construction (km) 400 530 1166 Reconstruction (km) 299 Capex ($US mill) (excl new capacity) 720 954 2248

The combined capital and operating cost of the Choibalsan option is compared with the Airag and Ulaanbaatar options for a range of traffic volumes in Figures 6.3. The costs are to the respective Russian borders, with those for the Naushki options increased by the difference between the tariffs from Naushki and Solievski to Vostochnoy ($8.55/tonne).

0

10

20

30

40

50

60

70

80

5 7 9 11 13 15 17 19 21 23 25 27 29


Cos

t/ton

ne ($

US)

Option 3 Option 4 Option 5

Figure 6.3 Total cost – northern exports via Naushki/Solievski (US$/tonne)


DRAFT

BULLPIN

63

The Choibalsan route (Option 5) is significantly more expensive until the export volume reaches around 15 million tonnes p.a. and only becomes slightly cheaper than the Airag route (Option 3) when volumes reach 30 million tonnes p.a. However, the Ulaanbaatar route (Option 4) remains cheaper even at this level of traffic as the much larger capital cost on the Choibalsan route, including the rebuilding of the existing branchlines which are used, only generates very small ($3-5/tonne) reductions in operating cost.

6.3.8 Identification of preferred routes and timing

The analysis in the previous section shows that, whilst the cross-border routes to China are likely to be viable at around 5 million tonnes p.a., extending the network to connect to the TMR needs to wait until the export potential for Mongolian coal is clearer. It seems unlikely that any significant volume of exports to China will use the TMR via Erlian unless the line is built for other reasons but it is difficult to see what other traffics might justify construction of a connecting line, as most other developments in the Tavan Tolgoi – TMR corridors are metal mines generating comparatively low volumes of product which would normally be transported by road to an existing railhead.

Whichever route is constructed, it will be better for some traffics than others. If exports via Russia can be guaranteed and it is agreed all exports to China will use Gauhuun Sukhait, then the Ulaan Bataar route is the best but if it is required to keep options open then an alternative approach may be needed in which the route which gives the least penalty for a mix of traffics. Figure 6.4 shows the cost of each route option (including the upgrading of the TMR) for a 50:50 split of traffic going north and south.

0

5

10

15

20

25

30

35

10 14 18 22 26 30 34 38 42 46 50 54 58


Cos

t/ton

ne ($

US)

Option 1 Option 2 Option 3 Option 4

Note: Options 2 and 3 are superimposed in this graph

Figure 6.4 Total cost of options – 50:50 north-south split of traffic (US$/tonne)


DRAFT

BULLPIN

64

In this situation, Options 2 and 3 (which are indistinguishable on the graph) are better than either of the extremes (Option 1 and Option 4). However, Option 3 (Airag) is more robust as, whilst it is never the best option for either direction, it is also within about $2/tonne for both directions, a small penalty compared to the total cost of transport to the ports. It also provides the opportunity for future extension to Choibalsan should traffic volumes expand or if the network structure develops to provide a more direct link to the TSR.

However, practical route selection will also need to take into account any engineering features of these lines. The analysis has assumed the same unit construction costs for all routes and the same operating conditions and unit costs; in practice these will need to be reviewed and refined prior to any final decisions on route selection.

The timing of any new construction remains unclear. The distance to ports of Mongolian exports leave them very vulnerable to changes in world prices or to unilateral changes in rail tariffs in Russia and China. Long-term contracts would be required before any substantial new construction would be justified, or attractive to private investment, covering not only the end-user but also port and rail access in China or Russia.

The cross-border lines are being planned as standard-gauge lines. However, it is likely that most of the traffic using the connections to TMR will be to or from the north and these lines should therefore be broad-gauge. If the Airag route is selected, direct from Tavan Tolgoi, Oyu Tolgoi will not have a broad-gauge connection but it seems unlikely much concentrate traffic would move north and any that does could easily be transported by road to a railhead at Tavan Tolgoi; an option if traffic volumes increase is to construct a broad-gauge link between the two.


DRAFT

BULLPIN

65

7. PRIVATE INVESTMENT IN MINERAL TRANSPORT

7.1 INTRODUCTION

The development of Mongolia’s strategic mines, including Oyu Tolgoi and Tavan Tolgoi, requires electricity plants and railways costing well over $1 billion. As the Government can finance at most about $100 million of infrastructure (electricity, transport, water services, housing) per year in South Gobi using its own revenues, it well be necessary to mobilise private sector investment if these projects are to go ahead. This chapter summarises potential options for private sector investment in infrastructure and highlights key issues relevant to the South Gobi.

7.2 OPTIONS

There are many different ways in which the private sector can be involved in the construction and operation of railway lines (Table 7.1).

Table 7.1 Main forms of private-sector involvement in new railway projects

Finance and build rail line

Operate and maintain rail line

Finance and maintain trains

Operate train services

Train availability contract

Public Public Private Public or private (hire payments to

private) Train operating concession

Public Public Private Private (pay access charges to public)

Infrastructure concession Private Public or private (lease payments to

private)

Public or private Public or private

Infrastructure build and operate concession

Private Private Public or private Public or private (pay access charges

to private) Integrated concession Private Private Private Private

The private sector can be involved in any, or all, of infrastructure construction, infrastructure operation (i.e. signalling and train control), rollingstock provision and train operation through five basic types of concession: • Train availability contract – examples of these contracts, where rolling-stock is

provided by a third party (who also normally also does the scheduled maintenance) include the Roscos (both passenger and freight rollingstock companies) and the ‘power-by-the-hour’ contracts which have been negotiated with some major manufacturers. These enable what would otherwise be large lumps of capital expenditure to be spread over the life of the asset, as well as guaranteeing rollingstock availability and reliability.

• Train operating concession – examples of these are the various European passenger rail concessions in which operators either pay Government (or are paid by Government) to operate specific services. Such operators pay access charges (normally a combination of a rate per train-kilometre and a rate per gross tonne-kilometre) to the infrastructure provider. In some cases, and this is the normal situation for freight, there is ‘open access’,


DRAFT

BULLPIN

66

i.e. an operator can operate trains without any financial payment to or from Government by merely complying with technical standards and paying access charges

• Infrastructure concession – where an organisation undertakes to build infrastructure which is then leased to Government (either directly or to a Government-owned railway) for an annual lease fee. This arrangement is similar financially to leasing rollingstock in that it avoids having to make a large initial capital expenditure and it also greatly reduces Government’s risk of project over-runs and delays. However, it is normally Government’s responsibility to provide the right-of-way and to supervise resettlement. In the early days of railways many railways were built in this way and then contracted to a neighbouring (and generally larger) railway to operate, normally in return for either a fixed fee or for a share of revenue, and it is possible some of the new Chinese dedicated passenger lines may be financed in this way.

• Infrastructure build and operate concession – where an organisation both builds the infrastructure and ‘operates’ it, similar to the arrangements used for toll-roads. There are comparatively few examples of such arrangements, although there is currently a proposal for such a scheme for a new mining development in Australia.

• Integrated concession – this is the classic form of concession which flourished in the 19th century. A company is given the right to build and operate a railway, normally for a defined time period (historically 50 or 100 years) subject to certain conditions such as tariff control, service obligations or (on the Government side) a dividend guarantee. Recent examples include the Alice Springs – Darwin line in Australia (although this is subject to open-access provisions) and the Saudi landbridge project.

There are a number of variations on these basic models; for example, a line can be financed by a mine developer, even if it is built and operated by the public sector, with the cost of the line being offset against freight revenue received from the mine.

One of the fundamental distinctions is that between railways built by an individual company (or companies) which they intend to use to transport their own products as part of an overall logistic chain and those which third parties can use – either using and operating their own rollingstock or using services provided by railway operator. Many of the new railways constructed in the last fifty years have been of the first type: examples include the Carajas ore line in Brazil and the Mauritania ore line in West Africa. Both were built by mining companies to transport their ore as part of an integrated mine-rail-port project (and both incidentally with World Bank assistance). They are both effectively private railways with no obligations to provide a service for third parties, although some informal passenger transport is done by both of them.

However, in other cases, the railways have either been explicitly built to carry third-party freight or it is an obligation imposed by the Government as part of the permission to construct the railway. In these cases, which include a number of examples in Australia, as well as a coal line in Colombia, governments have been concerned to ensure that smaller mines that cannot justify building a railway line on their own account can be provided with access at a reasonable price.


DRAFT

BULLPIN

67

7.3 RAIL NETWORKS AND ACCESS

A fundamental issue is whether, if a line is built by a particular mine, third-parties will be allowed to transport their product on it for a reasonable charge (i.e. cost plus reasonable profit). Many of the heavy-haul lines currently operated are solely used by the company that built them; others, such as the Pilbara lines in Australia, have erected significant legal and technical barriers to prevent third parties taking advantage of what appeared at the time to be a simple requirement; still others, as in the case of Gabon in Africa, appear to severely discriminate against third-party users15. The first decision to be made is therefore whether the line is to be common-user and, if so, what arrangements are proposed for this to be enforced in practice as well as in theory.

There are three broad streams of work associated with administering a rail network with multiple users:

• Tasks associated with accreditation, licensing and safety • Tasks associated with establishing the procedures for capacity allocation and network

management • Tasks associated with establishing the charging framework and prices

Each of these streams of work has two phases: • Defining the principles to be adopted • Developing the procedures by which they will be implemented

These are normally summarised for by third party operators (TPOs) in a document which in Europe is known as the ‘network statement’.

7.3.1 Accreditation, Licensing and Safety

This covers the following areas: • Accreditation of the operator, to ensure he has the necessary management capacity

and financial resources • Certification of the rollingstock, to ensure it complies with GOM standards • Licensing of the operating staff (drivers etc) to ensure they are suitably qualified • Ensuring that the operator will meet the necessary safety standards

All these functions would normally be seen as the responsibility of MOR (as the regulatory authority). Safety in particular requires considerable thought. The trend in many countries is for this to be the responsibility of the operator through the preparation and approval of a Safety Plan, for which the senior managers are ultimately personally responsible. It is important these regulatory functions be separated from the main operating railway, both from the viewpoint of corporate governance but also because there will almost inevitably be differences of opinion between it and TPOs, for example in the introduction of newer and more innovative rollingstock. The full benefits of third-party involvement will only be achieved if such differences are resolved independently.

7.3.2 Capacity Allocation and Network Management

These tasks are concerned with:

15 The access charge for third-party users in 2005 was ten times that for the operator, and the operator’s traffic reportedly had

priority over other users.


DRAFT

BULLPIN

68

• Establishing the principles by which available capacity is distributed between the infrastructure owner; this task is addressing medium- and long-term procedures (e.g. grandfather rights, impact of re-timetabling etc)

• The rules governing day-to-day network operation (e.g. which trains have priority over others under what circumstances, procedures in case of accident or disruption etc)

These are especially important where train control is in the hands of the major operator as there is often a natural tendency to favour one’s own trains when problems arise. Experience has shown that it is important TPOs can be confident that they will get fair treatment in such situations and a proven way of doing this is to set down clear train control guidelines.

7.3.3 Charging structures

These tasks are concerned with establishing the charging principles, structure and prices. There are three key issues:

• The extent to which revenue from TPOs is to be maximised as compared to volume • The structure of charges (e.g. different charges for different types of services, basis on

which charges are made – train-km or gross tonne-km, say – or whether they are commodity-based, as in Russia - and the conditions under which charges are levied e.g. take-or-pay, rebates for failures by the infrastructure operator to provide the contracted level of service etc)

• The absolute level of charges The first of these, at least, will require discussion with GOM while the second and third will involve discussions with potential TPOs as well as the marketing and costing staff of the infrastructure operator. Hopefully, the work undertaken to support the Millennium Challenge project will provide a sound basis for this.

7.4 ENCOURAGEMENT OF TPOS

Even though the regulatory and operational obstacles for TPOs may be overcome, they may still face difficulties in obtaining rollingstock. Most rail rollingstock has a life of over 20 years, a long period to which to commit given that comparable road vehicles may only be used for 3-5 years and then easily disposed of through the secondhand market.

A key factor where TPOs have flourished has therefore been the existence of third-party rollingstock which can be leased or hired for periods of one or two years. In some countries this has been through specialist leasing companies; in others (particularly Australia) through other railways. An obvious option in theory in Mongolia is for TPOs to contract with one of the Russian railway operators for the provision of rollingstock.

Although there are currently no rail leasing organisations in Mongolia, two alternatives that would significantly increase the participation of TPOs for at least the medium-term are:

• Establishment of a public leasing company using new rollingstock. This could also lease to PR instead of the current practice of effectively providing free rollingstock which is in practice financed by MOF

• Establish a facility whereby any TPO who uses ‘approved’ rollingstock has the option, after a set period of time (say 3 -5 years) of having it purchased by a public


DRAFT

BULLPIN

69

rollingstock company which can then lease it to either another TPO or to PR (or even to India/Sri Lanka/Bangladesh).

Once a market has been established, this company could be either sold or liquidated, taking care to avoid the less desirable aspects of the UK ROSCOs.

An important part of the development of track access policy will therefore be to research the market for potential TPOs and establish to what extent and under what conditions they are prepared to fund rollingstock themselves.

7.5 OTHER ISSUES

Other issues to be considered include whether any line that is constructed by a private company will eventually revert to the state (either at the end of the mining lease or after some more general time period) or whether ownership of the line will continue in perpetuity. This is often linked to who is developing the line: if it is a mining company then the life would presumably be linked to the mining lease but if it is an independent body or consortium then the life would generally be related to the presumed life of the infrastructure (50 years in the case of one Australian line).

If the line is to be handed back, then the handback arrangements need to be clearly stated, as well as the conditions under which the government can step in and takeover if the concessionaire is not performing as required (as happened in Colombia). To do this successfully, governments must have lawyers at least as good as those acting for the concessionaire and must also have a monitoring procedure with is non-intrusive but observant.

The process of tendering, concessionaire selection and negotiation of railway concessions are similar in those in other sectors. Care must be taken to ensure that the concessionaire is financially and technically capable of performing his obligations; this has often been a problem in railway concessions, especially where the concessionaire is not linked to a user of the railway. This is especially so if the Government is contributing a significant share of any capital expenditure as there is often pressure to ‘get the project moving’ while the concessionaire organises his finances (as happened in Colombia).

Once the line is in operation, interface arrangements with other operators will need to be clear and straightforward. In the case of the Gobi mines, will they be able to run their own trains over the existing Mongolian network (and then onwards over the Russian network?). Will the existing Mongolian network be upgraded to accept 25 tonne axle-load wagons (probably the minimum that any new line should be constructed to) – and if so, when? What will the charges be for on-carriage and so on.

Developing a concession is often a lengthy process. Few have been arranged in under two years from project concept to agreement; many have taken far longer. In doing so, almost all governments need advice from experts experienced in this type of transaction; concessionaires, many of whom will themselves be experienced, will be obtaining the best advice they can, both locally and internationally. IFIs have often played a useful role; several of the major world mineral railways were partly funded by IFIs and their corporate experience can help to ensure any process is both transparent and in the best interests of the country as a


DRAFT

BULLPIN

70

whole. And finally, organising and completing a transaction is not cheap; few will have been done without at least a million dollars having been spent on transaction advisors of one sort or another.

7.6 POTENTIAL FOR PSP IN SOUTH GOBI

The potential for PSP in the rail network in South Gobi ranges from very good to negligible. The most important thing for an investor in infrastructure is the reliability of the traffic forecasts. The lines linking the proposed mines with the industrial complexes just over the border in China will be carrying traffic that is subject to long-term contracts. Whilst the recent fluctuations in commodity prices have shown that such contracts cannot be relied on absolutely, they are the closest thing to certainty that the mines are going to experience, especially if the industrial complexes themselves have a stake in the mines. For many of the plants in Gansu and Nei Monggol, Mongolia is the closest and cheapest source of their inputs and this also provides some reassurance that substantial traffic volumes can be expected in the long-term.

The lines linking the mines to the existing Mongolian network unfortunately are unlikely to have any such long-term guarantees. Exporting coal through Russian orts must overcome two uncertainties: firstly, the inherent uncertainty of the international market in which, for practical purposes, Mongolian coal will be competing against Chinese, Australian and Russian coal for the Japanese and South Korean markets; secondly, the uncertainty of exporting through Russia, which has shown several times in recent years that it is prepared to use energy as a political weapon.

Whilst it may be possible to interest some local businesses in lines connecting the mines to the existing network, it seems likely that any such interest would require some guarantees from the government. The other option is to interest some of the Russian resource companies in the South Gobi mines.

It is important that any lines that are built allow access to third parties at a reasonable rate. Whilst Tavan Tolgoi is clearly large enough to justify a line in its own right, other mines with a production capacity of, say, 2-5 million tonnes p.a. will not be and will naturally use road. At the margin, this will mean that there will be small mines which will not be developed unless they can obtain rail access and it is thus in Mongolia’s national interest to ensure this is so.


DRAFT

BULLPIN

71

8. TRANSPORT INFRASTRUCTURE DECISION FRAMEWORK

This chapter presents an outline decision framework (Table 8.1) for developing transport infrastructure within the South Gobi, based on the analysis and discussion in the previous sections.

Table 8.1 Outline decision framework

Issue Decision Comments

Road access to new mines

Privately-funded as far as public network

Determine threshold volumes above which access roads must be sealed to preserve environment. Specify construction standards

Use of public roads for hauling product

Allow up to specified volume subject to loading standards and charges

Volume limit will depend on level of other usage. Loading should be enforced at corporate level. Charges must be sufficient to cover long-run maintenance and renewal costs

Rail access to new lines Allow

Require access to third parties under specified conditions. Develop standard procedures and charging methodology. Specify handback requirements.

Connection to TMR

Allow if privately-financed. Otherwise defer until there is clear evidence of long-term demand

Undertake marketing and operations study to establish long-term potential for export to Russia. May require commercial agreement with RZhD and ports to obtain reduced long-term access and rate for export coal transitting from Mongolia. As China appears to currently have $70/tonne advantage, a better option may be to develop alternative customers in China; the ownership of the connecting border lines in China should be monitored.

Connection between Ovoot Tolgoi and Tavan Tolgoi

Allow if privately-funded.

Any Government funding should be restricted to the Tavan Tolgoi – Dalanzadgad section and only considered after a connection to TMR has been constructed.

Connection to Choibalsan Long-term option only Monitor progress of proposed Chinese

line through Choibalsan

Initially, most new mines will use road to transport their product; this will involve privately-funded mine access roads with subsequent use of the public network. The private roads should be sealed once volume reaches a minimum value to minimise environmental damage whilst use of the public road network should be subject to specified loading standards and road maintenance charges. These requirements should be enforced on a regular basis to avoid long-term damage to the pavement structure.

Rail line connecting the mines, either cross-border or to the main network, should be privately-funded but should be required to provide access to third parties who may desire it


DRAFT

BULLPIN

72

under standard procedures and charges developed by the regulatory authority. Any agreement for line construction should specify handback arrangements and the conditions under which the government can step in if the operation of the line is prejudicing the interests of other users.

If a private investor is prepared to finance a connection to the TMR, it should be allowed subject to the same conditions as for mine connections. Otherwise this connection should be deferred until there is clear evidence, e.g. through a marketing and operational study, that there is sufficient demand for the long-term export of coal through Russia (e.g. if a major Russian producer invested in the Mongolian mines). Any investment by the Government in the connecting line should be kept to a minimum (e.g. it might reflect the share of general traffic) and arrangements should include charges for any upgrading to increase the capacity of the TMR.

The link between Ovoot Tolgoi and Tavan Tolgoi appears to have little traffic potential and should only be considered after a TMR connection has been established. Any Government contribution to funding should be limited to the Tavan Tolgoi – Dalanzadgad section only.

The route via Choibalsan appears a long-term option only and is also subject to the same general caveats as the TMR connections.

Land Transport Report - World Bank Group

Documents

Transcript of Land Transport Report - World Bank Group