This record is a partial extract of the original cable. The full text of the original cable is not available.

UNCLAS SECTION 01 OF 05 PRETORIA 000113 
 
SIPDIS 
 
STATE PLEASE PASS USAID STATE PLEASE PASS USGS 
 
E.O. 12958: N/A 
TAGS: EPET ENRG EINV EIND ETRD ECON SF
SUBJECT: RESTRUCTURING OF THE SOUTH AFRICAN ELECTRICITY 
INDUSTRY -- THE SUPPLY SECTOR 
 
REF: PRETORIA 00094 
 
THIS CABLE IS A CORRECTED COPY OF REF A:  FORMATTING 
PURPOSES 
 
¶1. (U) Introduction:  In light of recent government 
initiatives to restructure the South African energy 
industry, post will submit separate reports on the 
supply, distribution and transmission sectors.  Each 
cable will serve as a general review of that sector, and 
be the launch-pad from which subsequent cables on 
specific topics are prepared, as developments take place 
in each sector.  The first, dealing with the supply 
sector, follows.  The purpose of these cables is to 
indicate the potential opportunities for investment, 
technology and service providers created by the energy- 
restructuring program.  End introduction. 
 
--------------------------------- 
Electricity Supply Industry (ESI) 
--------------------------------- 
 
¶2. (U) Summary: At the beginning of this decade, in the 
absence of new supply, the South African Government 
concluded that demand for electricity would soon overtake 
existing generation capacity.  Government tasked the 
National Energy Regulator (NER) to develop a National 
Integrated Resource Plan (NIRP) for insuring security of 
future supply.  The first NIRP was completed and 
published in March 2002.  At the beginning of 2003, the 
NER established an Advisory and Review Committee (ARC) to 
solicit stakeholder contributions to the NIRP process. 
The updated study (October 2003), generated under the 
guidance and approval of the NER, was carried out by a 
NIRP team comprising the Eskom Resources and Strategy 
Group (headed by their Managing Director, Dr Steve 
Lennon), the Energy Research Institute (ERI) of the 
University of Cape Town, and the NER. 
 
¶3. (U) In the early seventies Eskom forecast an 
electricity growth rate of 8%, and consequently built and 
commissioned a number of six-pack (six generation units 
per station) coal-fired stations.  In the middle of the 
construction program, the demand growth rate dropped as 
low as minus 0.4%, causing a serious over capacity, and 
three of the stations - Camden, Komati and Grootvlei - 
were "mothballed".  Decommissioning ended in about 1990. 
The excess capacity allowed Eskom to become the lowest 
cost producer of electricity in the world.  Eskom's 
tracking of electricity consumption identified a 
significant increase in demand (from about 2000), and 
they forecast that new peaking capacity would be required 
by 2007, and new base load capacity by 2011.  In October 
2003, the NIRP-team updated the 20-year energy supply 
plan based on the new demand projections, and estimated 
capital expenditure of some $17 billion to 2010.  The 
main purpose of the plan was to identify the most cost- 
effective and environmentally friendly combination of 
options and technologies available to ensure South Africa 
of a timely, reliable and quality supply of electricity 
in the future.  Schedules and technologies will 
inevitably change over time as new developments occur. 
End summary. 
 
Current Capacity and Technology Mix 
----------------------------------- 
 
¶4. (U) South Africa's total licensed generation capacity 
is about 44,000 MW of which Eskom (state-owned 
electricity utility) owns 42,000 MW.  Eskom mothballed 
3,600 MW of capacity in 1990 and their total net 
operating capacity (NOC) at December 31, 2003 was 36,200 
MW.  Coal-fired stations generate 86% of the electricity, 
nuclear 5%, pumped-storage 4%, hydro 2%, emergency gas 
turbines 1%, and 2% is imported from the Cahora Bassa 
hydro station in Mozambique. 
 
¶5. (U) Eskom currently operates 10 large coal-fired power 
stations, a two-reactor 1,800 MW nuclear station (Koeberg 
located 30 kilometers north of Cape Town), six small 
hydro stations, and two pumped-storage schemes that play 
a critical role in meeting peak demand.  Municipalities 
own 22 small power stations and back-up gas turbines, but 
these total only 5% of national generation capacity and 
generally run at low load factors.  Municipalities also 
own the transmission lines and the transformers located 
within their boundaries.  Private generators comprise the 
remaining 2% of capacity.  In 2003, the peak demand was 
32,000 MW, equal to 89% of NOC.  Estimated demand growth 
is at least 1,000 MW per annum, and the current growth 
rate stands at around 4.6%. 
¶6. (U) According to Mike Cary, Managing Director of Rotek 
Engineering an Eskom group engineering company, a 
complicating factor to the capacity problem is that the 
power stations are ageing.  The ages of the newest 
stations vary between 10 and 24 years, and the equipment 
on these stations can be as old as 30 years.  Although 
Eskom has a comprehensive maintenance program, the old 
equipment is more prone to failure.  Part of the risk- 
management strategy is to carry spares, but financial 
constraints limit this option.  Over the past 12 months, 
Eskom has experienced an abnormally high failure rate (14 
major supply disruptions in the Johannesburg area alone) 
and some of the plant transformers have failed to such an 
extent that some critical spares are no longer available. 
 
----------------------- 
Eskom's Expansion Plans 
----------------------- 
 
¶7. (U) Government decided in 1998 to restructure Eskom 
and to establish a multi-market model (MMM) for 
electricity trade in South Africa.  During the four-year 
model planning process, Eskom submitted its investment 
program for new electricity capacity (as its contribution 
to the NIRP) on an annual basis.  The estimated cost of 
the expansion program was about $32 billion (over 20 
years), and aimed to ensure that South Africa had enough 
electricity to power its growing economy. 
 
¶8. (U) However, during this time Eskom became a virtual 
"on-looker" as government decided that Eskom would not be 
allowed to build any new generation capacity as the 
government wanted to have the MMM in place before the 
excess capacity ran out.  Eskom, however, remained the 
supplier of last resort and obligated to ensure 
sufficient supply to the country.  Part of the plan to 
restructure Eskom was to sell 30% of their generation 
capacity, valued at more than $4 billion, to foreign 
power utilities.  Due to a policy shift, this process 
remained in limbo.  Eskom and Government now face crucial 
decisions about new power plants, and the balancing of 
financial, operational and environmental criteria in 
selecting the type and mix of energy projects it should 
develop. 
 
¶9. (U) Following the April 2004 elections, new Public 
Enterprise Minister Alec Erwin (Eskom's shareholder 
department) rescinded the "restructuring" of Eskom.  This 
was part of a general policy shift aimed at strengthening 
State Owned Enterprises to ensure that the social 
objectives of government would be met first.  Erwin fast- 
tracked the go-ahead for new investment in electricity 
supply, as outlined in the NIRP.  On October 20, the 
Cabinet approved $13.5 billion in capital expenditure for 
Eskom to return to service the mothballed stations, build 
70% of new capacity, and upgrade existing stations, 
transmission lines, and distribution networks. 
Government expects Eskom to source much of this capital 
from the markets.  Erwin also "offered" independent power 
producers (IPP's), particularly foreign utilities, a 30% 
slice of the proposed new generation capacity.  In this 
way, he hopes to ensure healthy competition for Eskom and 
to attract foreign direct investment into the 
infrastructure sector of the country without damaging 
Eskom's ability to lead the expansion drive.  Steve 
Lennon, Eskom's Managing Director for Resources and 
Strategy, has stated that the investment decision was in 
time to avoid early supply disruptions in 2006 and 2010. 
 
Short-Term Expansions to 2010 
----------------------------- 
 
¶10. (U) The latest update of the NIRP's power expansion 
strategy (October 2003) deals with expansions envisaged 
up to 2022.  The first phase deals with immediate 
electricity needs over the next five years.  This 
requires Eskom and IPP's to add at least 1,000 MW of 
capacity every year, from 2005 to 2010, to avoid 
shortages during peak usage time.  The investment package 
portions the $13.5 billion as $9.7 billion for 
generation, $2.2 billion for distribution, and $1.6 
billion for transmission.  The plan is for Eskom to 
return to service three coal-fired stations between 2005 
and 2011 -- total base-load capacity of 3,600 MW.  In 
addition, Eskom is to build a new 1,000 MW open cycle gas 
turbine (OCGT) plant by 2007 (location not specified), 
and by mid 2005 the DME plans to request the private 
sector to tender for a 1,000 MW OCGT plant (location not 
specified), fueled by a light diesel distillate, to be 
operational by 2008. 
 
Longer-Term Expansions - 2011-2022 
---------------------------------- 
 
¶11. (U) By 2010 Eskom will have to start commissioning 
power plants that will add to the base load for which 
planning will start in 2005.  According to Lennon, the 
program was based on the forecast of an increase in 
demand of 1.5%-4% per year over the next 20 years. 
However, there are contingency plans for an uptake of 5% 
should demand exceed estimates.  The long-term base load 
capacity would include a combination of new power plants 
and imports of electricity from a number of African 
projects.  Capacity increases include: 
-- a $0.7 billion, 1,330 MW pumped-storage scheme at 
Braamhoek in KwaZulu/Natal Province, by 2013, 
-- a $0.8 billion, 1,000 MW pumped-storage scheme at 
Steelpoort in Mpumalanga province, by 2014, 
-- a third 1,000 MW pumped-storage facility at Monontsa 
in the Free State province, and planned for 2019, 
-- a $1.4 billion, 1,500 MW combined cycle gas turbine 
(CCGT) facility, possibly located near Coega (new deep 
water port under construction) in the Eastern Cape 
Province, by 2013, 
-- three generation units (total rated capacity of nearly 
2,000 MW) added to the six already operating at the 
Mathimba power station in Limpopo Province, at a cost of 
$3.2 billion, 
-- a 4,000 MW greenfields coal-fired power station, at an 
estimated cost of almost $6.4 billion, near coal reserves 
in either Limpopo or Mpumalanga provinces, or in 
neighboring Botswana or Mozambique. 
 
Summary of Capacity Expansions to 2022 
-------------------------------------- 
 
¶12. (U) Some electricity expansion options considered by 
the NIRP team extend beyond the 2022 time-period, but are 
shown for information completeness.  Obviously, these 
technologies and time schedules will change over time as 
energy demand, innovations and other developments occur. 
 
Generation Stations            Implementation New (MW) 
and Technologies               Schedule Capacity 
--------------------            ---------------------- 
3 De-mothballed coal-fired      2005-2011    3,600 
(possibly fast-tracked to 2009) 
10 Single/Open Cycle Gas (SCGT) 2006-2010    2,640 
1 SCGT                             2022        240 
3 Dry-cooled coal-fired         2016-2022    9,630 
2 Pumped-storage                2012-2014    2,330 
1 Pumped-storage                   2019      1,000 
5 Fluid Bed Combustion          2012-2014    2,330 
1 IPP Open Cycle Gas (OCGT)        2008      1,000 
Total new capacity                          22,770 
 
Other possible options for the future (probably beyond 2022) include: 
3 Combined Cycle natural gas                    3,470 
1 Nuclear Pebble Bed cluster of 8 reactor units 1,320 
1 Coal-fired                                    3,780 
1 Pumped-storage                                1,000 
Total possible new capacity                     9,570 
 
PBMR 
---- 
 
¶13. (U) According to Lennon, environmental criteria are 
critically important in deciding the combination of new 
power plants to be developed.  This could benefit non- 
coal projects, as Eskom's coal-fired plants are the chief 
emitters of carbon dioxide and other pollutants. 
Environmental considerations could also provide a boost 
to development of South Africa's fourth-generation 
nuclear-based pebble bed modular reactor (PBMR), which 
could be ready for commercial launch by 2013, and 
features in Eskom's long-term expansion plans, from about 
2015 onwards.  The capacity of a single PBMR reactor unit 
is 165 MW, and that of an 8-pack cluster, 1,320 MW. 
Johan Kriek, CEO of PBMR Ltd, plans to start construction 
of a PBMR demonstration unit in the second quarter of 
2007, the first commercial unit by 2013, and to supply 
Eskom with 24 units (three clusters) between 2013 and 
¶2023.  Lennon said that negotiations have started in the 
process to develop a PBMR supply agreement between PBMR 
Ltd and Eskom. 
 
Imports 
------- 
 
¶14. (U) The African electricity import projects on the 
Eskom drawing board include Cahora Bassa in Mozambique, 
the proposed Inga3 hydro-electric project in the DRC, and 
buying surplus power from the Southern African Power 
Pool, as available.  Lennon said that part of Eskom's 
strategy could be to put equity into Cahora Bassa, 
depending on the financial details of such a deal.  In 
addition, in November 2004, the South African utility was 
one of five Southern African utilities that bought equal 
equity stakes in Westcor, the company earmarked to build 
the $5 billion power station at Inga3, and the associated 
transmission and interconnect supply lines.  The project 
has the potential to supply 3,500 MW of electricity by 
2011, but many experts believe this to be an unrealistic 
target, given the political (and investment) 
uncertainties in the DRC. 
Status of the Expansion Plan 
---------------------------- 
 
¶15. (U) The first phase of the expansion, dealing with 
new capacity over the next five years, is well under way. 
Eskom has commissioned the return to service of three 
mothballed power stations at a total cost of $1.9 
billion.  These three facilities, when refurbished, 
should supply 3,600 MW to the grid.  The largest of the 
three, Camden, is set to return to service in 2005.  The 
DME has also authorized Eskom to build a new 1,000 MW 
open cycle gas turbine (OCGT) plant by 2007 (location not 
specified), and a 1330 MW pumped-storage station 
(Braamhoek) by 2013, as part of a $13.5 billion 
investment.  In December, the DME issued an IPP 
"Expression of Interest" document for a 1,000 MW OCGT 
facility (location not specified), fueled by a light 
diesel distillate, to be operational by 2008.  The tender 
document should be available by August 2005, and Eskom, 
as the purchaser of the electricity, cannot bid for the 
project.  The final leg of Eskom's short-term revamp will 
be to upgrade a number of its older base-load stations, 
including a $160 million upgrade of the 2,100 MW Arnot 
station.  Through demand and supply side efficiency- 
improvement strategies, Eskom management plans to take 
4,000 MW out of the system by 2013. 
 
Financing the Expansions 
------------------------ 
 
¶16. (U) Eskom spokesman, Fani Zulu, has indicated that 
Eskom would finance the $13.5 billion expansion for the 
first three years through a combination of cash flows and 
loan financing.  After that, Eskom is likely to revert to 
the local and offshore capital markets.  Eskom, has a 
relatively low 30% gearing, and has some room on its 
balance sheet to raise its debt level.  A crucial part of 
the financing would be the electricity price, which the 
NER determines.  The regulator has recently granted Eskom 
below-inflation tariff increases, but Minister Erwin has 
backed Eskom's call for more realistic prices to enable 
the huge capital expenditure requirement. 
 
¶17. (U) In summary, the main conclusions drawn from the 
latest NIRP study by energy experts, are that: 
-- Options for diversifying away from coal-fired plants 
are insufficient to meet forecast electricity demand over 
the next 20-years, 
-- Economic justification for diversification plants 
would prove difficult in the absence of persuasive 
measures such as penalties and subsidies for non-use or 
use, 
-- Clean coal technologies and demand side efficiency 
strategies are required to meet environmental standards, 
-- Base load plants would be required for commercial 
operation after 2010, 
-- At assumed future costs, and after returning the Eskom 
mothballed plants to service, fluidized bed combustion 
(FBC) technology offers the most economical option, 
followed by coal-fired plants, and then CCGT plants (in 
the western Cape), using imported gas/LNG, 
-- NIRP plans indicate that 920 MW OCGT peak load plants 
must begin commissioning from 2008, 
-- NIRP plans assume the attainment and sustainability of 
demand side management targets, power plant availability, 
imports, and interruptible supply strategies, 
-- short-term peak-load requirements can be facilitated 
using single cycle gas turbines fueled by locally- 
produced synthetic gas. 
 
------------------------------------------ 
Summary of South African Energy Statistics 
------------------------------------------ 
¶18. (U) Projected Electricity Supply/Demand (MW) 
 
                      2004    2008    2012    2016    2022 
Generating Capacity 38,620  41,990  46,060  50,700  57,540 
Peak Demand         34,620  38,530  42,020  45,660  51,890 
Reserve Margin (%)    11.6     9.0     9.6    11.0    10.9 
New capacity           -     3,390   8,460  13,140  19,990 
 
Current Supply (MW) 
 
Eskom: 
Coal-fired               32,070 
Nuclear                   1,800 
Pump Storage              1,400 
Hydro                       570 
Gas Turbines                340 
Total                    36,180 
 
Non-Eskom: 
Coal-fired                1,320 
Pump Storage                180 
Hydro                        70 
Gas Turbines                 90 
Total                     1,660 
 
Imports: 
Cahora Bassa (hydro)        780 
Total                    38,620 
 
South Africa's Energy Mix 
 
Product                       Electricity   Energy 
               End Use %      Generation %  Supply % 
Coal           86             69%           30 
Crude          -              18%            - 
Biomas         -              9              8 
Nat gas        1              3%             - 
Hydro          2