Revamping the Regional Railway Systems
in Eastern and Southern Africa
Mark Pearson and Bo Giersing
Regional Integration Research Network
Open Dialogues for Regional Innovation
www.
trademarksa.org/rirn
Regional Integration Research Network
Discussion Paper (RIRN/DP/12/01)
Published by TradeMark Southern Africa | September 2012
About the Regional Integration Research Network
TMSA supports the COMESA-EAC-SADC Tripartite to generate innovative approaches, practices and solutions to
vexing regional integration questions e.g. the programmatic approach to regional transport network development;
integrated management approaches to the functioning of border crossing problems; an electronic approach to
non-tariff barrier identification and reporting for real-time solution; introducing new and revamped methodologies
and approaches to measuring progress and success, particularly along transport corridors. The RIRN will
provide a platform for exchanging information and facilitating dialogue on these and many other pertinent regional
integration issues and practices.
The RIRN dialogue platorm is:
•
•
•
•
Inclusive - broadening the traditional circles for dialogue from policy makers and their think tanks to also
include stakeholders such as private sector service providers, financiers and users / beneficiaries of regional
integration activities.
Focused on topics such as trade, trade and transport facilitation, regional infrastructural development,
industrialisation within the region and linkages among these topics. In these areas the emphasis will be on
the ‘what’ and ‘how to’ of regional integration skewing dialogue to approaches, best practices, innovations,
evidence, experiences, lessons learnt, impacts, risk mitigation and removal of constraints to regional
integration. Even where traditional areas related to the political economy, scope and depth of regional
integration are debated, the focus will be on practical aspects of the ‘what’ and ‘how’ of integration.
Focused on regional integration particularly in the Tripartite Region, but also beyond.
Complementary to existing dialogue forums and initiatives which have different purposes and approaches to
the same functional areas.
The RIRN will facilitate information exchange and dialogue through:
• Generation of information and knowledge,
• Using peer-reviewed commissioned research, own research, and analytical pieces by academia, private
sector and other potential contributors; and
• Evidence and lessons learnt from pilot / demonstration projects / initiatives;
• Sharing of such information and knowledge through publication in TMSA Trade and Regional Integration
Discussion Notes, which will be downloadable from the TMSA Website; TMSA Trade and Regional Integration
Network Symposiums; and participation in forums in this field that are initiated by other stakeholders
For more information and enquiries about the RIRN, please contact Lolette Kritzinger-van Niekerk
at TradeMark Southern Africa
Phone: +27 12 349 7500
Email: [email protected] | [email protected]
Web: www.trademarksa.org/rirn
Address: First Floor Building 41, CSIR Campus, Brummeria, Pretoria, 0001, South Africa
Contents
Acronyms and abbreviations......................................................................................................................... 2
1Background................................................................................................................................ 5
2
Description of the region’s railway systems............................................................................... 7
2.1Malawi........................................................................................................................................ 7
2.2
Democratic Republic of the Congo............................................................................................ 9
2.3Zimbabwe................................................................................................................................ 10
2.4Zambia..................................................................................................................................... 11
2.5Mozambique............................................................................................................................ 14
2.5.1Beira........................................................................................................................................ 14
2.5.2Nacala...................................................................................................................................... 16
2.6Tanzania ................................................................................................................................. 16
2.7Botswana................................................................................................................................. 18
3
Revamping the railways in ESA............................................................................................... 19
4
Conclusion and the possible way forward............................................................................... 22
Bibliography................................................................................................................................................. 27
Annexure 1................................................................................................................................................. 29
List of figures and tables
Figure 1: Capitalisation of various railway concessions in SSA.................................................................... 6
Figure 2: Railway concessions awarded in Africa, 1990–2011...................................................................... 7
Figure 3: Track assessment of the ESA railways, June 2012....................................................................... 8
Table 1: Rail track condition in Malawi.......................................................................................................... 9
Table 2: Rail track condition in the DRC...................................................................................................... 10
Table 3: Rail track condition in Zimbabwe................................................................................................... 11
Table 4: Rail track condition in Zambia........................................................................................................ 13
Table 5: Rail track condition in Mozambique............................................................................................... 15
Table 6: Rail track condition in Botswana.................................................................................................... 19
Table 7: Track access fee annual income per km per annum..................................................................... 29
Table 8: Track access rates based on the ARTC report, AUD/US$ prices.................................................. 30
Table 9: Maintenance costs of US railroads converted to ZAR................................................................... 30
Table 10: Maintenance strategy assumptions............................................................................................. 31
Table 11: South African benchmark maintenance costs.............................................................................. 32
Table 12: Calculation of track access fees.................................................................................................. 33
RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa 1
Acronyms and abbreviations
AUD
Australian dollar
BBR
Beitbridge Bulawayo Railway
BOT
Build-Operate-Transfer
BR
Botswana Railways
CCFB
Companhia dos Caminhos de Ferro da Beira (Beira Railway Company)
CDN
Corredor de Desenvolvimento do Norte (Northern Development Corridor)
CEAR
Central East African Railways
CFB
Caminho de Ferro de Benguela (Benguela Railway)
CFM
Portos e Caminhos de Ferro de Mocambique (Mozambican Ports and Railways)
DFID
UK Department for International Development
DP
Discussion Paper
DRC
Democratic Republic of the Congo
EIB
European Investment Bank
ESA
Eastern and Southern Africa
est.
estimated
EU
European Union
GE
General Electric
hp
horsepower
IBRD
International Bank for Reconstruction and Development
IFI
international financing institution
IRCON
Indian Railways Construction Corporation
JICA
Japan International Cooperation Agency
JISR
Joint Infrastructure Sector Review
mgtk
million gross tons kilometre
mntk
million net tons kilometre
mtpa
million tons per annum
NOCZIM
National Oil Company of Zimbabwe
NPLI
New Limpopo Bridge Projects Investments
NRZ
National Railways of Zimbabwe
OPIC
Overseas Private Investment Corporation
RITES
Rail India Technical and Economical Services
RSZ
Railway Systems of Zambia
RVR
Rift Valley Railways
SADC
Southern African Development Community
SLA
service level agreement
SNCC
Société Nationale des Chemins de fer Congolais (DRC National Railways)
SPV
Special Purpose Vehicle
SSA
sub-Saharan Africa
TAZARA
Tanzania and Zambia Railway Authority
teu
20-foot equivalent unit
2 RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa
TFR
Transnet Freight Rail
TMSA
TradeMark Southern Africa
ton-km
ton kilometre
tpa
tons per annum
TRL
Tanzania Railway Limited
US$
American dollar
USA
United States of America
USTDA
United States Trade and Development Agency
ZAR
South African rand
ZR
Zambia Railways
RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa 3
4 RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa
1Background
1.In virtually all respects, most regional railway systems
in Eastern and Southern Africa (ESA) are not functioning
as they should. Their reliability is extremely poor; accident
and failure rates are high; operating costs are high; and
volumes of goods transported are low compared with
those transported by road (only about 5% of regional traffic
volumes – excluding those of South Africa – travel by rail);
and they generally make a financial loss. It can be safely
said that, in their present state and condition, the railway
systems in ESA, except South Africa, are not operationally
sustainable.
2.
For several of the ESA’s regional railway systems,
the levels of traffic volume and income have dropped
below those required for sustainable operations. The
income generated is spent first on salaries and fuel, with
inadequate funds remaining for maintenance and repair
of infra-structure and equipment. This is clearly shown
by the budgets and performance indicators. The railways
continue to decline and lose customers, and are unable
to attract the necessary funding required to return to
reliable and competitive levels. On average, traffic and
income levels would have to increase three to four times if
financial viability and a sustainable level of operations are
to be achieved.
3.The reasons for the decline in the region’s railways
have been well documented, namely poor management
and a lack of investment in railways, coupled with the rise
in importance of the road sector, which has received high
levels of public sector investment and subsidies. In most
sub-Saharan African (SSA) countries, roads are seen as
a public good and their construction, maintenance and
rehabilitation are covered by public donor funds and the
state budget. Although recently there have been moves to
introduce taxes, such as fuel tax, to finance the upkeep of
the road network, road users are still subsidised through
public funds. Conversely, railways operate on a userpays-all basis, with no or limited public sector subsidies.
In fact, because fuel tax is an across-the-board tax, the
road system is usually subsidised by the railway system
through the tax on diesel.
4.
Unlike the rail sector, the road sector has been
deregulated. This fact, coupled with technological
advances, has allowed trucks to carry higher payloads
at lower costs. This situation has introduced competition
between service providers in the road sector, as well as
between the road and rail sectors. The former sector is
now much more competitive than the latter, so that the
railways have lost traffic and business to the road sector.
This has resulted in a decline in railway revenue and
deferred maintenance, leading to further unreliability, loss
of capacity, loss of business and revenue, and a spiral of
decline.
5.
Throughout the ESA region, governments have
responded to this crisis mainly by concessioning their
railways to private sector operators – one for each rail
system. Concessioning countries such as Kenya, Malawi,
Mozambique, Tanzania, Uganda, Zambia and Zimbabwe
have followed a basic model developed by the World
Bank. The challenges they face in making concessions
operate effectively and efficiently have more to do with the
way the concessions were negotiated and the text of the
final agreement, rather than with the act of concessioning.
6.
An initial constraint for concessions was that
concession companies started with a far too limited capital
base, in part to lower the risk perception of private investors
(Figure 1). As a result, many concessions rapidly fell into
a cash-strapped situation as projected positive cash flows
did not materialise (World Bank, 2011).
RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa 5
Figure 1: Capitalisation of various railway concessions in SSA
Source: World Bank (2011)
7.After many years of neglect, governments and other
stakeholders are showing a renewed interest in their
countries’ railways. Efforts are being made to explore how
best to revive the railway systems so that they are part
of a more efficient multi-modal transport system and take
the pressure off the road transport sector. The road sector
is generally at a stage where it is increasingly difficult to
meet the region’s surface transport needs. This results in
high wear and tear on the road network, associated high
costs of road rehabilitation and maintenance, congestions
at borders that delay freight movement, and escalating
costs of imports and exports. In addition, there are
environmental, safety and economic benefits to moving
certain goods, such as fuel, acid, coal, minerals, cement
and grain, by rail rather than by road.
8.The privatisation of railway systems through long-term
concessioning was in many cases flawed. The process
took far too long, during which time no provision was made
for funding; the agreements were generally weak; and the
choice of concessionaire was often poor in that there was
a lack of serious bidders with the appropriate skills and
resources.
9.
At present it is not clear where funding to revamp
the region’s railway systems will come from. National
governments on their own do not have the required level
of resources. It is also unlikely that large international
financing institutions (IFIs) or development banks will be
able to finance a complete revitalisation of the railways.
The private sector will not invest in railways to the required
level unless some form of guarantee, in terms of securities
and revenue-earning opportunities (such as contracts with
mines, fuel distributors, etc.) can be locked in, in advance.
10.The solution possibly lies in creating financing
packages involving public funds from government budgets,
and grants and concessionary loans from donors, IFIs
and development banks to fund track infrastructure, with
the private sector providing the bulk of the financing and
management for the equipment and operations.
11.
A future model for seamless cross-border general
freight railway operations in the region could be similar
to that for road services, with the state providing the
track infrastructure and the private sector being mainly
responsible for the provision of equipment and operations
on the basis of a track access fee, similar to toll roads.
Scheduled rail services have the advantage of better
security than roads, allowing goods to be customs
cleared at the commercial end stations and thus avoiding
the problem of congestion and delays at border posts –
railways traditionally operate in this manner.
6 RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa
Figure 2: Railway concessions awarded in Africa, 1990–2011
Source: Bullock (2005).
Note: Since 2005, railway concessions have been concluded in Kenya (RVR in 2007, with a change of main commercial
shareholder in 2010) and in Tanzania (TRL in 2007, cancelled in 2010/11). In addition, the CDN and CEAR concessions
on the Nacala Corridor in Mozambique underwent a change of majority shareholding in 2009. The RSZ concession in
Zambia was cancelled in September 2012.
2
Description of the region’s railway systems
2.1Malawi
12.The 787 km railway in Malawi was historically linked
to Beira in Mozambique via a southern connection to the
Sena line in Mozambique. In 1970, a connection was built
to the inland CFM line from Nacala and it thus gained
a second outlet. In 1982, when the southern route was
closed due to war in Mozambique and damage to the
Zambezi bridge, Malawi became totally dependent on the
Nacala line. However, this line was also damaged and was
closed for several periods during the war. After the war, the
European Union (EU) funded the rehabilitation of the line
over most of its length and as far as Cuamba.
13.However, the Cuamba-Entre Lagos connection to the
border remains in a very poor condition, with an average
speed of no more than 10–15 kilometres per hour.
This has been a severe constraint on train operation on the
corridor. The southern leg of the Malawian network was
further affected when, in 1997, the Shire River washed
away a 300 m section of earth embankment north of the
sugar terminal at Bangula, 77 km north of the border.
Prior to this, the rail link to Nacala via Limbe had been
used for sugar exports, as the alternative much shorter
rail link to the port of Beira to the south had been out of
commission since 1985. As a result, there has been no
rail traffic of any kind from Bangula for the last 14 years,
and more than 100,000 tons per annum of sugar are
transported over 900 km by road to Beira via Blantyre, at a
cost of approximately US$80 per ton. This is about US$60
per ton more than a rail tariff on the 360 km link to Beira
would be.
RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa 7
Figure 3: Track assessment of the ESA railways, June 2012
Note: This map is based on a desk study by Bo Giersing of the track condition of the region’s railways by section and is a general assessment.
14.
The northern arm of the network runs due west
through Lilongwe to the Zambian border, and was recently
extended to Chipata. There is a long-standing proposal
to extend this line further into Zambia as part of a major
east-west corridor running from Nacala into central Africa,
and then to the Atlantic. However, it is unlikely that the
extension will be economically viable given the existing
potential for traffic.
15.In 2012, the Japan International Cooperation Agency
(JICA) completed a feasibility study looking at opening up
the road/rail route from Blantyre southwards and to the
port of Beira by building a bridge or large culvert where
the wash-away occurred, and by upgrading the road
and railway line to the southern border of Malawi with
Mozambique. If this section of rail were to be rehabilitated,
it would then be necessary to upgrade the spur from
the Malawian border to the recently upgraded Sena line
from Tete to Beira so that Malawi once again has rail
access to the port of Beira. However, future rail links and
operations from Malawi to the ports of Beira and Nacala,
and possibly a new bulk port, are likely to be influenced
or even governed by the development of alternative and
competing rail lines for exporting coal from the Moatize
coal mines in Mozambique.
16.In December 1999, the Central East Africa Railways
(CEAR) took over the 20-year concession on Malawi
railways, renewable in five-year tranches – this is effectively
the same consortium operating the concession to Nacala
on the Mozambican side of the border. The concession
operated reasonably well for the first two years, although
the delay in closing the concession in Mozambique meant
it was still dependent on the Mozambican Ports and
Railways (CFM) for the Mozambican leg of the trip. This
clearly created operational difficulties, particularly as CFM
evidently carried out little heavy maintenance during this
period, thus creating a poor level of service for the corridor
as a whole. This situation fed into CEAR finances and the
consortium was cash-negative until 2002, which was a
better year with a working ratio of 96%.
17.In January 2003, however, a bridge over the Rivi Rivi
River was damaged by Cyclone Delfina and the northern
8 RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa
arm of the network, including Lilongwe, was cut off from
the rest of the network. This effectively killed off almost all
local traffic, which dropped from 183,000 tonnes in 2002, to
about 20,000 tonnes per annum and also marooned rolling
stock and two locomotives north of the bridge. The bridge
was repaired using grant funds from the UK Department
for International Development (DFID), but this took more
than two years and the line only became operational again
in May 2005.
18.
The Government of Malawi recently signed a
concession agreement with the Vale mining company that
addresses two issues. The first is the construction of a
new section of railway line through Malawi to shorten the
distance from Moatize to Nacala, for the export of up to 25
million tons per annum (mtpa) of coal from the Vale coal
mines. The second is the rehabilitation of the existing line
in Malawi that is part of the Nacala line; that is, the CEAR
concession, of which Vale is a part.
Table 1: Rail track condition in Malawi
2.2
Democratic Republic of the Congo
19.The DRC railway system consists of two separate rail
operations: Onatra in the west between Kinshasa and the
Port of Matadi, and the state railway SNCC in the south
and east, connecting to the Southern African Development
Community (SADC) rail system in Zambia and Angola.
During the 1990s, the SNCC was privately operated. It
was a short-lived ‘concession’ – more precisely, a fiveyear performance-based management contract – in which
a consortium of Transurb and Transnet/Comazar assumed
responsibility for the network from Zambia in the north
into the DRC minerals belt, operating as Sizarail. The
arrangement lasted for about two years, at which point the
foreign management was arbitrarily ejected by the new
Kabila government after the exit of President Mobutu.
20.The SNCC rail system is undergoing a rehabilitation
programme, sponsored by the World Bank with an
investment of around US$280 million. Management
support is provided by the Belgian firm Vecturis under a
rolling contract, staffed partly by previous Transurb and
Comazar employees. The line from Mutshatsha to Dilolo,
connecting to the Angolan Benguela Railway (CFB) line
to Lobito, is not included in the World Bank rehabilitation
project, although it is essentially intact but in poor
condition. Rehabilitation of the 1,345 km Angolan section
of the Benguela Railway is expected to be completed
by the end of 2012. Cross-border rail traffic between
the DRC and Zambia in 2011/12 was insignificant, but
this is likely to change with the implementation of the
SNCC rehabilitation programme and the restructuring of
Zambia Railways.
RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa 9
Table 2: Rail track condition in the DRC
2.3Zimbabwe
21.
The Beitbridge Bulawayo Railway (BBR) is the
only recent regional concession in the classic BuildOperate-Transfer (BOT) form of constructing a new line,
as opposed to rehabilitating an existing line. The route
runs from Beitbridge on the South African border on the
Limpopo River to Heany Junction near Bulawayo, where
it rejoins the main Zimbabwean system. There has been
about 150 km of new construction at the eastern end, with
the western end consisting of the existing Nicholson West
branch of 170 km, which is a sharply graded branch line
that has received some upgrading. If the 28 km track of
the National Railways of Zimbabwe (NRZ) from Heany
Junction to Mpopoma Yard in Bulawayo is included, BBR
trains operate over a distance of 345 km. This is 184 km
shorter than the previous NRZ route via Rutenga and
Somabhula.
22.This substantial saving in rail distance was cited as
the main motivation for the new BBR railway. The existing
‘traditional’ route between Gauteng in South Africa and
Bulawayo via Mafikeng in South Africa and Botswana is,
however, only about 20 km longer than the BBR route.
Spoornet in South Africa, which had an agreement to
operate the BBR line, adopted the principle of the ‘shortest
distance rule’ to divert all traffic originating to the east of
a defined line to the west of Gauteng, via the BBR route.
This was seen to maximise the distance travelled and thus
the income generated for Spoornet and NRZ lines, but at
the expense of Botswana Rail. Shortly thereafter, rail tariffs
on the BBR route were increased.
23.The project dates back to the 1960s, but until
Botswana Railways was separated in 1987, there was
little incentive for NRZ to construct the line. They would
have had to create a shortcut reducing their haul to South
Africa by about 400 km, compared with the existing route
via Mafikeng.
24.BBR is a privately owned railroad company registered
in Zimbabwe. New Limpopo Bridge Projects Investments
(NLPI) owns 85% of the shares and NRZ the remaining
15%. The line was opened in July 1999 at a reported cost
of US$85 million, having been constructed in 17 months.
The line was designed to Spoornet standards, with a 20ton axle-load, and has a carrying capacity of 4 million tons
per annum.
25.
The terms of the concession were never made
public. The agreement is believed to have included
clauses requiring (or encouraging) NRZ to direct as
much traffic as possible over the line. For example, the
concession agreement may have included a ‘take or pay’
clause covering oil carried for NOCZIM, the National Oil
Company of Zimbabwe. At times, oil traffic destined for
Harare has been diverted south onto the BBR and then
re-consigned from Bulawayo, adding over 100 km to the
distance travelled. A similar clause is thought to apply to all
traffic between the south and Zambia and DRC, thus not
permitting the use of the Botswana route.
10 RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa
26.The establishment of BBR under the concession has
resulted in an almost complete cessation of transit traffic
through Botswana, declining from 800,000 to 1 million tons
in the late 1990s, to its current level of under 100,000 tons.
Traffic levels for Botswana Railways therefore consistently
fall below those required to operate on a sustainable basis.
NLPI is the main commercial shareholder in BBR and it
also controlled the Railway Systems of Zambia (RSZ)
concession in Zambia, which has now been cancelled. This
could lead to a change in regional rail traffic flows, as well
as the development of a more competitive environment.
27.Rail tracks in Zimbabwe are generally in a fair to good
condition, with 40 kg and 45 kg rail on concrete sleepers.
There was a reported total of 69 speed restrictions of
down to 10 km/hr in 2009 (over a total length of 320 km).
Relative to RSZ and Tanzania Railway Limited (TRL), NRZ
is in a reasonably good condition. However, the signalling
system is no longer functioning and the total freight traffic
in 2010 was about 2.5 mtpa.
Table 3: Rail track condition in Zimbabwe
2.4Zambia
28.Zambia Railways (ZR) consists of a north-south main
line linking with Zimbabwe at Victoria Falls in the south
and with the DRC at Sakania in the north. The TAZARA
line to Dar es Salaam joins the line at Kapiri Mposhi, and
is dependent on ZR to collect and distribute its traffic to its
customers some 200 km north on the Zambian Copperbelt.
The system serves two main purposes: carrying imports
and exports overwhelmingly to and from the south, and
hauling mineral-related traffic within the Copperbelt, some
of which are very short hauls.
29.The standard of the infrastructure and assets of ZR
had steadily deteriorated over time, and for several years
before the concession it was managed by expatriate
senior management under contract. During this period,
ZR received a loan from the International Bank for
Reconstruction and Development (IBRD), which was
used to help upgrade the track with concrete sleepers
and rehabilitate four locomotives. Immediately prior to the
concession, the railway was handling about 1.8 million tons
of freight and 400,000 passengers annually. About 30% of
RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa 11
Cargo Train Between Zambia and Zimbabwe
the freight by tonnage consisted of local traffic between
industrial plants within the Copperbelt.
passenger rail for seven years. The terms of the concession
are highly dependent on traffic volumes:
30.In December 2003, operation of most of the Zambian
Railway Corporation network (excluding the Mulobezi
branch) was handed over to RSZ, a company controlled
by NLPI, the parent company operating the BeitbridgeBulawayo line, but also partially owned by Spoornet.
Spoornet undertook the actual management under contract
to RSZ, but very quickly pulled out of the concession and
as a shareholder. Since then, RSZ has had no other
railway technical partner.
•
31.Spoornet initially agreed to a contract under which it
would be paid a fixed rate per net ton-kilometre (ton-km).
As infrastructure-related costs are largely independent
of traffic volumes, and should be covered whatever the
tonnage carried, Spoornet was exposed to considerable
traffic risk over which it had little direct influence and
started to lose money almost immediately. As a result,
Spoornet extracted itself from the concession. According
to Transnet Freight Rail (TFR), Spoornet handed over 12
mainline locomotives, valued at about US$20 million, as
compensation to RSZ and NLPI for its withdrawal from the
RSZ concession.
32.Two separate concessions were negotiated, one for
freight for 20 years, extendable to 30 years, and one for
•
•
For freight transport, an entry fee of US$0.75 million
applied, together with 5% of gross railway income.
There was also a fixed fee depending on a threshold
profit being achieved. On reaching this threshold, half
the additional profit would be paid as a fee. This was
the basis of the very large price (about US$250 million)
reported in the press at the time.
For passenger transport, RSZ agreed to gradually
increase the minimum service frequency on the main
line from 3–7 trips per week, in return for which it would
be entitled to reduce the fixed fee (assuming that this
was being paid) by an amount increasing from US$0.7
million per annum at the start of the concession, to
US$1.3 million after about three years.
In addition, RSZ agreed to invest US$14.8 million in
the freight business over five years and about US$0.5
million in the passenger business over four years.
33.
During the time between the concession being
awarded and the concessionaire taking over, one of the
main potential new sources of traffic, the development of
the Konkola Deep Mining Project in the Copperbelt, was
cancelled. This directly impacted the freight forecasts
underpinning the threshold profit in the concession
agreement.
12 RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa
34.According to the Zambia Railway Concession Review
Study (CPCS, 2010): ‘RSZ has been mostly compliant
to the concession agreement and non-compliance has
not been significant or central to their obligations. Noncompliance is limited to their obligations with respect
to reporting of financial and operating information;
maintenance and inspection practices of rolling stock and
fixed infrastructure; and off-setting monies owed to them
against variable concession fee payments. Relative to
four sub-Saharan African railway concessions compared/
benchmarked against in Annex A, the RSZ agreement is:
•
•
•
•
The same, in that it provides RSZ freedom to set
tariffs; though it is unique that it is the only that does
not include provisions for a minimum (or floor) tariff
Unique, in that it does not give the Government the
ability to license another operator in the event that the
Concessionaire seriously fails to address the transport
needs of one or a group of clients
Unique, in that it includes no provisions for ending the
contract for no cause.
The Agreement (Section 18-4) states that no party
shall have any right to terminate this Agreement other
than in accordance with the express provision of this
Agreement. And as there are no provisions for ending
the contract for no cause, we have based the default
amount on the event of GRZ default. We estimate the
default to be USD 12m. As well, we estimate the NPV
of the opportunity cost to the GRZ of cancelling the
agreement (and operating a government railway) to be
USD 30m relative to continuing with the concession.’
35.
On 11 September 2012, the state-owned holding
company Zambia Railways Limited resumed control of the
national rail network following the government’s decision to
revoke the operating concession awarded to RSZ in 2003.
Finance Minister Alexander Chikwanda had announced
the previous day that the government had decided on
‘compulsory acquisition’ of the concession rights, claiming
that RSZ had ‘blatantly disregarded the provisions of the
agreement’ and had been ‘acting in a manner prejudicial to
the interests of Zambians’. Chikwanda said failure to invest
in the maintenance and renewal of both infrastructure and
rolling stock had led to deterioration of the state-owned
assets, an ‘unacceptable’ level of derailments and poor
safety, including loss of life and property. Increased tariffs
and the poor quality of service had seen a significant drop
in rail use, with much traffic switching to road transport.
ZR undertook to absorb the 900 staff employed by RSZ
and would embark on renovation work ‘within the next two
weeks’.
Table 4: Rail track condition in Zambia
RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa 13
2.5Mozambique
2.5.1Beira
36.
This concession covers the 317 km Mozambican
section of the Zimbabwe-Beira route from the border at
Mutare (the Machipanda line), together with the 578 km line
north towards Malawi and Moatize (the Sena line), where
there are large coal resources, and the Marromeu (88 km)
and Vila Nova Frontiera (39 km) branches. In August 2004,
a concession was signed with the Beira Railway company
(CCFB) and included Rail India Technical and Economical
Services (RITES), the Indian Railways Construction
Corporation (IRCON) at 51%, and the Mozambican Ports
and Railways (CFM) at 49% (of which 16% is held by
CFM for subsequent disposal to Mozambican investors) to
rehabilitate and operate the lines for a 25-year period.
37.
The CCFB concession has been disappointing.
Operating within an atmosphere of conflict with the
conceding authority CFM, it was subsequently cancelled
in 2011, with control being returned to CFM. A report by the
European Investment Bank (EIB) during 2012 concluded
that CFM operations (the Sena and Machipanda lines)
would be highly profitable on the basis of Moatize coal
exports through Beira. Railway operations on the Sena line
are largely being carried out by the main coal exporters,
Vale and Rio Tinto, which operate their own trains and pay
a track access fee to CFM.
38.For many years, the Machipanda line has been the
main route for Zimbabwe’s overseas trade and over much
of its distance it runs parallel to the oil pipeline from Beira
to Zimbabwe. The Machipanda line continued to operate
throughout the civil war in Mozambique. The planned
upgrading of and investment in the line, and also the
projected increase in rail volumes have not taken place.
This is partly due to the poor economic climate in Zimbabwe,
but also to the limited depth of the Beira port owing to
siltation and lack of maintenance dredging. This situation
has previously prevented Beira from attracting freight from
the Copperbelt, despite its substantial advantage of being
400 km closer than any other port by road. A dredging
contract to return the access channel to its design depth of
12 m on the tide was completed in 2011, thus increasing
the future competitiveness of the Beira corridor.
39.The Sena line was closed in 1983 due to damage
during the war and was reopened in 2010, with a reported
capacity of 7 million tons per annum. The completion of
Disused Railway Line, Mozambique.
14 RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa
rehabilitation has been delayed, with severe cost overruns.
The commissioning of the line was completed in 2012, but it
has a limited capacity of about 3 mtpa. Ongoing upgrading
is being carried out to provide a capacity of 6 mtpa in the
first phase, to be followed by additional planned phases to
handle up to 18 mtpa, including a new coal terminal.
40.One of the main problems has been the inability of the
CCFB and CFM to negotiate a rail tariff for coal exports
with the two main developers – Vale and Riversdale – and
also the Sena Sugar Estates at Marromeu. This impasse
of about five years has finally resulted in mining companies
operating their own trains, at a track access fee of more
than US 3 cents per ton-km, which is about three times
more than a reasonable world benchmark. The CCFB’s
tariff is approximately 250% more than that offered by the
coalmines. The Sena line has been upgraded to 45 kg/m
rail on concrete sleepers and 20.5 ton axle loads, permitting
70 wagon trains carrying 4,500 tons of coal. The CCFB
has stated the capacity at 7 mtpa, but with the possibility of
increasing it to about 12 mtpa by adding additional passing
loops and some realignment, at an additional investment
of about US$150 million. However, it is clear that the Sena
line will not be able to accommodate all the projected coal
exports from Moatize.
41.
The link from Sena to the Malawi system remains
closed, despite the fact that the rail systems on both
sides of the border are concessioned, and that the cost of
reopening the line is likely to be financially viable and also
much cheaper than the longer road route through Tete.
Table 5: Rail track condition in Mozambique
RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa 15
2.5.2Nacala
42.In the 1970s, after completion of the rail link to Malawi,
Nacala was an important port for Malawi, but during the
1980s this route was closed due to the Mozambican civil
war. Currently most of Malawi’s international imports and
exports travel by road through Beira in Mozambique, while
most imports come in by road from South Africa. Because
of the discounted truck return loads to South Africa, some
Malawi exports are still routed through Durban. The
rail service between Malawi and Nacala is considered
unreliable and a direct surfaced road connection is
currently being built.
43.
The Northern Development Corridor (CDN), a
consortium led by CEAR, signed agreements with the CFM
for a 15-year concession of both the Nacala railway and
Nacala port in early 2000. This concession links directly
with CEAR operations in Malawi (from Nacala to Malawi)
and is intended to generate significant economies of scale
for both concessions. Finalisation of the concession was
delayed for five years while funding was being arranged
for rehabilitation of the 77 km stretch of line between
Cuamba and the Malawi border at Entre Lagos. The
rehabilitation was done with a US$29.7 million loan by the
Overseas Private Investment Corporation (OPIC) in 2003,
which also covered other works in Malawi and at the port.
The loan, although guaranteed by the two governments, is
based on commercial rather than concessional terms, with
interest at about 500 points above the base rate. Owing to
the delays and difficulties with concluding the concessions
and seamless operations between CDN and CEAR, this
project was not implemented for some time.
44.The concession finally began in January 2005. It
covers both the 610 km main line and the 262 km Lichinga
branch line. Although the main line inland to Cuamba is
in a good condition, much of it having been rehabilitated
with concessional loans, the remainder of the line to the
Malawian border and the branch line to Lichinga are both
in a very poor condition, with speeds limited to 10–20 km/
hr. The concession essentially provides operating rights
only. No rolling stock was involved in the sale, although the
CFM provided about 220 wagons as its equity contribution
to the concessionaire.
45.The concession payments include a fixed fee and a
variable component. The fixed fee is set at US$0.5 million
for years two to five, after which it increases to US$1.5
million until year ten, and to US$1.5 million thereafter. The
variable component is 5% to year five; then progressively
increases to 15% after year ten. During 2009 it was reported
that the Mozambican company Insitec had taken over the
interests of US investors RDC and Edlow Resources. The
main coalmining developer at Moatize, Vale, subsequently
took control of the concession by acquiring the Insitec
shares, and commenced with the construction and
upgrading of the railway line to Nacala to export up to 25
mtpa of coal.
2.6Tanzania
46.The performance of the Tanzania and Zambia Railway
Authority (TAZARA) and its volumes of freight have
remained fairly static over the past 15 years. TAZARA
has been unable to take advantage of the increased trade
volumes associated with the commodity boom behind
mining expansion in the Copperbelt. TAZARA suffers from
severe operating capacity constraints, primarily due to the
very low availability of mainline locomotives and a severe
shortage of working capital. Poor reliability and low asset
utilisation prevent TAZARA from attracting new customers
and increased freight volumes without a substantial
injection of working capital. It has sufficient equipment
and track capacity to carry four times the current freight
volumes.
47.
In recent years, TAZARA has carried 0.450 to 0.6
mtpa of freight annually, of which about 75% has been
made up of transit traffic to and from Zambia, the DRC
and northern Malawi via Mbeya. The maximum volume
carried was 1.2 mtpa in 1993, when there was an urgent
regional demand for maize imports for drought relief. Since
then, freight volume has fluctuated around the present
low level of 0.5 mtpa, which does not generate enough
income to cover the full costs of operations. Consequently,
TAZARA has been unable to fund the replacement and
maintenance of equipment and track infrastructure, with
deferred maintenance leading to poor reliability and
frequent breakdowns. This, in turn, has affected the
competitiveness of the service and the price that can be
charged to customers. The upper limit of freight pricing
is determined by alternative competitive routes and road
services on the Dar es Salaam corridor. If TAZARA could
perform better in respect of reliability and consistent transit
times, the demand for freight services would increase and
customers would likely be willing to accept a higher tariff.
Expansion of the TAZARA market is linked to increasing
the capacity of the system for a given performance
benchmark.
16 RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa
48.
TAZARA’s infrastructure and equipment were
designed to handle up to 5 mtpa of freight, and would
be able to accommodate more with the provision of
additional equipment and more frequent passing loops.
The original 60 Chinese 2,000 hp mainline locomotives
did not perform well and were supplemented in 1983 by
29 GE U30C locomotives of 3,200 hp. These locomotives
now constitute the mainline fleet, with 24 units remaining
serviceable. Of these, ten are operational on average, with
two for passenger services and eight for freight services.
TAZARA has more than 2,300 wagons of various types,
of which about 1,200 are of the open type, including 700
drop-sided wagons that are capable of carrying heavy
bulk goods, such as copper cathodes (50–60 tons) and
containers (2 teus).
cover the following essential elements:
i. Repair of the existing GE U30C locomotives and
the rectification of deferred maintenance on all
operating units, with the aim of achieving a locomotive
availability of more than 90%. Initially, a fleet of 12
locomotives should be available at all times, with
a high degree of reliability (measured in distance
between failures), rising to 20 locomotives as freight
levels rise to 1.5 mtpa within the two-year period. The
increased availability of locomotives will significantly
reduce train turnaround times, and consequently
improve the utilisation of equipment and also reduce
operating costs. For its locomotive repair programme,
TAZARA should enter into a strategic partnership with
a locomotive and railway equipment specialist.
49.If open wagon trains can be operated on a ten-day
turnaround trip between Dar es Salaam and Kapiri Mposhi,
which means they can be loaded three times per month,
then the freight carrying capacity of the open wagon fleet
is more than 2 mtpa, subject to wagon and locomotive
availability. A freight volume of 2 mtpa, with a 60:40 ratio
for imports and exports, could be achieved on a ten-day
turnaround trip with 20 locomotives and 650 wagons –
this is less than 90% of the current TAZARA fleet. Transit
times of four days have been achieved on the TAZARA
line in recent years, but not consistently, demonstrating
that a ten-day transit time (or less) could be achieved if
the problem of low equipment reliability can be resolved. A
four-day transit time translates into an average line speed
of less than 20 km/hr.
ii. Repair of workshop equipment and building up of
strategic stock spares and materials, assumed to cost
about US$2 million and to be funded by the workshop
contractor or partner.
50.The most important operational and capacity constraint
on the TAZARA system is the poor availability of mainline
locomotives. This is partly due to lack of funds to purchase
and maintain a critical stock of replacement / service spare
parts. Key workshop and track maintenance equipment
has broken down for the same reason -a lack of spare
parts. Poor availability of locomotives also translates into
wagon shortages and consequent poor wagon utilisation,
which are loaded only about once a month on average.
51.
A ‘steady state’ operational and costing model has
been prepared to illustrate the current and short-term
future business prospects for TAZARA. The model is
based on utilising TAZARA’s existing infrastructure and
equipment, without any major new purchases or major
track upgrades within the first two-year period. China is
to provide ten new locomotives (six in 2012) and 90 new
low-sided wagons. However, TAZARA cannot pick itself up
and improve its performance without additional funding.
The model assumes that a US$30 million working capital
loan funding will be made available to TAZARA, either in
the form of cash and/or guarantees, to be repaid at 8% per
annum over a ten-year period. The funding is intended to
iii. Repair of track maintenance equipment, such as
tamping machines, ballast cleaners and also rescue
cranes, assumed to cost US$3 million. Ongoing
maintenance of this equipment is covered in the track
maintenance budget.
iv. Emergency upfront track repair in critical sections
to ensure safe operations (US$5 million), including
an operational communications system. The model
makes provision for an annual track maintenance
budget of US$4.5 million per annum in the initial 0.5
mtpa freight volume, increasing to US$11.2 million per
annum when freight levels of 1.5 mtpa are achieved,
which could be within two years.
v. An allowance of US$5 million for working capital,
but determined by the initial cash flow analysis. It is
expected that, within a few months of the locomotive
repair programme, cash flows will improve due to
better service levels, higher income and lower unit
costs.
52.
During 2012, a series of customer meetings and
negotiations took place between TAZARA and its main
customers –copper exporters and fertiliser importers. Talks
centred around a proposal that customers fund TAZARA’s
critical working capital requirements through short-term
loans governed by performance-based contracts. These
negotiations have been ongoing since September 2012
and are likely to be influenced by the cancellation of the
RSZ railway concession in Zambia. This could provide
TAZARA with direct access to the Copperbelt. TAZARA
has accepted the principle of open access for multiple
operators..
RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa 17
2.7Botswana
53.
The Botswana railway system was originally built
as part of the North-South railway from the Cape to the
Copperbelt. It provided the main railway service between
South Africa and the SADC region north of Zimbabwe until
1999, when the traffic to and from Gauteng was switched
to the new BBR route. The Botswana system was part
of the NRZ system until 1986, when it was taken over by
Botswana. Originally, the same railway system operated
all the way to the Congo border, until it was split into
separate national systems, which required locomotive
changes at the borders for all regional rail services. There
is now a tendency to combine the systems again, partly
through operational agreements to offer seamless railway
services, such as the Maputo Corridor.
54.The route length is 888 km, located on the eastern side
of Botswana. It previously carried about 2.5 mtpa, falling
more recently to 1.8 mtpa, with the loss of transit traffic
to the BBR route and using about 1,000 wagons and 34
locomotives. All the equipment is in need of upgrading or
replacement, and Botswana Railways (BR) has developed
a turnaround strategy in an attempt to motivate the
government to provide the necessary financing for this.
55.
Prior to 1999, BR was the traditional railway route
between South Africa and the north, rather than the route
through Beitbridge, Rutenga and Bulawayo, which is more
than 200 km longer. A direct line between Beitbridge and
Bulawayo was concessioned to the NLPI on a BOT basis,
with Spoornet as initial contracted operator. This BBR
route effectively closed the route through Botswana on
the basis of the ‘shortest distance rule’, even if the cost
or tariff on the new route is substantially higher. The new
route was shorter by a nominal 30 km, and this was clearly
a profit-motivated manipulation - serving to increase
regional transport costs - contributing to loss of rail market
share. Freight is still moved from Botswana to the north
but very little comes in from NRZ because of the switch to
the BBR route. BR traditionally had low rail tariffs and was
considered efficient and the track in good order. BR was
also profitable when transit traffic was more than 300,000
tpa, but financing is currently assisted by the state. TFR
recently entered into discussions with BR to develop a
closer working relationship, with the intention of making
increased use of the Botswana route for transit freight
traffic destined for the north. This is likely to be incentivised
by the recent cancellation of the RSZ rail concession.
56.BR has suffered operating losses for more than 12
years, for the following reasons:
• Loss of revenue as a result of the decline in transit
traffic between South Africa and Zimbabwe, routed
•
•
•
•
•
•
through BBR after 1999
Operational inefficiencies, leading to low asset
utilisation and availability
High fixed costs, resulting in a high breakeven point
Severe maintenance backlogs, resulting in a high level
of equipment failures
Escalating costs of fuel, spares and labour against a
declining revenue base
Lack of growth in freight traffic
Increasing competitiveness from road hauliers, which
has captured almost the entire general freight and fuel
transport market.
57.In order to address these issues, BR effectively has
three strategic options for the future:
• Continue to operate as present, but attempt to cut
costs – this is likely to lead to further loss of the
freight market due to non-performance, which could
eventually lead to bankruptcy.
• Transformation – through a new focus on transforming
management, applying a specific action plan, obtaining
substantial investment in new equipment by the
government and rectifying the maintenance backlog –
will be difficult to sell and achieve.
• Restructuring through, for example, privatisation,
strategic partnerships with selected operators, or new
investment and resources from outside. Strategic
partnerships may be easier, quicker and less risky.
58.
The isolation of BR and the desire to attract new
markets for freight has prompted the consideration of
three new railway projects:
• A direct link to the Zambian system via Kazungula in
Zambia, in order to bypass Zimbabwe and the BBR/
NLPI operations. The economic and financial viability
of this option has always been in question.
• A link across the Kalahari to the Namibian system,
which is a distance of more than 1,000 km. Potential
traffic volumes indicate that this would not be viable,
even for possible coal exports from Botswana, and
that it would not attract private sector funding.
• A direct link to the TFR system at Ellisras to shorten
the distance for possible future coal exports from
Botswana. This project may be realistic within the
next few years, depending on coal prices and regional
demand for new coal-fired power stations.
59.Botswana is increasingly providing key road routes
between South Africa and the Copperbelt, through the
Martin’s Drift border post and the Zambezi ferry crossing
at Kazungula in Zambia. The construction of a new road
bridge at Kazungula is now being implemented.
18 RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa
Table 6: Rail track condition in Botswana
3
Revamping the railways in ESA
60. In order for the region’s railways to be revamped and operated as a single network, the following considerations
need to be taken into account:
i)
Cost structures of railways
iii)
Operating speed
The fixed costs for railway operations generally vary
between 60% and 80%, mostly depending on the freight
volumes and related asset utilisation. For road services,
fixed costs make up about 40% of operating costs. This
is why it is important for rail to achieve a very high level
of infrastructure and equipment utilisation. The freight
volume on the coal line in South Africa is about 60 mtpa,
the train turnaround time is about two days and the tariff
is about US$0.015 per ton-km. The operation is very
profitable. For TRL in Tanzania, the volume is 0.5 mtpa, the
train turnaround time is 17 days, the tariff is US$0.065 per
ton-km, and the operation is loss making. An operational
model has indicated that if TRL were able to reduce the
turnaround time to seven days and increase traffic to 1.5
mtpa (the previous level in 2004), tariffs would be reduced
to less than US$0.05 per ton-km. This would mean an
operating margin of 12%, after servicing a loan of US$105
million at 8% per annum over ten years.
It should be noted that an increased operating speed will
not necessarily result in shorter journey times. For example,
the average transit time on Rift Valley Railways (RVR)
from Mombasa to Kampala is ten days. The actual train
travelling time, at an average speed of 20 km/hr, is 2.5 days.
The train is therefore effectively standing still for 7.5 days,
owing to scheduling, breakdowns and derailments. If a 2.5day guaranteed transit time can be offered to customers at
a competitive tariff, then RVR is likely to capture a large
proportion of the cargo currently travelling by road. The
first priority for the railways would be to make operations
safe for a given speed restriction (e.g. 20–30 km/hr) and to
prevent the breakdown of equipment. Properly maintained
equipment hardly ever suffers breakdowns – the problem
is usually deferred maintenance, where the equipment is
used until it breaks down.
ii)
iv)
Indicative capital costs for track
The cost of a complete upgrade of a railway, with new
sleepers and rails to a 20 ton axle load, is in the order of
US$400,000–500,000 per km. The cost of a new track of
1,067 mm gauge (‘Cape gauge’, which is the predominant
gauge in use in ESA), with new formation and structures, is
approximately US$1–1.5 million per km and up to US$2.5
million per km for heavy haul high-speed railway lines. It
is therefore much cheaper to upgrade an existing rail line
than to build a new line.
Threshold traffic volumes
A preliminary analysis has indicated that existing railway
systems require minimum freight volumes of around 1.5–2
mtpa to be financially viable, taking into account that most,
or all, of the existing infrastructure is ‘sunk capital’. For new
railways, minimum traffic volumes should be more than 10
mtpa, and more than 20 mtpa for high-speed heavy haul
lines. The region’s railways need to attract much larger
traffic volumes – at least twice the current volumes – to be
economically viable.
RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa 19
v)Capacity
All the regional railway systems are operating at well below
their original design capacities. They are suffering from
severe capacity constraints due to poor track condition
and poor locomotive and wagon availability, with many
units stabled. The railway systems will therefore be unable
to handle more traffic without substantial investment in the
repair and upgrading of track and equipment, as well as
the provision of working capital.
vi)Financing
Despite the content of their concession agreements,
concessionaires are unlikely to be willing or able to fund
major upgrades of track infrastructure. Initial investments
from the government to pay for rectifying deferred
maintenance and upgrades have often been based on
the loan repayments being serviced by the concession
payments. In many cases this has not materialised, and
hence the funding has been held back. If concessionaires
cannot see how their investment will be returned,
irrespective of what the agreement says, they will also
stop funding. If one party is deemed to be in default, then
the other party often withholds payment as well – this
atmosphere of conflict can carry on for many years.
vii)
New projects and standard gauge
The policy has been adopted that all new railway projects
in Africa should be built to standard-gauge specifications.
This is a perfectly logical approach, assuming that
new projects can be ring-fenced and can be shown
to be economically and/or financially viable, based on
committed or guaranteed traffic volumes and income
streams rather than speculative projections. The new lines
will be implemented when they are shown to be bankable,
whether they are standard gauge or not. The first priority
should be to improve the existing operations to the point
where they are financially viable. In most cases, this will
require a substantial investment of US$100–200 million to
cover deferred maintenance in track and equipment, new
equipment and track upgrades. Information from proposals
to upgrade the East African railways from narrow to
standard gauge suggests that upgrades will cost US$2–4
million per km, depending on the amount of realignment
needed.
One current proposal is that existing tracks should be
upgraded to standard-gauge specifications in respect of
track formation, structures, track ballast and sleepers;
allow a third rail to be installed in a progressive manner;
and possibly permit the simultaneous operation of both
the narrow and standard-gauge systems. However, the
standard-gauge specifications of higher speeds and axle
loads will require a new track to be realigned in areas of
difficult topography, and it may be more practical to build
the whole line on a new alignment. If a programme of
progressive upgrading of existing narrow-gauge track to
standard gauge is adopted, this will require that a detailed
design for the whole standard-gauge line be completed
first, showing which sections are to be realigned and
which structures strengthened. The track formation would
also have to be strengthened, at additional cost, to handle
substantially heavier axle loads.
In addition, there are a number of proposed new railway
projects, some of which have been around for many
years (e.g. the line from Kafue in Zambia to Lion’s Den in
Zimbabwe that was first suggested in the 1920s but is still
a project proposal). The important lessons learnt are:
• not to base project decisions on political criteria but
to ensure that a project is economically, financial and
environmentally viable;
• to take account of previous proposals and feasibility
studies, as new projects have often been researched
and studied previously, sometimes more than once;
• to take full account of the impact of a new investment on
existing and alternative surface transport investments.
viii)
Renegotiation of concession agreements
Almost all concession agreements are not performing as
well as expected and there is frustration on both sides.
Often both parties to the concession agreement (the
government and the concessionaire) are at fault, in that
neither party implements the agreed concession in full. In
Zambia, this dissatisfaction with the concession and the
concessionaire has resulted in the state-owned holding
company Zambia Railways Limited resuming control of the
national rail network following the Zambian government’s
decision to revoke the operating concession awarded to
RSZ in 2003. However, it is not clear whether, or by how
much, the concessionaire will be compensated.
In general, it is unlikely that the blame for the failure of a
concession to reach acceptable levels of service delivery
can be laid entirely at the door of either party. Legal action
by one party against the other could therefore result in
a long and expensive litigation process, with no obvious
predetermined outcome. During this process the railway
may well decline even further, perhaps reaching a point
where it is no longer a viable business operation. Under
these circumstances, and where there is still a choice in the
actions to be taken, it may be more effective to renegotiate
the concession using an independent arbitrator, so that it is
clear what the responsibilities of each party are and binding
performance indicators can be built in. If railways that are
concessioned are to be turned around, the following has
to be considered:
20 RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa
•
•
•
In some cases, the agreement may need to be
revised and/or renegotiated. The concessionaires
are generally operating in an atmosphere of conflict
and there are complaints and accusations of
non-performance on both sides. The necessary
investments in infrastructure and equipment are not
being made, and traffic projections and budgets are a
long way off the stated targets. The present situation
is unsustainable and there is a need to chart the way
forward. The concession may need to be adjusted so
that it is clear that the government owns the track and
the concessionaire provides the service. This will allow
service providers to agree a ton-km tariff with users,
and for the government to have a guaranteed revenue
stream. This, in turn, would allow the government to
borrow money for maintaining the infrastructure and
so allow the private sector to provide the service on
the track owned by the public sector.
Railway operators need to prepare realistic, detailed
business plans, initially focusing only on the core
activities necessary to increase targeted bulk and
intermodal freight volumes, and omitting all items not
absolutely necessary for the efficient operation of the
railway – such as studies for new lines, state-of-theart signalling systems, new locomotive and wagon
fleets, upgrading to standard gauge, etc. Detailed
cash flow projections should be prepared, linked to
performance targets and agreements or memoranda
of understanding with key customers and showing the
long and short-term financing requirements.
In simple terms, regional railways will all have to
increase their freight volumes substantially in order
to become viable. Building new lines and linkages
will not redress the situation, unless linked to specific
contracted anchor projects (e.g. Moatize coal). Prior to
the 1980s, railways were partially protected in respect
of their volumes and the tariffs charged. This is no
longer the case unless there is market interference
by governments. In most cases, there is not enough
bulk traffic to sustain the railways, and the railways
will have to win back both bulk and intermodal traffic
from road transport. This will require much improved
reliability and associated investment.
ix)
Access agreements
Starting from the premise that private sector participation
is the best way to open up development opportunities
in the region’s railways, there is currently a window of
opportunity for the private sector to become involved in the
operation and maintenance of its own trains (consisting
of a locomotive and wagons), only paying the stateowned railways for access to the track. Where relevant,
maintenance facilities would have to be leased, which
would be managed and resourced by a private company.
There is a strong demand from the private sector to
operate its own trains and for the railways to enable such
open access arrangements, meaning that any operator
that conforms to a specified set of standards governed by
a railway regulator should be able to operate a scheduled
service.
This open-access model allows the private sector to
identify relevant rail freight transport demand and to
operate and maintain a rolling stock fleet profitably to
meet this demand. However, for this model to be effective,
it is vital to set the access fee (the amount charged by
the owners of the track per ton-km of freight moved) low
enough to allow rail freight to be competitive as opposed
to road transport, but not so low as to not cover the cost of
maintenance and make a small profit.
An assessment (see Annexure 1) was done of the access
fees charged in Australia, Europe, South Africa and the
United States of America (USA) with a view to establishing
an ESA benchmark access fee. Apart from the cost of
maintaining the track, other components should also be
taken into account when calculating track access fees:
•
•
•
Cost of operating the network (train control systems
and staff)
Cost of repaying any capital investment in the new
installed capacity, over and above normal renewal
maintenance
Network operator’s profit.
It was found that, assuming a capital investment of
US$500,000 per km and a 50% operator’s mark-up, the
access fee for the region would be approximately one US
cent per net ton-km.
RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa 21
4
Conclusion and the possible way forward
61.
Revitalisation of the ESA railway system is not an
option; it is essential if the region is to reach the levels of
economic growth needed to trade its way out of poverty
and into sustainable growth and income generation. It is
clear that the volumes of freight that will need to be moved
will increase significantly in the short to medium term,
which will place additional strain on the region’s transport
system. No single mode of transport will be able to meet
the region’s requirements.
62.The national railways making up the regional network
need to cooperate so that it can be operated as a single
network, much in the same way that the regional road
network operates as a single regional network with
harmonised axle loads, transit systems, etc. The current
practice of having to change locomotives and sometimes
wagons at the border should be addressed if the railways
are to be able to offer an alternative transport service to
road transport.
63.There are many significant challenges to be overcome
in bringing railways to a level at which they are able to
provide an efficient, competitive and reliable service, but
these are not insurmountable. The railways are currently
not viable businesses but they have not deteriorated to such
an extent that they cannot be turned into viable sustainable
businesses that play a vital role in the economic recovery
of the region.
•
65.The World Bank has been involved in supporting the
successful restructuring of ongoing concession contracts
in Cameroun and Madagascar. The main pillars of this
restructuring are the following:
•
•
•
•
64.
A number of useful studies have focused on how
to revitalise the region’s railways, and some have listed
reasons why railways have been unable to compete
against road transport:
•
•
•
Overestimation of the serviceable freight markets
available to railways due to competition from road
freighters and the uneven playing field for rail vs road
Underestimation of rail investment needs, with longerterm investment needs being higher than anticipated
owing to the poor state of the railways at take-over by
the concessionaires
Undercapitalisation of concessions, with some
concessionaires becoming cash-strapped as projected
positive cash flows do not materialise and the debt
burden becomes harder to accommodate over the
long term. This is also complicated by the structure
of the concession. If, for instance, a concession
consists of a consortium of two or more companies,
then each partner will be expected to capitalise the
concession according to the ratio of its ownership of
the consortium. If a partner, even a junior partner, is
unwilling or unable to contribute to the capitalisation of
the concession, this will delay all partners.
Undue expectations regarding rail passenger services
– since 1996, none of the rail passenger services
managed by private operators have been financially
viable and misunderstandings and friction have
been created between host governments and rail
concessionaires.
•
•
Private operators take on the exclusive responsibility
of financing maintenance and renewal of rolling stock,
and shoulder only the cost of track maintenance.
Host governments agree to finance the long-term
cost of track renewal but also become business
partners based on the implementation of profit-sharing
agreements designed to partially compensate them
for their financial support.
Host governments’ financial commitments to
infrastructure investment are partially secured by the
concessionaire’s payment of an infrastructure renewal
fee into a secured account it manages on behalf of the
host government.
Concession contracts acknowledge upfront estimated
infrastructure amounts for at least 15 years in order
for host governments to grasp their long-term net
commitments to the concession (after payment of the
infrastructure renewal fee, profit-sharing agreements,
and any other concession fee payments).
Intermodal competition policies are instituted to
rebalance the playing field between road and rail
transport (e.g. enforcement of axle loading for trucks
along competing corridors, tolling of roads, etc.).
Separate accounting of passenger and freight services
is put in place in order to recognise the cost imposed
by the government’s public service obligations for
passenger services.
66.
Not all of these lessons learnt in restructuring the
concessions in Cameroun and Madagascar may be
appropriate for all concessions in ESA, but a number of
them are applicable. For the ESA situation the following
general principles are suggested:
i.
Track and immovable railway infrastructure (including
22 RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa
marshalling yards, communications equipment,
maintenance sheds, stations, workshops, etc.) should
remain the property of the state where it is already the
property of the state.
ii. The state should establish its own track maintenance
company, possibly through a Special Purpose Vehicle
(SPV), or it should concession only the maintenance
of the track and communication equipment to a
private sector operator with a proven record of track
maintenance and railway operations, and under a
performance-related contract. The company operating
the track should not be involved in any way with the
running of the trains.
iii. The company owning the track will need to make
significant investments in the rehabilitation and
upgrading of the track to allow service operators to
provide an efficient service. These investments have
to be made to ensure that the service operators can
use the most appropriately sized locomotives (e.g.
the 3,000 hp GE U30Cs have an axle load of 20 tons
per axle, which exceeds the present capacity of the
Zambian railways). Service operators must be able to
load the wagons to capacity, as this is a function of the
power of the locomotive (the size of the train), as well
as of the axle load limit of the track. This will ensure
that trains endure less wear and tear and are more
reliable. Experience with the 16 running locomotives
on the TAZARA line from July 2011 to June 2012
shows that 45% of the locomotive failures were a result
of traction motor failures, while 37% of the locomotive
failures were due to loose gear case bolts, hot axles
and failing wheel sets. Traction motor failures may be
the result of poor winding and rewinding, but traction
motors can also fail if they are stressed as a result
of going slowly and up a gradient. Poor track quality
could be partly responsible for loose gear case bolts,
hot axles and failing wheel sets.
iv. Private sector service operators should be able to
buy a license to operate their own locomotives and
rolling stock on the track. The level of private sector
participation may be in the form of ownership of
wagons; ownership of locomotives; ownership and
operation (including employment of drivers and
private fuel pumping); or ownership, operation and
maintenance. Also, various options exist for the
outsourcing of track maintenance and signalling –
typically to a joint venture company with state equity,
or a totally privately owned company with an agreed
fee payable to the government.
v. The company in charge of track maintenance rents
out the track to private sector operators at a fixed
rate per ton-km. As a guideline, track access should
consist of a rate per ton-km (covering the infrastructure
owner’s costs to maintain and renew infrastructure,
plus a reasonable profit) and a fee per train (covering
the cost of network operations staff that control train
movements on the network, plus a reasonable profit).
The latter cost could cover the cost of breakdown
services, or such services could be charged if and
when they are required. The access fee should not
exceed US$0.01 per ton-km.
vi. The company in charge of the track may operate on
a take-or-pay scheduled service level agreement
(SLA). This means that the track operator would be
able to allocate a required number of train slots for
each private sector operator to run a scheduled train
service. If, for example, the track operator allocates
three slots a week to a service operator to carry a
fixed amount of cargo from A to B, then the service
operator would be guaranteed this access and the
track operator would be guaranteed this income,
whether or not the operator carries the agreed tonnage
of freight. The track operator will have a guaranteed
and predictable income against which it could borrow
money to upgrade and maintain the track and also
make loan repayments. The service operator would be
able to guarantee the service to itself or its customers,
as it will have guaranteed slots and be able to run a
guaranteed scheduled service.
vii. Each private sector operator will need to ensure that
its rolling stock and locomotives conform to a minimum
set of safety standards. These will vary, based on the
operating criteria of the railway system.
viii. Private companies and freight owners should be able
to purchase track access and operate and maintain
their own trains on the region’s national networks.
Countries would probably still need to operate at
least suburban passenger services and mainline
passenger services, but could outsource all onboard
and maintenance services to private contractors under
an SLA, as these normally need to be subsidised by
the government.
RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa 23
ix. Regional governments would need to facilitate an
independent regulator to control safety, including
technical and operational interfaces, and ensure an
equitable commercial environment via a regional
multilateral agreement.
x. Agreements would need to be made between national
railway systems, so that a train operator can operate a
train without having to change either the locomotive or
the wagons at a national border. The region’s national
railway networks could be operated as a single
regional railway system and so provide an efficient
service that is able to compete with road transport in
terms of the time taken to travel and the costs per tonkm, as well as ensuring the safety of the cargo from
origin to destination.
67.
The single most important motivating factor for
continued long-term investment in regional railways is likely
the fact that diesel-powered rail transportation accounts
for about 25%–30% of the diesel fuel used by road
transportation (per net ton-km). If fuel prices double relative
to other operational costs, railway costs would increase by
about 15%, whereas road costs would increase by about
40%. Rail transportation is likely to become good business
again, and will therefore attract increasing interest from
the private sector in future. Interestingly, Warren Buffett,
widely regarded as one of the most successful investors in
the world, recently purchased BNSF Railway, the largest
railway company in the USA.
24 RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa
Bibliography
References
Australian Rail Track Company (ARTC), 2008. 2007/08 Annual report of the Australian Rail Track Company (ARTC).
Adelaide, South Australia.
Baumgartner, JP, 2001. Prices and costs in the railway sector. EPFL, Lausanne.
Bullock, R, 2005. Results of railway privatisation in Africa. World Bank Transport Papers TP-8. World Bank,
Washington, DC.
CPCS, 2010. Zambia Railway concession review study: Final report. Prepared for the World Bank, August 2010.
CPCS Transcom Limited, Ottawa, Canada.
Giersing, B & Howard, M, 2011. Draft Tanzania-Zambia Railway (TAZARA) strategic plan: Interim report – initial
operating assessment. Discussion document not for distribution, July 2011. TradeMark Southern Africa, Pretoria,
South Africa.
Grimes, GA & Barkan, CPL, 2006. Cost-effectiveness of railway infrastructure renewal maintenance. Journal of
Transportation Engineering, 132(8): 601–608.
Pozzo di Borgo, P, 2011. Africa railway concessions: Lessons learned and potential solutions for a revival of the
sector. PowerPoint presentation, March 2011. World Bank, Washington DC.
Additional sources
BNSF Railway, 2009. Upgrade of the Dar es Salaam to Isaka Railway – Feasibility study, September 2009. Done
on behalf of the Tanzania Ministry of Infrastructure Development and financed by the United States Trade and
Development Agency (USTDA).
CPCS, 2009. East African Railways master plan study, January 2009. CPCS Transcom Limited, Ottawa, Canada.
Phipps, L, 2009. Technical report: Review of the effectiveness of rail concessions in the SADC region, March 2009.
Report done for the Southern Africa Global Competitiveness Hub, Gaborone, Botswana.
TAZARA, 2010. Rail subsector review paper for the Fourth Joint Infrastructure Sector Review (JISR) in Transport,
September 2010.
World Bank, 2006. Review of selected railway concessions in Sub-Saharan Africa. World Bank, Washington, DC.
Websites
TradeMark Southern Africa – www.trademarksa.org
RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa 25
26 RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa
ANNEXURE 1
A note on the calculation of access fees
Australian Track Corporation benchmark
The Australian model is considered relevant for benchmarking fees in the ESA region, as it is based on private rail
operators gaining access to a state-owned rail network (ARTC, 2008). In the 2007/08 financial year, ARTC had 77.7
billion gross ton-km on its 10,000 km of track and its revenue from track access charges was AUD329,181,000, which
is what it charges for maintaining the track and upgrading where necessary, as well as controlling the train movements
from the central traffic control. The private train operators employ their own drivers and deploy their own rolling stock.
The above figures translate into AUD0.00424 per gross ton-km, which equates to US$0.424 cents per gross ton-km
(1 US$ to 1 AUD) or, in terms of net ton-km (assuming a gross to net ratio of 1.33 for a 20 ton per axle service with 20
ton tare wagons) US$0.56 cents per net ton-km. Table 7 shows this track access in terms of annual income per track
kilometre for various annual net tonnages.
Table 7: Track access fee annual income per km per annum
The ARTC report indicates a cost of maintaining rail infrastructure of US$35,000 per km per annum, which is high
compared with South Africa. Nevertheless, the analysis shows that for a well-maintained network, relatively low track
access fees (US$0.0056 per net ton-km) exceed the cost of maintaining the network at above about 6 mtpa.
Table 8 shows the costs per gross ton-km and the price per train for different categories of train. The least expensive end
of the ARTC price range is probably more applicable for conversion to the ESA region, as the Southern African region
will likely be at a maximum of 20 ton per axle and low traffic.
RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa 27
Table 8: Track access rates based on the ARTC report, AUD/US$ prices
North American benchmark
Grimes & Barkan (2006) give a comparison of the major North American railroads’ annual ordinary and renewal
maintenance costs. The data is repeated in Table 9 below and shows an average for all the railroads of US$2,507.33 per
million gross ton-miles (in 2001 US$). It can be seen from the table that maintenance costs for North American railroads
are about US$0.36 cents per net ton-km.
Table 9: Maintenance costs of US railroads converted to ZAR
European benchmark
Baumgartner (2001), whose study gives orders of magnitude only of costs in a European context (and whose figures
are of a similar order of magnitude to the North American context) estimates that for 7.5–11 mgtk (million gross tons-km)
per year, track maintenance costs €10,000–20,000 per km per annum. Taking the mean of €15,000 and adding 2% for
structures gives €15,300 per km per annum. Escalating this figure at 5% per annum gives €26,168 in 2012 terms. At an
exchange rate of ZAR10 = €1, this gives ZAR261,682 per km per year. For 10 mgtk per year (or 7.5 mntk per year), this
equates to ZAR3.5 cents per net ton-km (US$0.438 cents).
28 RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa
South African maintenance costs
The maintenance strategy assumptions (from which the South African maintenance costs are calculated) are given in
Table 10 below. South African benchmark network maintenance costs are shown in Table 11. These assumptions and
costs were adapted from work done in association with R&H Railway Consultants in the SADC region. It can be seen
from the table that the cost of maintenance is ZAR193,347.21 per km per year.
Table 10: Maintenance strategy assumptions
RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa 29
Table 11: South African benchmark maintenance costs
Calculating rail track access fees
As mentioned, apart from the cost of maintaining the track, other components must be taken into account when calculating
track access. These are the cost of operating the network (train control systems and staff); the cost of repaying capital
investment in new installed capacity (over and above normal renewal maintenance costs); and the network operator’s
profit.
Table 12 opposite shows an example of track access calculation for the same 500 km example considered in the section
above, assuming a US$250 million capital investment (US$500,00 per km) and a 50% operator’s mark-up. The result in
the table is a track access fee of US$1.1 cents per net ton-km.
30 RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa
Table 12: Calculation of track access fees
RIRN Discussion Paper - Revamping the Regional Railway System in Southern and East Africa 31
“
Revitalisation of the East and Southern African railway system is
essential if the region is to reach the levels of economic growth needed
to trade its way out of poverty. Freight volumes are going to increase
significantly - placing additional strain on the region’s transport
system... No single mode of transport will be able to meet the region’s
requirements.”
Regional Integration Research Network | TradeMark Southern Africa
+27 12 349 7500
www.trademarksa.org/rirn | [email protected]
First Floor Building 41, CSIR Campus, Brummeria, Pretoria, 0001, South Africa