Namibia Electricity Supply Demand Options March 2008 FINAL

Renewable
Energy &
Energy MINISTRY OF MINES AND ENERGY
Efficiency
REEECAP
Institute Renewable Energy and Energy Efficiency Capacity Building Programme
Electricity Supply and Demand

Management Options for Namibia.
A Technical and Economic Evaluation.
FINAL REPORT
March 2008
Prepared by:
P O Box 1900
Windhoek
Namibia
Tel + 264 - 61 – 224 725
Fax + 264 - 61 – 233 207
Email contact@emcongroup.com
Electricity Supply and Demand Management Options for Namibia. A technical and economic evaluation.
Executive Summary
The Executive Summary starts with a presentation of key findings – a summary of the
summary – before giving the full executive summary:
KEY FINDINGS
The Namibian electricity industry stands at a cross roads. Domestic electricity generation is
not adequate to meet current and future projected demand. To complicate this, supply
constraints in South Africa are resulting in Eskom not being able to meet South African
electricity needs let alone those of Namibia. The cost of sourcing electricity has almost tripled
in the last eight years.
The study developed eight generation scenarios that aimed at illustrating different
applications and interpretations of the Namibian energy policy as defined in the 1998 White
Paper on Energy Policy. This energy policy spells out a number of goals that should
influence the decisions on how to procure electricity.
This report will show that decisions need to be taken and these decisions need to be taken
quickly. The price for a lack of action today will be very high indeed.
This study set out with three objectives.
The first objective was to determine whether Namibia can meet its electricity demands.
• It was found that this can be done but urgent action is required.
• Continued reliance on imported electricity could result in blackouts equal to 10% of

electricity demand by 2012. This has serious economic implications. It will impact on
livelihoods and destroy jobs.
• Demand side management is economically efficient. It has a role to play in avoiding

blackouts and should be implemented as a matter of urgency.
The second objective was to identify generation and demand management options within the
context of the White Paper. It was found that:
• Relying on imported electricity is the worst option. It is economically inefficient, results

in the most blackouts and pollutes the global village.
• The most economically efficient mix is a balanced combination of a wide variety of

generation options. This is cost effective, creates jobs, generates incomes and keeps
emissions low.
• Other, more specific, policy options like energy security or maximising the diversity of
supply involve difficult tradeoffs between cost, blackouts, emissions, etc.
The third objective was to determine the role that renewable resources could play. It was
found that:
• The use of some renewable resources is economically efficient.
Page i of xii
• Within a balanced generation mix:
o Renewables, other than hydro power, could supply up to 20% of demand at a

slight price premium.
o When hydro power is included, renewables could supply up to 90% of

electricity demand.
o Renewables reduce future price risk, resource risk, supply risk and emissions.
o Renewables create jobs and contribute to gross value added (GVA).
INTRODUCTION TO THE EXECUTIVE SUMMARY

The Namibian electricity industry stands at a cross roads. Internationally high oil and
commodity prices have had a ripple effect on the price of gas and coal. Energy prices are
escalating at alarming rates. Global warming and a growing awareness of the damage
generated by fossil fuels is seeing increasing pressure on countries to turn to other ways of
generating energy. Within Namibia there are growing misgivings about the Van Eck dry coal
fired power station on the outskirts of Windhoek. And Eskom, the South African power utility
that supplies half of Namibia’s electricity, has not been able to keep pace with electricity
demands in South Africa and is being forced to reduce power to parts of South Africa and
neighbouring countries. Compounding this is the historic availability of cheap electricity from
Eskom which has been an inhibiting factor to the construction of any new power stations in
Namibia and to this day dominates the price expectations in Namibia.
At the same time Namibia has very limited options with using the traditional generation
technologies of coal fired, hydro or nuclear power. Namibia does not have any coal reserves
that have been proven to be exploitable and must import coal. It is simply cheaper to import
electricity from South Africa than import South African coal. Namibia is a very dry country
which limits hydro options. Nuclear is an option but is constrained by high capital costs and a
relatively small skills base.
Yet there are opportunities. Namibia is one of the sunniest countries in the world. There are
feasible wind resources at the coast. There is a major offshore gas field. In the north of the
country there are thousands of hectares of alien vegetation that desperately need clearing
and are an obvious fuel source. There is the Kunene River that holds promise for additional
hydro power.
This report will show that decisions need to be made and these decisions need to be made
quickly. The price for a lack of action today will be very high, already in the near future.
This study develops eight generation scenarios that aim at illustrating different applications
and interpretations of the Namibian energy policy as defined in the 1998 White Paper on
Energy Policy. This energy policy spells out a number of goals that should influence the
decisions on how to procure electricity. The Energy White Paper identifies self-sufficiency,
security of supply, inclusion of renewable energy source, sustainability and cost
effectiveness and efficiency as potential policy objectives.
This study therefore aims to inform the policy interpretation and application process in
Namibia by illustrating what the various policy drivers might mean in practice, both at a
technical and financial level and in terms of the economy. To our knowledge this is the first
comprehensive study for Namibia that considers alternative energy sources and fossil fuel
Page ii of xii
and hydro electricity generation options. It also explores in detail the financial and economic
implications of various possible supply mixes.
THE CHALLENGES:
Namibia produces power locally and imports about half its electricity, with the bulk of this
being supplied by Eskom in South Africa and the balance by ZESCO in Zambia. NamPower
also imports on a smaller scale from Zambia for supply to the Caprivi Region and exports on
a small scale to Angola and Botswana. There are currently three power stations in Namibia
which are owned and operated by NamPower, i.e. the Ruacana hydro-electric power station,
Van Eck coal-fired power station, and the Paratus diesel power station.
The challenge faced by Namibia is that domestic generation is not adequate to meet current
and future projected demand. To complicate this, supply constraints in South Africa are
resulting in Eskom not being able to meet South African electricity needs let alone those of
Namibia. For the past twenty years Eskom has had surplus generating capacity and has sold
electricity at very low prices by world standards. However, slow and stagnant electricity
supply industry (ESI) reforms and government intervention in South Africa have delayed
Eskom capacity extension plans, resulting in the capacity shortages that are now prevalent.
Namibia has in the past had very favourable long-term contracts with Eskom, based on the
excess capacity in South Africa. A new contract has been negotiated recently, but this no
longer offers firm supply and prices are escalating rapidly due to the change from excess
capacity to supply shortages in South Africa. The nominal cost of sourcing electricity as
reported by NamPower has almost tripled in the last eight years.
Namibia therefore finds itself in a difficult position. Its security of supply is compromised and
power shortages are likely during the coming years because of insufficient local generation
capacity and constrained imports. This is an unprecedented occurrence in a country that has
enjoyed a reliable and stable power supply in the past. Power outages were an uncommon
and rare occurrence in recent Namibian history. At the same time electricity prices are rising
dramatically due to a conglomeration of factors. This places strain on people and business
as well as on the utilities struggling to fund the necessary investments and still provide
quality service.
THE APPROACH
This study set out to address potential solutions to the pending electricity crisis and to
explore the role that renewable resources could play in the solution. The results to the
analysis were technical results, on the one hand, to ensure the supply of electricity is
adequate to meet growing demand at a reasonable price. On the other hand the analysis
also aimed at showing the economic, social and environmental implications of the technical
analysis.
The approach that was adopted was to start with the technical analysis and use the results
as an input into the economic analysis.
The technical analysis starts with a demand forecast about the likely increase in demand
for electricity in Namibia. Second a range of generation options and demand management
options were identified, assessed and costed. Finally, these inputs were processed through
an electricity dispatch model.
Page iii of xii

Nine feasible generation options were identified and three demand side management
options. The generation options are:
• Baynes Hydro-Electric Power Generation

• Coal-Fired Power Generation
• Natural Gas Power Generation
• Nuclear Power Generation
• Electricity Generation from Biomass
• Concentrating Solar Thermal Power Plants
• Solar Photovoltaic Power Plants
• Wind Energy Power Generation
• Integrated Solar Combined Cycle Plant
On the demand management side the options are:
• Compact Fluorescent Lamp roll out

• Ripple Control for water heating
• Solar Water Heater roll out
Following the determination of likely future demand and identification of potential generation
and demand management options an electricity dispatch model was developed. The model
takes a specified constellation of generators, demand side management options and load
growth and solves the dispatch equation until supply and demand balance. If it cannot match
demand with existing supply then there is a residual called unserved energy.
The dispatch of generators and the revenue requirement for transmission and distribution
define the cost to be recovered through electricity sold and thus determine the end consumer
costs. As distribution consumers are price sensitive higher prices result in reduced
consumption, and a different dispatch pattern for the generators. The model is designed to
iterate this sequence and solve for the lowest cost and least unserved energy.
Following this the generation and demand side management options were grouped into eight
scenarios where each scenario attempts to depict a particular policy option. The scenarios
are:
Scenario 1: Base Case. This scenario depicts a policy of using proven and conventional,
large scale, centralised generation options. In this scenario only two power plants are built,
namely a 350MW coal plant commissioned in 2011 and 360MW of additional hydro on the
Kunene River in the form of Baynes in 2017. This is augmented by increasing imports from
Eskom, based on the assumption that Eskom will be able to supply increasing capacity again
starting from 2012. Van Eck power station is retired as soon as replacement capacity can be
obtained from Eskom. This scenario does not implement demand side management or
energy efficiency options.
Scenario 2: No New Local Generation. This scenario centred on the analysis of doing as
little generation as possible within Namibia and relying primarily on electricity imports. It
could also be seen as depicting a scenario where no investment decision is made even
though there are possible options. This scenario produces severe amounts of unserved
demand, meaning that wide spread load shedding cannot be avoided. This scenario also
does not implement demand side management
Scenario 3: Price minimised. This scenario centred on providing the cheapest electricity
possible while still avoiding extensive load shedding. A biomass power plant is built in 2009
and expanded in 2010. A small coal plant in 2011 and a small concentrating solar plant is
built in 2010. This is necessitated in order to cover the time before Eskom capacity again
Page iv of xii
becomes available. Baynes hydro opens in 2017 to harness additional the hydro potential of
the Kunene River.
Scenario 4: Maximum Supply Diversity. This scenario centred on the analysis of

diversifying the generation options as much as possible in order to minimise the risk
exposure to any one particular energy carrier or plant technology. In this scenario a coal fired
station opens in 2011; biomass production starts in 2009 and is scaled up over the next three
years; concentration solar starts in 2009 and is doubled in 2013; wind power takes off in
2009; and a combined cycle gas turbine (CCGT) from Kudu commences in 2017. Solar PV is
not build because of cheaper alternatives nor is nuclear because of the very long lead time.
Van Eck is retired and Baynes is not built because it utilises the same river as Ruacana,
posing correlated risk. Demand side management is included in this scenario.
Scenario 5: Generation Self Sufficiency. This scenario centres on generating as much

electricity within Namibia as possible. Imported fuels are part of the mix as are the imported
components of capital and operating expenditure on both the supply and demand side. This
scenario calls for the construction of a small coal plant which, because of the short lead time
opens in 2011; a gas power plant which opens in 2012 as well as the Baynes hydro plant
that would come on line in 2017. These are augmented with smaller renewable plant, which
offer short implementation times and contribute to long term supply stability and diversity.
Demand side management is implemented.
Scenario 6: Energy Imports Minimised. This scenario is similar to scenario 5 but apart
from just limiting electricity imports it aims at also minimising imports of fuel and coal. This
generation mix consists of biomass starting in 2009 and scaling up over the next two years;
concentrating solar and wind both coming on line in 2010 and Kudu gas coming on line in
2012. It would also be necessary to build a small coal station as base load support until the
gas power station can be completed. Demand side management is included.
Scenario 7: Maximum Renewable Energy Options. This scenario centred on using as

many renewable energy generation options as possible and seeks, as far as possible, to
avoid fossil fuel generation. The resulting generation portfolio relies heavily on concentrating
solar (with storage) as well as biomass and wind, while also building Baynes hydro and
continuing reliance on imports from Eskom. This is one of the few scenarios where even a
solar PV plant is built despite its high price. Van Eck is retired later than in the other
scenarios. Demand side management is implemented.
Scenario 8: ‘All options considered’. This scenario was compiled because it mixes a
strong renewable focus with a strong concern for supply diversity to minimise risk exposure
to energy carriers and technology types. It is intended as the main “counterpoint” to the ‘base
case’ scenario, and also seeks to bring in strong decentralised elements. This scenario relies
on a small coal plant, biomass and Baynes hydro as the backbone for 24 hour electricity
supply. This is complemented by an array of diverse renewable generators, each limited in
size to ensure grid stability and cost effective grid integration as well as a balance between
dispatchable (firm) and intermittent generators. This scenario implements demand side
management. The scenario has excluded the building of nuclear or gas plants. Nuclear is
excluded because of its environmental issues and long lead time. Gas is not built because of
the current difficulties in getting the Kudu project going.
Two types of economic analysis were used. These are a cost benefit analysis and a
macroeconomic analysis. Cost benefit analysis is a means of taking all the direct costs and
all the direct benefits of a proposed project and comparing these. It is the conventional
method that is used in project appraisal. The outcome of this analysis is the reporting of a net
present value (NPV), a benefit cost ratio (BCR) and an internal rate of return (IRR). This
analysis is done by adjusting for shadow prices and wages and removing the potential
Page v of xii
distortions caused by taxes and subsidies. A high BCR is usually a good indicator that it
would be possible to raise finance to implement a project. In the case of a private sector
investment the good BCR would be part of the business case to funders. If it is a public
infrastructure project, a high BCR should give confidence that it is worth funding the project
directly from its Treasury or, alternatively, to make suitable institutional arrangements for the
involvement of the private sector in the project’s funding.
The second type of analysis is macroeconomic analysis. This focuses on the overall
contribution of the proposed project to the national economy. It reports on contribution to
gross domestic product (GDP), job creation, tax generation, etc. The size of a national or
regional economy is measured in terms of the sum total of all economic activities taking
place within the area concerned, both in the public and private sectors. For countries like
Namibia, this necessarily includes measures of informal sector activity as well. The name
given to the measure of the size of the economy is GDP. Sometimes this is referred to as
(GVA) which is a slight variation on GDP. The unit of measurement is the national currency.
While there are a number of different types of macro-economic effects, the two most
important are contribution to GDP and the creation of jobs. The importance of job creation is
obvious. Increases in GDP are synonymous with increases in peoples’ economic standards
of living. Increased GDP – i.e. increased production – is experienced in the form of more
jobs, higher wages and reduced economic hardship.
ANALYTICAL RESULTS
The analytical results of this study are presented in nine parts:
1. Cost of the Generation Options
A calculation was made of the capital cost for each generation option as the cost per MW of
generating capacity. These costs include the overall capital costs, which consist of the
overnight capital cost, the interest during construction and the decommissioning costs. It was
found that the generation option with the lowest capital cost is the combined cycle gas
turbine, followed by wind, biomass and clean coal. The most expensive option is
concentrating solar, followed by nuclear, Baynes and solar PV.
2. Cost benefit analysis:
All amounts are given in N$ million, 2007 prices

NPV PV PV B:C
Comparison Table
Total Costs Benefit Ratio
Base Case Without EE 15,385.3 14,086 29,471 2.09
No New Local Generation -13,325.1 13,362 37 0.00
Price Minimised 20,750.1 14,804 35,554 2.40
Maximum Supply Diversity 20,578.0 17,225 37,803 2.19
Generation Self-Sufficiency 18,800.0 18,183 36,983 2.03
Minimum Energy Imports 18,114.0 18,098 36,212 2.00
Renewables Maximised 20,463.3 15,859 36,322 2.29
All Options Considered With EE 23,168.8 13,697 36,866 2.69
Table 0-1: NPVs, incremental costs, incremental benefits and BCRs
The results of the economic cost benefit analysis are presented in Table 0-1. The cost
benefit analysis is a comparative exercise which is why the ‘no new local generation’
scenario has a BCR of zero - the benefits of all the other scenarios are compared to this
Page vi of xii
scenario. The scenario with the highest BCR is the ‘all options considered’ (2.69) followed by
‘price minimised’ (2.40), ‘renewables maximised’ (2.29), ‘maximum supply diversity’ (2.19)
and the ‘base case’ (2.09). The ‘minimum energy imports’ has the lowest BCR (2.00). It
should be noted that from an economic perspective this scenario is still more desirable than
the ‘no new local generation’ scenario because the BCR is greater than 1.
The following conclusions are drawn:
• The ‘all options considered’ scenario is the most desirable scenario from an economic
efficiency perspective.
• This is followed by the ‘price minimised’ generation mix, the ‘renewables maximised’
scenario and the ‘maximum supply diversity’.
• The least desirable scenarios are ‘no new local generation’ and ‘minimum energy
imports’ in that order respectively.
• The ‘all options considered’ scenario remains the most desirable scenario even when
testing sensitivity to the inclusion of energy efficiency; changes in economic growth
rates; changes in Eskom prices; increased or decreased Eskom shortages, the value
of unserved energy; and the value of carbon credits.
3. The key conclusions from an electricity price perspective are:
• The ‘no new local generation’ scenario has the lowest overall electricity price but the
greatest price volatility and the highest risk of unserved energy.
• The ‘maximum supply diversity’ and ‘maximum renewable’ scenarios have the
highest electricity prices but are less volatile than the ‘no new local generation’
scenarios.
• The ‘all options considered’ scenario results in a price level only about 5% higher
than the ‘price minimised’ and ‘base case’ scenarios.
• From a sensitivity perspective the ‘all options considered’ scenario has a significantly
lower exposure to variations in resource costs than the ‘base case’; and has a lower
sensitivity to South African import constraints than the ‘base case’.
4. As far as contribution to GDP is concerned:
• The ‘renewables maximised’ scenario, at NAD 14.0bn, makes the greatest

contribution to GDP. This is the result of the high construction costs involved with
constructing additional capacity at the Baynes hydroelectric power station (more than
in any of the other scenarios), as well as requiring significant plant size from
Concentrating Solar power stations.
• The next highest contribution is from the ‘minimum energy imports’ and the
‘generation self sufficiency’ scenario at NAD 13.1bn.
• The ‘all options considered’ scenario makes the fourth largest contribution to GDP at
NAD 12bn.
• The ‘no new local generation’ scenario, at NAD 3.0bn makes the lowest contribution
to GDP. The reason for this is the extremely low expenditure that would go into
producing electricity in Namibia.
Page vii of xii

• The ‘maximum supply diversity’, at NAD 7.5bn has the second lowest contribution to
GDP.
• Both the ‘base case’ and ‘no new local generation’ have high amounts of unserved
energy which poses a significant cost to the economy.
5. From a social perspective:
Two issues are quantifiable – job creation and promotion of small, medium and micro
enterprises (SMME).
For jobs the following conclusions are drawn:
• The number of unskilled jobs outweighs the number of skilled jobs for each
generation scenario.
• The ‘renewables maximised’ scenario at 8 553, creates the most jobs. This is
followed by the ‘minimum energy imports’, the ‘generation self-sufficiency scenarios
at 5 851 and the ‘all options considered’ scenario at 5 600.
• The main reason for the high job creation for these four scenarios is the
implementation of the Biomass (invader bush) generation option. This is a labour
intensive generation option, particularly at the unskilled level.
• The option with the least job creation potential is the ‘no new local generation’
scenario, followed by the ‘base case’ scenario. The reason that these two scenarios
create so few jobs is that they do not have the Biomass (invader bush) generation
option in their mixes and they are also more heavily dependent on imports from
Eskom and the Hwange power station in Zimbabwe.
The following conclusions are drawn on the degree of SMME promotion:
• The biomass (invader bush) generation option has the greatest potential to promote
SMME development, followed by the Baynes hydroelectric scheme and then the
Kudu gas field and the gas/solar hybrid generation options.
• Many of the generation options will not promote much SMME development.
• Scenarios ‘generation self-sufficiency’ and ‘minimum energy imports’ scenario score

the highest as far as potential SMME development is concerned. This is largely due
to the Kudu, Baynes and biomass (invader bush) generation options contribution.
• As would be expected the no new local generation scenario (mainly imports) scores
the lowest, followed by the ‘base case’.
6. From a renewable energy perspective:
• At the moment nearly one half of all Namibia’s electricity generation comes from
renewable generators in the form of hydro power. In time the ‘maximum renewable’
scenario could contribute between 90% and 100% of all of Namibia’s electricity
needs. In contrast, and as could be expected, the ‘no new local generation’ scenario
has the lowest renewable component.
Page viii of xii

• If one excludes hydro generation then the non hydro renewable part of the ‘maximum
renewable’ scenario stands at 46% in 2014 to 2017 before dropping to 25% as
Baynes comes into the generation mix.
7. As far as emissions are concerned:
• On average the ‘maximum renewables’ scenario has the lowest emissions, followed
by the ‘generation self-sufficiency’ and ‘minimum energy imports’ scenarios.
• The ‘all options considered’ scenario has the fourth highest level of emissions but is
not considerably larger than the lowest three.
• On average the ‘no new local generation’ scenario has the most emissions, followed
by the 'base case' scenario.
• The ‘maximum renewables’ generation scenario continues to have the lowest levels
of emissions.
• Considered over time the 'all options considered' scenario achieves the second best
level of emissions in the latter part of the model horizon preceded by the “maximum
renewables’” scenario.
• The ‘no new local generation’ scenario continues to have the highest level of
emission even though these do take place in South Africa and not in Namibia.
• All scenarios that include the Baynes hydro display a sharp drop in direct emissions
when that power plant comes on-line.
8. There are broader social and environmental issues that need to be taken into
account:
• When comparing the ‘base case’ and the ‘all options considered’ scenarios, it is
apparent that the ‘base case’ relies more heavily on fossil fuels both within Namibia
and outside its borders through Eskom imports. Although the Van Eck coal-fired
power station is likely to retire after 2012, until that time the power station is likely to
release high levels of air pollution into the atmosphere. Additional power could be
available in South Africa after 2012 and available for import to Namibia. The lack of
demand-side management and energy efficiency alternatives in the ‘base case’
scenario avoids the role that the public and private business alike should play in the
effective use of electricity. Without such a programme in place the general public
does not take ownership for their electricity use nor understand the relationship that
the generation of electricity has in its impacts upon the environment.
• However, there are additional environmental aspects associated with the

transmission of power that will need attention both in South Africa and Namibia.
These include:
o The transmission line losses and thus, ineffective use of natural resources
from the generation source to the point of use,
o The cost of operation and maintenance of land beneath the long-distance
transmission lines,
o The risks of fire under these lines, with the potential loss of power to Namibia,
and
o Wildlife and bird inter-actions of power lines and the loss of power or electricity
supply interruptions to Namibia.
Page ix of xii
• Due to the mix and diversity of supply options proposed under the ‘all options
considered’ scenario, the environmental and social impacts should be less than those
of a larger, centralised power station. Under this scenario, the general public will take
greater ownership of the electricity they use through the use of diverse renewable
energy options and demand side management (DSM). For Namibia, the location of
the biomass will not necessarily be where the need for electricity is on the system.
However, more localized, distributed power supply options could be considered to
avoid the high carbon footprint and cost of transporting the biomass over great
distances to a centralized power station. Also, smaller, localized systems could
support communities in the area of the station by buying fuel and/or employing people
from the surrounding areas. If planned properly, land for biomass-for-energy
production in Namibia should not have to conflict or compete with land for food
production for the country. However, this would have to be verified and is outside the
scope of this study.
9. The final set of conclusions relates to a comparison of the ‘base case’ to the ‘all
options considered’ scenario.
• The ‘all options considered’ scenario has a cumulative contribution to GDP of

NAD 12.05bn, compared to the NAD 8.98bn of the ‘base case’ although the ‘all
options considered’ scenario does reduce economic productivity because the
electricity price is higher than that of the ‘base case’.
• At 5 600 average annual jobs the ‘all options considered’ scenario creates many
more jobs than the ‘base case’ at an annual average of 2 200. This is particularly the
case for unskilled job because of the inclusion of the Biomass (invader bush)
generation option in the ‘all options considered’ scenario.
• As far as the price of electricity is concerned the ‘all options considered’ scenario has
a lower electricity price than the ‘base case’ in the early years. However, from
2011/12 onwards the electricity price in the ‘all options considered’ scenario is more
expensive than the ‘base case’.
• The ‘all options considered’ presents a lower risk profile due to less exposure to fuel
price fluctuations, greater diversity in the generation profile and higher level of
generation self-sufficiency compared to the ‘base case’.
• Due to the utilisation of smaller generation plant and concomitant shorter lead times
the ‘all options considered’ scenario results in lower levels of unserved electricity than
the ‘base case’, in particular in the short term.
THE POLICY IMPLICATIONS

This study set out with three objectives. The first was to determine the electricity outlook that
Namibia faces. The second was to identify generation and demand management options
within the context of the 1998 White Paper on Energy Policy. Here it was important to
determine the technical feasibility and economic implications of the various options. The third
was to explore the potential and role of renewable energy resources and energy efficiency
measures in these generation mixes.
Namibia faces a pending electricity crisis. If Namibia continues to rely on imported electricity
and does not invest in local generation capacity then unserved energy, power outages in
other words, could be as high as 9.7% of total demand. Peaking in 2012/13 cumulative
unserved demand could be as high as 10.8% of total demand. It has also been shown that,
Page x of xii
based on some specific assumptions that a 24 hour blackout for one day a month, the
equivalent of a 3.29% drop in electricity supply, would reduce Gross Value Added by up to
3.7%. These calculations suggest that there could be some dramatic impacts on the
Namibian economy because of the shortage of electricity.
The Energy White Paper of 1998 identifies a number of policy options including self-
sufficiency, security of supply, inclusion of renewable energy source, sustainability and cost
effectiveness and efficiency. Each of these has been explored separately in this study within
the context of eight separate scenarios. What has been found is that there is no, one, single
scenario that satisfies all of the policy criteria. Some of the options are economically efficient
but do not maximise the use of renewable resources. Others are cheap but do not guarantee
security of supply. Others, in turn, maximise the use of renewable resources but are
expensive. The consequence is that there is trade off between the various options and the
final decision needs to taken at the political level. What this report does is to inform on the
consequences of the various policy decisions.
There is however one conclusion that is unequivocal. Demand side management measures
should be introduced as a matter of urgency. They have been shown to be cost effective.
They are desirable from an environmental perspective. Probably the most important short
term advantage is they will help to alleviate the imminent power outages and in the long term
reduce the need for generation capacity. The analysis indicates that the investment needed
for demand side management breaks even at an unserved energy value between NAD2/kWh
and NAD10/kWh. Both these values are substantially less than the “low” value attached to
unserved energy in South Africa which has been assumed at R20/kWh in the 2004 NIRP.
There is a second conclusion that is almost as unequivocal although more muted.

Generation from renewable resources is desirable. It is environmentally responsible, will
create jobs, make an important contribution to the economy and ranks well from an economic
efficiency perspective. However this option is not without risks and cannot be expected to
supply all Namibia’s electricity need, at least in the short term. The “all options considered”
scenario demonstrates that, excluding hydro, it is possible to have renewable energy
producing 20% of electricity needs without escalating end consumer prices more than 7%
above the cost of importing electricity from South Africa. It also avoids the power outage
problems having unserved energy of as little as 0.4% of total demand.
From an economic perspective the least desirable option is relying on imports from South
Africa. While this option has the lowest average price it also has the highest amount of
unserved energy and therefore the worst benefit cost ratio. Probably the most important
observation that can be made is that such an option would continue to leave Namibia at the
mercy of South African politics and policy with a similar risk exposure currently experienced
in the capacity shortages in the South African electricity sector.
One policy option is that of security of supply. What has been found in this study is that it has
a higher cost, because it means more diversified supply sources. It also means making use
of both the cheapest supply sources as well as small amounts of alternate sources that have
little or no correlated risks with the mainstream supply sources. This option has one of the
worst impacts on the current account, has the second lowest benefit cost ratio and one of the
highest electricity prices, indicating that security of supply as dominant policy driver
What this study has shown, and where the research moved to as the results became evident,
is that a balanced approach to electricity generation and demand side management is
probably the most desirable option. In this study this became known as the ‘all options
considered’ scenario. This generation mix relies on a small coal plant, biomass and Baynes
hydro as the backbone of base-load electricity supply. This is complemented by an array of
diverse renewable generators, each limited in size to ensure grid stability and cost effective
Page xi of xii
grid integration as well as a balance between dispatchable (firm) and intermittent generators.
This scenario also implements demand side management. This option is the most efficient
and desirable from an economic perspective. It makes a contribution to the economy which is
little different to the top contributing scenarios. It creates a handsome number of jobs
although not as many as the ‘renewables maximised’ scenario. It has the least power
outages, low environmental emissions and average electricity price that it not that far out of
line with the cheapest options. It presents a lower risk exposure to unserved demand through
a higher level of generation diversification, less reliance on fossil fuels and more self-
sufficiency.
Expanding the generation capacity by including small scale renewable options allows
Namibia to grow generation modularly with demand growth thus reducing the size of the
projects and associated risks as well as the need to sell excess capacity in the region.
Namibia stands at a cross roads and the price of not taking action could be very high indeed.
We trust that the work that has been done in this study will assist policy makers in making
informed decisions that will benefit all Namibians.
Is has become clear that an integrated resource plan for electricity in Namibia is needed.
This needs to be developed under the leadership of the policy maker with full participation of
all major stakeholders to ensure that the end result is a balanced plan drawing on the
wisdom of diverse specialists and viewpoints.
Page xii of xii

Acknowledgements
The Team would like to acknowledge the following persons for their contributions to the
study:
Petja Dobreva, for drafting the sections on nuclear power plants, concentrating solar plants,
integrated solar combined cycle plants and biomass to electricity plants.
Robin Sherbourne, for peer reviewing the draft report and being a sounding board on
economic issues in Namibia.
Ralf Tobich, for peer reviewing the draft report.
Penda Hangala, for reviewing the Vision 2030 document.
The Team would like to thank Dr Detlof von Oertzen and Harald Schütt for providing
extensive comments on the draft report.
Page 1 of 222
Authors
This report was prepared by:
Emcon Consulting Group: technical analysis
Uli von Seydlitz
Axel Scholle
Economics Information Services: economic analysis
Barry Standish
Antony Boting
Geo cc, environmental analysis
Catherine Fedorsky
Page 2 of 222
Disclaimer
While every effort is made to ensure the accuracy of analysis performed, no responsibility is
taken for errors or financial losses or any other incidental or consequential damages arising
as a result of its use.
Page 3 of 222
Table of Contents
Executive Summary................................................................................................................ i
Key findings......................................................................................................................... i
Introduction to the Executive Summary .............................................................................. ii
The Challenges: .................................................................................................................iii
The Approach.....................................................................................................................iii
Analytical Results .............................................................................................................. vi
The Policy Implications....................................................................................................... x
Acknowledgements................................................................................................................ 1
Authors .................................................................................................................................. 2
Disclaimer.............................................................................................................................. 3
Table of Contents .................................................................................................................. 4
List of Tables ......................................................................................................................... 9
List of Figures ...................................................................................................................... 11
Abbreviations ....................................................................................................................... 15
1 Introduction................................................................................................................... 17
2 Current Situation and Potential Outlook ........................................................................ 20
2.1 Electricity Usage.................................................................................................... 20
2.2 Electricity Supply ................................................................................................... 23

2.2.1 Generation...................................................................................................... 23
2.2.2 Transmission and Distribution......................................................................... 25
2.2.3 Electricity Tariffs ............................................................................................. 25
2.3 Demand and Supply .............................................................................................. 26
2.4 Environmental Issues ............................................................................................ 28
2.5 Potential Outlook ................................................................................................... 29

2.5.1 Cost................................................................................................................ 29
2.5.2 Supply options ................................................................................................ 31
Page 4 of 222
2.5.3 Environmental considerations......................................................................... 32
3 Description of Methodology .......................................................................................... 36
3.1 Technical Methodology.......................................................................................... 36

3.1.1 Demand Forecast ........................................................................................... 36
3.1.2 Generation and Demand Management Options.............................................. 39
3.1.3 Dispatch Modelling ......................................................................................... 53
3.2 Description of Generation Scenarios ..................................................................... 60

3.2.1 Scenario 1: Base Case ................................................................................... 61
3.2.2 Scenario 2: No New Local Generation............................................................ 62
3.2.3 Scenario 3: Price minimised ........................................................................... 63
3.2.4 Scenario 4: Maximum Supply Diversity........................................................... 64
3.2.5 Scenario 5: Generation Self Sufficiency.......................................................... 65
3.2.6 Scenario 6: Energy Imports Minimised............................................................ 66
3.2.7 Scenario 7: Maximum Renewable Energy Options ......................................... 67
3.2.8 Scenario 8: ‘All options considered’ ................................................................ 68
3.3 Economic Methodology ......................................................................................... 69

3.3.1 Cost Benefit Analysis...................................................................................... 69
3.3.2 Macro-Economic Analysis............................................................................... 70
4 Analytical Results, Sensitivity Analysis and Interpretation............................................. 77
4.1 Analytical Results .................................................................................................. 77

4.1.1 Economic Cost Benefit Analysis ..................................................................... 77
4.1.2 Electricity Price Projections and Sensitivities .................................................. 83
4.1.3 Economic Impacts .......................................................................................... 90
4.1.4 Social Impacts ................................................................................................ 93
4.1.5 Supply Interruptions and Unserved Demand................................................... 95
4.1.6 Contribution of Renewable Resources to Electricity Generation ..................... 99
4.1.7 Grid Cost: Centralised vs. Decentralised ...................................................... 101
4.1.8 Emissions ..................................................................................................... 101
4.2 Comparison of the ‘base case’ and the ‘all options considered’ scenarios ........... 105
4.2.1 Contribution to GDP...................................................................................... 105
4.2.2 Job Creation ................................................................................................. 107
4.2.3 Balance on Current Account ......................................................................... 109
4.2.4 Electricity Price ............................................................................................. 111
4.2.5 Macro Economic Sensitivity Analysis ............................................................ 111
4.2.6 Environmental assessment........................................................................... 118
Page 5 of 222
4.3 Summary and Discussion .................................................................................... 121

4.3.1 Summary of results....................................................................................... 122
4.3.2 Comparative Discussion ............................................................................... 124
5 Conclusions ................................................................................................................ 126
References ........................................................................................................................ 128
Appendix A. Terms of Reference .................................................................................. 135
Appendix B. Discussion of the existing Regulatory and Power Supply Infrastructure .... 142
B.1. Regulatory Environment ...................................................................................... 142
B.2. Roles and Responsibilities of ESI Institutions ...................................................... 142

B.2.1. Policy Level .................................................................................................. 142
B.2.2. Regulatory Level........................................................................................... 142
B.2.3. Licensees ..................................................................................................... 143
B.3. Power Supply Infrastructure................................................................................. 143

B.3.1. Generation.................................................................................................... 144
B.3.2. Transmission and Distribution Network......................................................... 146
Appendix C. Detailed discussion of Generation Options................................................ 148
C.1. Technical issues .................................................................................................. 149
C.2. Environmental aspects ........................................................................................ 150
C.3. Cost issues.......................................................................................................... 155
C.4. Hydro-Electric Power Generation......................................................................... 155

C.4.1. Technical Issues........................................................................................... 155
C.4.2. Generation Potential ..................................................................................... 156
C.4.3. Environmental and social aspects................................................................. 156
C.4.4. Financial Overview ....................................................................................... 160
C.4.5. Risks ............................................................................................................ 160
C.5. Coal-Fired Power Generation .............................................................................. 162

C.5.2. Resource ...................................................................................................... 163
C.5.3. Generation potential ..................................................................................... 164
C.5.4. Environmental and Social Aspects................................................................ 164
C.5.6. Risks ............................................................................................................ 171
C.6. Natural Gas Power Generation ............................................................................ 172
Page 6 of 222

C.6.2. Resource & Generation Potential.................................................................. 173
C.6.5. Risks ............................................................................................................ 177
C.7. Nuclear Power Generation................................................................................... 178

C.7.1. Technical issues ........................................................................................... 178
C.7.2. Resource ...................................................................................................... 178
C.7.6. Risks ............................................................................................................ 183
C.8. Electricity Generation from Biomass .................................................................... 185

C.8.2. Resource ...................................................................................................... 186
C.8.6. Risks ............................................................................................................ 190
C.9. Concentrating Solar Thermal Power Plants ......................................................... 192

C.9.1. Technology Status........................................................................................ 192
C.9.2. Resource ...................................................................................................... 193
C.9.6. Risks ............................................................................................................ 198
C.10. Solar Photovoltaic Power Plants ...................................................................... 199

C.10.2. Resource ...................................................................................................... 199
C.10.6. Risks ............................................................................................................ 204
C.11. Wind Energy Power Generation ....................................................................... 205

C.11.2. Resource ...................................................................................................... 205
Page 7 of 222

C.11.6. Risks ............................................................................................................ 209
C.12. Integrated Solar Combined Cycle Plant............................................................ 210

C.12.3. Risks ............................................................................................................ 210
Appendix D. Detailed discussion of Demand Side Management Options ...................... 212
D.1. Compact Fluorescent Lamp Roll Out ................................................................... 214

D.1.1. Anticipated Impact ........................................................................................ 214
D.1.2. Costs and Investment ................................................................................... 214
D.1.3. Environmental and Social Impacts................................................................ 214
D.1.4. Risks ............................................................................................................ 215
D.1.5. Summary ...................................................................................................... 216
D.2. Solar Water Heater Roll Out ................................................................................ 217

D.2.3. Environmental and Social Impacts................................................................ 217
D.2.4. Risks ............................................................................................................ 218
D.2.5. Summary ...................................................................................................... 219
D.3. Ripple Control...................................................................................................... 220

D.3.3. Risks ............................................................................................................ 221
D.3.4. Summary ...................................................................................................... 222
Page 8 of 222
List of Tables
Table 0-1: NPVs, incremental costs, incremental benefits and BCRs ................................... vi
Table 2-1: The carbon intensity of a number of countries..................................................... 33
Table 2-2: Externality costs for a range of supply options .................................................... 35
Table 3-1: Estimated Generator Capacities and Step Sizes................................................ 49
Table 3-2: Technical Model Key Characteristics .................................................................. 56
Table 3-3: Technical Model Key Assumptions...................................................................... 58
Table 3-4: Technical Model Limitations................................................................................ 59
Table 3-5: Scenario Definitions ............................................................................................ 60
Table 3-6: Comparison of GDP multipliers ........................................................................... 72
Table 3-7: Growth rates, elasticities of demand and production effects from electricity
outages per economic sector ........................................................................................ 76
Table 4-1: NPVs, incremental costs, incremental benefits and BCRs .................................. 78
Table 4-2: Demand Side Management Sensitivity................................................................ 79
Table 4-3: Economic Growth Sensitivity............................................................................... 80
Table 4-4: Commodity Price Sensitivity................................................................................ 80
Table 4-5: Eskom Price Sensitivity....................................................................................... 81
Table 4-6: Eskom Shortages Sensitivity............................................................................... 82
Table 4-7: Cost of Unserved Demand.................................................................................. 82
Table 4-8: Value of Carbon Credits...................................................................................... 83
Table 4-9: SMME contribution of each generation scenario ................................................. 95
Table 4-10: Effect of supply interruptions on GVA................................................................ 97
Table 4-11: Contribution of the ‘base case’ to Gross Domestic Product............................. 105
Table 4-12: Contribution of ‘all options considered’ to Gross Domestic Product................. 106
Table 4-13: ‘Base case’ Contribution to Direct Jobs ........................................................... 107
Table 4-14: ‘Base case’ Contribution to Indirect Jobs......................................................... 107
Table 4-15: ‘Base case’ Contribution to Total Jobs ............................................................ 108
Page 9 of 222
Table 4-16: ‘All options considered’ Contribution to Direct Jobs......................................... 108
Table 4-17: ‘All options considered’ Contribution to Indirect Jobs ...................................... 108
Table 4-18: ‘All options considered’ Contribution to Total Jobs .......................................... 109
Table 4-19: Scenario evaluation matrix.............................................................................. 125
Table C-1: Potential Environmental and Social Impacts of Energy supply Options ............ 154
Table C-2: Social Impact Checklist – Baynes Hydro .......................................................... 160
Table C-3: Environmental challenges at coal fired power stations...................................... 167
Table C-4: Social Impact Checklist – Coal Station at Walvis Bay....................................... 168
Table C-5: Social Impact Checklist – Gas Power Station in Oranjemund Area................... 176
Table C-6: Average Emission Levels ................................................................................. 180
Table C-7: Social Impact Checklist – Nuclear Power Station at Walvis Bay ....................... 182
Table C-8: Social Impact Checklist – Biomass to Electricity ............................................... 190
Table C-9: Concentrating Solar Power Technical Potential ................................................ 195
Table C-10: Social Impact Checklist – Solar Concentrating Plant in Namib Desert ............ 197
Table C-11: Social Impact Checklist – Solar PV Plant in Namib Desert ............................. 203
Table C-12: Social Impact Checklist – Wind energy plants near Lüderitz and Walvis Bay.. 208
Table D-13: Summary information on CFL measure .......................................................... 216
Table D-14: CO2 Emissions Reduced and Avoided Kuyasa CDM Energy Upgrade Project
................................................................................................................................... 218
Table D-15: Summary information on SWH measure......................................................... 219
Table D-16: Summary information on ripple control measure............................................. 222
Page 10 of 222
List of Figures
Figure 2-1: Electricity Revenue by Sector (NAD millions) in 2002 ........................................ 20
Figure 2-2: Electricity usage as a percentage of turnover by user type ................................ 21
Figure 2-3: Contribution to GVA and electricity usage per economic sector in 2002 ............ 22
Figure 2-4: 2005 Load Curve ............................................................................................... 22
Figure 2-5: Electricity Demand and Supply Trend by Source ............................................... 23
Figure 2-6: The Namibian Transmission Grid and Power Stations ....................................... 24
Figure 2-7: Namibia Installed Capacity and Peak Demand .................................................. 26
Figure 2-8: Namibia Demand and Supply Forecast.............................................................. 27
Figure 2-9: South African Demand and Capacity ................................................................. 27
Figure 2-10: NamPower Average Electricity Procurement Cost ........................................... 30
Figure 2-11: South African Generation Cost Forecast.......................................................... 30
Figure 2-12: Regional Electricity Prices ............................................................................... 31
Figure 3-1: Dispatch Model Overview .................................................................................. 36
Figure 3-2: Average weekday demand profile for winter and summer.................................. 37
Figure 3-3: NamPower Demand Forecast............................................................................ 37
Figure 3-4: Current Demand Forecast – System Peak......................................................... 38
Figure 3-5: Full estimated capital cost for generating options NAD million per MW.............. 48
Figure 3-6: Estimated base tariffs for generating options ..................................................... 48
Figure 3-7: Timeline of generators that could be built........................................................... 49
Figure 3-8: ‘Base case’ generation mix ................................................................................ 61
Figure 3-9: ‘No new local generation’ generation mix........................................................... 62
Figure 3-10: ‘Price minimised’ generation mix...................................................................... 63
Figure 3-11: ‘Maximum supply diversity’ generation mix ...................................................... 64
Figure 3-12: ‘Generation self sufficiency’ generation mix ..................................................... 65
Figure 3-13: ‘Energy imports minimised’ generation mix ...................................................... 66
Figure 3-14: ‘Maximum renewable’ generation mix .............................................................. 67
Page 11 of 222
Figure 3-15: ‘All options considered’ generation mix ............................................................ 68
Figure 4-1: Projected Average Electricity Price .................................................................... 84
Figure 4-2: Projected Electricity Price Development............................................................. 85
Figure 4-3: Average Electricity Price Sensitivity to Economic Growth .................................. 86
Figure 4-4: Electricity Price Sensitivity to Economic Growth................................................. 86
Figure 4-5: Average Electricity Price Sensitivity to Resource ............................................... 87
Figure 4-6: Electricity Price Sensitivity to Resource Costs ................................................... 88
Figure 4-7: Average Price Sensitivity to Import Constraints.................................................. 89
Figure 4-8: Electricity Price Sensitivity to Import Constraints................................................ 89
Figure 4-9: Projected Peak Demand at an economic growth rate of 3.7%............................ 90
Figure 4-10: Projected Electricity Demand at an economic growth rate of 3.7%................... 91
Figure 4-11: Projected Demand for Electricity at an economic growth rate of 2% and 6%. .. 91
Figure 4-12: Projected System Peak Demand at economic growth of 2% and 6%............... 92
Figure 4-13: Cumulative impact on GDP under different scenarios, NAD billion (2007 values).
..................................................................................................................................... 93
Figure 4-14: Total jobs within each scenario. ....................................................................... 94
Figure 4-15: Electricity Demand Profile................................................................................ 96
Figure 4-16: Potential Cumulative Unserved Electricity Demand ......................................... 98
Figure 4-17: Potential Unserved Electricity Demand ............................................................ 99
Figure 4-18: Projected Renewable (including hydro) Generation Contribution ................... 100
Figure 4-19: Projected Renewable Generation Contribution excluding hydro..................... 100
Figure 4-20: Cumulative environmental emissions............................................................. 102
Figure 4-21: Projected CO2 Emissions............................................................................... 103
Figure 4-22: Projected NOx Emissions .............................................................................. 104
Figure 4-23: Projected SOx Emissions .............................................................................. 104
Figure 4-24: Comparative Cumulative Contribution to GDP ............................................... 106
Figure 4-25: Comparison between Average Total Job Creation by Skill Level ................... 109
Figure 4-26: Balance of Current Account for the ‘base case’ and the ‘all options considered’
Scenario ..................................................................................................................... 110
Page 12 of 222
Figure 4-27: Electricity Price Comparison .......................................................................... 111
Figure 4-28: Cumulative Contribution to GDP .................................................................... 112
Figure 4-29: Average Total Job Creation ........................................................................... 113
Figure 4-30: Cumulative Contribution to GDP for Different Economic Growth Rates.......... 113
Figure 4-31: Average Total Job Creation for Different Economic Growth Rates ................. 114
Figure 4-32: Cumulative Contribution to GDP for Different Commodity Prices ................... 115
Figure 4-33: Average Total Job Creation for Different Commodity Prices .......................... 115
Figure 4-34: Cumulative Contribution to GDP for Different Eskom Prices .......................... 116
Figure 4-35: Average Total Job Creation for Different Eskom Prices ................................. 116
Figure 4-36: Cumulative Contribution to GDP for Different Economic Growth Rates.......... 117
Figure 4-37: Average Total Job Creation for Different Economic Growth Rates ................. 118
Figure B-1: The Namibian ESI ........................................................................................... 143
Figure B-2: The Namibian Electricity Network Map ............................................................ 147
Figure C-3: Risk chart for Baynes hydro power.................................................................. 161
Figure C-4: World Primary Energy Sources ....................................................................... 162
Figure C-5: Relative operating costs of systems at coal fired power stations ..................... 169
Figure C-6: Coal plant investment versus capacity ............................................................ 170
Figure C-7: Risk chart for clean coal power plant............................................................... 171
Figure C-8: CCGT Schematic Layout................................................................................. 172
Figure C-9: Gas Plant Fuel Cost Trend.............................................................................. 173
Figure C-10: Gas Plant Operating Cost Trend ................................................................... 174
Figure C-11: Risk chart for Gas CCGT power plant ........................................................... 177
Figure C-12: Risk chart for nuclear PBMR power plant ...................................................... 184
Figure C-13: Overview of biomass resources in Namibia................................................... 186
Figure C-14: Risk chart for biomass to electricity power plant ............................................ 191
Figure C-15: Map of average daily irradiation levels for Namibia ....................................... 194
Figure C-16: Risk chart for concentrating solar power plant ............................................... 198
Figure C-17: Seasonal variation of the solar resource........................................................ 200
Page 13 of 222
Figure C-18: Average daily power generation profile of a 55MW peak plant .......................... 201
Figure C-19: Risk chart for solar PV power plant ............................................................... 204
Figure C-20: Risk chart for wind generation power plant.................................................... 209
Figure C-21: Risk chart for integrated solar combined cycle power plants ......................... 211
Figure D-22: Risk chart for CFL energy efficiency measure ............................................... 215
Figure D-23: Risk chart for conversion to SWH.................................................................. 219
Figure D-24: Risk chart for ripple control DSM measure .................................................... 221
Page 14 of 222
Abbreviations
BCR Benefit cost ratio
CBA Cost benefit analysis
CBEND Combating bush encroachment for Namibia’s development
CCGT Combined cycle gas turbine
CDM Clean development mechanism
CFL Compact fluorescent lamp
CSP Concentrating solar (thermal) power
DSM Demand side management
ECB Electricity Control Board
EA Environmental assessment
EE Energy efficiency
EIA Environmental impact assessment
EMP Environmental management plan
ESI Electricity supply industry
EWH Electric water heater
GDP Gross domestic product
GVA Gross value added, the same as Gross domestic product
GWh Giga watt hour
IDC Interest during construction
ISCC Integrated solar combined cycle
kWh kilo watt hour
LRMC Long run marginal cost
MET Ministry of Environment and Tourism
MME Ministry of Mines and Energy
MW Mega watt
Page 15 of 222
MWh Mega watt hour
NER National Electricity Regulator (South Africa)
NPP Nuclear power plants
OECD Organisation for economic co-operation and development
PBMR Pebble bed modular reactor
PV Photovoltaic
RE Renewable energy
RED Regional electricity distributors
Renewable energy and energy efficiency capacity building

REEECAP
programme
RSA Republic of South Africa
SAM Social accounting matrix
SAPP Southern African Power Pool
SMME Small medium and micro enterprises
SWH Solar water heater
USDOE United States Department of Energy
WACC Weighted average cost of capital
WEC Wind energy converter
Page 16 of 222
1 Introduction
The Namibian electricity industry stands at a critical cross road. Internationally high oil and
commodity prices have had a ripple effect on the price of gas and coal. Energy prices are
escalating at alarming rates. Global warming and a growing awareness of the damage
generated by fossil fuels is seeing increasing pressure on countries to turn to other ways of
generating electricity. Within Namibia there are growing misgivings about the Van Eck coal
fired power station on the outskirts of Windhoek. And Eskom, the South African power utility
that supplies half of Namibia’s electricity, has not been able to keep pace with electricity
demands in South Africa and is being forced to reduce power to parts of South Africa and
neighbouring countries. Compounding this is the historic availability of cheap electricity from
Eskom which has been an inhibiting factor to the construction of any new power stations in
Namibia and to this day dominates the price expectations in Namibia.
At the same time Namibia has very limited options with using the traditional generation
technologies of coal fired, hydro or nuclear power. Namibia does not have any currently
exploited coal reserves and must import coal. It is simply cheaper to import electricity from
South Africa than import South African coal. It is a very dry country which limits hydro.
Nuclear is an option but is constrained by high capital costs and a relatively small skills base.
Yet there are opportunities. Namibia is one of the sunniest countries in the world. There are
feasible wind resources at the coast. There is a major offshore gas field. In the north of the
country there are thousands of hectares of alien vegetation that desperately need clearing
and are an obvious fuel source. There is the Kunene River that holds promise for additional
hydro power.
This study was sponsored by the Renewable energy and energy efficiency capacity building
programme (REEECAP) with the intention of exploring the technical, economic, social and
environmental implications of existing supply generation, on the one hand, and other supply
and demand options, on the other. The intention is to inform policy makers about the range
of options, the costs of these options and the degree to which they can contribute to the
future economic and social well being of Namibia.
within the context of the 1998 White Paper on Energy Policy (MME 1998). Here it was
important to determine the technical feasibility and economic implications of the various
options. The third was to explore the potential and role of renewable resources in these
generation mixes.
This study therefore aims to inform the policy interpretation and application process in
Namibia by illustrating what the various policy drivers might mean in practice, both at a
technical/financial level but also in terms of the economy, and what their relative merits might
be. To our knowledge this is the first comprehensive study for Namibia that considers
alternative energy sources and fossil fuel and hydro electricity generation options. It also
explores the financial and economic implications of various possible supply mixes.
It will be shown that:
• Namibia faces a pending electricity crisis. If Namibia continues to rely on imported

electricity and does not invest in local generation capacity then unserved energy,
power outages in other words, could be as high as 9.7% of total demand. Peaking in
Page 17 of 222
2012/13 cumulative unserved demand could be as high as 10.8% of total demand. It

has also been shown that, based on some specific assumptions that a 24 hour
blackout for one day a month, the equivalent of a 3.29% drop in electricity supply,
would reduce Gross Value Added by up to 3.7%. These calculations suggest that
there could be some dramatic impacts on the Namibian economy because of the
shortage of electricity.
• There is no, one, single policy option that satisfies all of the policy criteria the White
Paper on Energy Policy identifies. Some of the options are economically efficient but
do not maximise the use of renewable resources. Others are cheap but do not
guarantee security of supply. Others, in turn, maximise the use of renewable
resources but are expensive. The consequence is that there is trade off between the
various options and the final decision needs to taken at the political level.
• There is however one conclusion that is unequivocal. Demand side management

measures should be introduced as a matter of urgency. They have been shown to be
cost effective. They are desirable from an environmental perspective. Probably the
most important short term advantage is that they will help to alleviate power outages
and in the long term reduce the need for generation capacity.
• There is a second conclusion that is almost as unequivocal although more muted.

Generation from renewable resources is desirable. It is environmentally responsible,
will create jobs, makes an important contribution to the economy and ranks well from
an economic efficiency perspective. However this option is not without risks and
cannot be expected to supply all Namibia’s electricity need, at least in the short term.
• From an economic perspective the least desirable option is relying solely on imports
from South Africa. While this option has the lowest average price it also has the
highest amount of unserved energy and therefore the worst benefit cost ratio.
Probably the most important observation is that such an option would continue to
leave Namibia at the mercy of South African politics and policy.
• A balanced approach to electricity generation and demand side management is the

most desirable option. In this study this became known as the ‘all options considered’
scenario. This generation mix relies on a small coal plant, biomass and Baynes hydro
as the backbone of hour electricity supply. This is complemented by an array of
diverse renewable generators, each limited in size to ensure grid stability and cost
effective grid integration by avoiding grid issues that would arise for larger versions of
the same plant type. This scenario also implements demand side management. This
option is the most efficient and desirable from an economic perspective. It makes a
contribution to the economy which is little different to the top contributing scenarios. It
creates a handsome number of jobs although not as many as the ‘renewables
maximised’ scenario. It has the least power outages, low environmental emissions
and an average electricity price that is not far out of line with the cheapest options.
This report has five sections:
• Section 1 introduces the report.
• Section 2 describes the current situation and potential outlook for the generation
industry and the implications of this for Namibia.
• Section 3 describes the methodology that was used in this study.
• Section 4 reports on the result of the technical and economic analysis.
Page 18 of 222
• Finally Section 5 reports on the main conclusions to the study.
Four appendices contain the following:
• An overview of the terms of reference for the study
• An overview of the electricity supply industry in Namibia
• Details of the new generation options considered in this study
• Details of the demand side management options considered in this study
Page 19 of 222
2 Current Situation and Potential Outlook

This section describes the current electricity situation in Namibia. It also paints some
potential future developments given existing knowledge and known policy decisions of the
country.
2.1 ELECTRICITY USAGE

Namibia has a surface area of almost 825 000 square kilometres and a population of around
2 million, giving it one of the lowest population densities in the world (2.5 people/square km).
Two thirds of the population live in rural areas with limited access to infrastructure,
specifically to grid electricity. According to the 2001 census only 32% of households have
access to grid electricity for lighting.
A description of electricity demand by the Namibian economy is shown graphically in Figure

2-1, Figure 2-2 and Figure 2-3.
Electricity Revenue by Sector
Government Services
$190
Other
$311
Mining
$100
Electricity Urban Households - Wage &

$40 Transport Salary
$49 $91
Figure 2-1: Electricity Revenue by Sector (NAD millions) in 20021
Figure 2-1 indicates that in 2002 government services, at NAD 190m, was the largest user of
electricity. This was followed by the mining sector (NAD 100m) and then urban households
(NAD 91m). These three users together made up nearly 50% of all electricity revenue in
2002. It might seem counterintuitive from Figure 2-1 that the mining sector uses less
electricity than government services. One of the reasons for this is that mines pay a lower
1
The charts are based on an analysis of the Namibian SAM
Page 20 of 222
tariff than other users (i.e. there are no distribution costs or local authority surcharge added
to the price they pay). The electricity sector in the figure refers to the electricity generating
sector, which in itself is a consumer of electricity. The urban households – wage and salary
label refers to those urban households where families earning salaries live. Households in
the social accounting Matrix (SAM) have been categorised into rural and urban households,
and then further subdivided according to their main source of income.
Figure 2-2 illustrates electricity usage as a percentage of turnover, i.e. the intensity of
electricity usage. The sector with electricity costs as the highest input proportion relative to
turnover was the electricity sector itself, followed by the water sector and then government
services. Hotels and restaurants are the fourth most intensive users of electricity.
Electricity Usage as % of Turnover per Sector

6.0%
5.1%
5.0%
Electricity Usage as % of Input Costs
4.0%
3.0%
2.3%
2.2%
2.0%
2.0%
1.3% 1.3% 1.2% 1.2% 1.2% 1.1%

1.0%
0.0%
Electricity Water Government Hotels and Commercial Transport Urban Mining Urban Commercial
Services Restaurants Animal Households - Households - Other Crops
Products Farm & Other
Business
Figure 2-2: Electricity usage as a percentage of turnover by user type
Figure 2-3 combines electricity usage and contribution to Gross Value Added (GVA)2. Here it
can be seen that although the government services sector is the highest user of electricity it
also contributes the most to GVA. Sectors such as the Fishing sector (not Fish Processing)
and Real Estate are negligible users of electricity yet are among the larger contributors to
GVA. In contrast, the Transport, Hotel & Restaurants, Heavy Manufacturing and Electricity
sectors are intensive users of electricity compared to their relative contributions to GVA.
2
Gross Value Added is the equivalent of Gross Domestic Product measured at factor cost rather than market prices.
Page 21 of 222
Contribution to GVA and Electricity Consumption

7,000 200
Electricity Consumption (N$ millions)

180
6,000
160
5,000 140
GVA (N$ millions)
Gross Value Added

Electricity Consumption
120
4,000
100
3,000
80
2,000 60
40
1,000
20
0 0
Pr sta ng
w rain icity
su ng
g
Fi An ate cts
an ro n
ot y M ran e
sh nic s
ts
s
ne Fis s
er g
g
O and fac rt
om er s
g
In ucts
R ops
r
s
te
m Re d P ice
m vice
ce
nt
Pu Co erc Pr uct
in
P w
ile
er ss S hin
T nc
in
on rin
io
io
th sin
ro ere rop
gh Pr atio
nu po
en
ai
er Re turi
on illi
a
u
-O
ad Min
at ura
ct
pt
an ess
um ext
vi
a
ep
C ctu
by Cr
r O G ctr
s
v
W
ci l E rod
ha me l O es
m at rod
id
d
Ab l C r C
ru
m
r
er
su
es
R
le
T
t M oc
c
st
al
a
S
tS
e P
e
o
u
uf
E
e,
e
nc al
st
en
d
o
a
na im
C
Fo
iv
nm
le
Tr
ad
n
& usi
C
a
Fi
tro
a
i
M
e
er
se rci
v
B
ea
Pe
al
s
th
ov
Li
el
th
ge &
Fo
G
om
er
e
s
H
ve tat
od
C
s
om
Be Es
Fo
ra
rc
l
ea
,R
ct
g
ire
tin
D
ke
ar
M
Figure 2-3: Contribution to GVA and electricity usage per economic sector in 2002
The final aspect of interest on the demand side is the load curve i.e. the varying demand for
electricity over the course of a day. Figure 2-4 (ECB 2006) shows the resulting set of load
curves for 2005. The 2005 load curve is characterised by a strong evening peak, a day-time
plateau and a night-time trough. The evening peak is largely attributed to household demand
stemming from cooking, electric water heaters, lighting and a variety of other energy usages.
2005 Load Curve (Categorised)
400
375
350
325
MW
300
275
250
225
200
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Hour
Low High Max
Figure 2-4: 2005 Load Curve
Page 22 of 222
2.2 ELECTRICITY SUPPLY

This section gives a brief description of the Namibian electricity supply, transmission and
distribution infrastructure. A more detailed presentation is given in Appendix B.
Namibia produces power locally and imports about half its electricity, with the bulk of this
being supplied by Eskom in South Africa. NamPower imports on a smaller scale from Zambia
for supply to the Caprivi Region and exports on a small scale to Angola, Botswana and South
Africa. The only current trading interconnection is to South Africa, with the Caprivi
transmission link being under construction3. This composition and the trends in this are
illustrated in Figure 2-5.
Namibia Grid Demand for Electricity

(excluding Skorpion Zinc Project)
3000
MWh per annum
2500
2000
Units sold in Namibia
1500 Supplied by NamPow er
Generation
1000 Supplied by Eskom
Supplied by Other
500
0
89
91
93
95
97
99
01
03
05
07
19
19
19
19
19
19
20
20
20
Financial Year ending June 20
Figure 2-5: Electricity Demand and Supply Trend by Source4
2.2.1 Generation
There are currently three major power stations in Namibia which are owned and operated by
NamPower, i.e. the Ruacana hydro-electric power station on the Kunene River, the Van Eck
coal-fired power station at Windhoek, and the Paratus diesel power station at Walvis Bay.
The location of these is shown in Figure 2-6 (NamPower 2006a).
3
Trading links are different from import and export links in that they support two way trade at high volumes. The small scale
import (such as to Caprivi from Zambia) and export (such as to Botswana, Angola and South Africa) links support one way
energy flow only and are not considered trading links. Trading links are normally at transmission voltages whereas export/import
links are at sub-transmission or distribution level.
4
Compiled from NamPower annual reports from 1993 to 2007. The Skorpion Zinc Project is supplied from Eskom with
NamPower wheeling the power across its transmission network. For this reason Skorpion is excluded from the normal statistics
for the NamPower interconnected system.
Page 23 of 222
Epupa/ Baynes
Ruacana Kati ma Muli lo
Omatando
Okatop e Rundu
Li vi ngstone
Divundu /
(ZAMBI A)
SVC P opa Fall s
Fo r Otji kot o
Otji koto Fo r Im po rt s/O tjik ot o
Gerus F or Ot jiko to
Omb uru Su bst at ion

We st o f Whk Ghanz i
V an Eck Omaere
Walmu nd Au as
Kuiseb
Paratus Rehoboth F o r Re ho bo t h/Ot jih ase
Hardap
Legend
Leg end
400 kV
400 kV
330 kV
Fo r Namde b SVC 330 kV
220 kV
220 kV
Kokerboom 132 kV
132 kV
66 kV
Nam ib 66 kV
Gen eration
Generati on
Exi sti ng
Existing
F or Ku d u Obi b Gen eration
Generati on
Rock Po tential
Poten ti al
Ku du Harib
Orange River Mi ni Hydros
Figure 2-6: The Namibian Transmission Grid and Power Stations
The Ruacana hydro-electric power station is located in the north of Namibia, where the
Kunene River becomes the border between Namibia and Angola. It has a rating of 240/249
MW. The station has black start up diesel generators and a 330 kV transmission line from
Ruacana to the Omburu transmission station which is 570 km in length. The station was
commissioned in 1978. The civil works for the addition of a fourth turbine are completed and
it was recently announced that this turbine would become operational. The station is mainly
operated as a run-of-river power plant (i.e. its output depends on the amount of water
available in the river), because upstream storage dams are either not completed or were
damaged in the Angolan civil war. A small diversion weir just upstream of Ruacana allows
the power station to provide full output for 8 hours (NamPower, pers. comm.., 22 Oct 2007)
when filled during times of low water flow. During the raining season (February to May) the
station is run on full load and operated as a base load power plant, while for the remainder of
the year it is operated as a run-of-river plant generating predominantly during peak demand
hours.
The Van Eck coal-fired power station is situated on the outskirts of Windhoek. It has a rating
of 120 MW using four 30MW generators and was commissioned in 1973. The station needs
external power to start generation. It is a dry cooled power station because of water
constraints in Windhoek. Coal is imported from South Africa, transported by ship to Walvis
Bay and then by rail or road to Windhoek. This is a costly exercise and the plant is normally
operated as a standby and peaking power station. During the more recent regional
Page 24 of 222
constraints it has been run mid-merit to base load5. The power station has very limited
emissions control equipment installed and thus emits high levels of air pollutants.
The Paratus diesel power station is located in Walvis Bay. It has a rating of 24 MW using four
6MW diesel generators. The station has a black start up diesel generator and was
commissioned in 1976. It is used mainly as a standby and peaking power station
respectively, but it is also contractually bound as an emergency standby plant for the city of
Walvis Bay. Paratus is in good workable condition, although it runs at very high short run
marginal cost of just under NAD 80c/kWh (NamPower, pers. comm., October 2007)
2.2.2 Transmission and Distribution

As a result of the fact that the existing domestic power stations are located so far apart, the
transmission system is a critical component in delivering power. This is compounded by the
great distances of in-coming supply coming from South Africa along high voltage power lines.
Transmission is divided into two businesses within NamPower - the wires business and the
supply business. The wires business manages and operates the transmission network, while
the supply business looks after the transmission customers. These transmission customers
include some large mines, the Regional electricity distributors (REDs), plus some smaller
supply points that are connected directly to the transmission grid largely for historic reasons.
The REDs are responsible for the distribution and supply of electricity to end consumers
within their respective areas. Three REDs have already been established and are operational
(NORED, CENORED and Erongo RED), while Central RED and Southern RED are delayed
indefinitely for various (largely political) reasons.
2.2.3 Electricity Tariffs

Electricity tariffs in Namibia are regulated by the Electricity Control Board (ECB). The end
user price is the sum of generation cost, transmission cost, distribution costs and a Local
Authority surcharge. The customers supplied directly by NamPower Transmission do not
have distribution costs and a Local Authority surcharge included in their price. The regulatory
approach to tariff determination is based on cost reflectivity plus a regulated return on assets.
The cost of demand side management initiatives undertaken and paid for by NamPower is
usually recovered through the tariff but is not accounted for separately.
In terms of tariff structure NamPower has very recently introduced time of use (TOU) tariffs to
its transmission customers, but not yet to distributors. The distributors typically offer credit
metered tariffs to commercial consumers, and credit metered or prepaid tariffs to residential
consumers.
New generators feeding into the NamPower transmission grid will compete price-wise mainly
with the Eskom imports, which are by far the cheapest current source of power except for
5
A mid merit power plant is typically operated 24 hours a day, but at varying output to follow the demand on the system. A
peaking plant only runs during peak times and is not utilised during off-peak times. A base load plant runs at full output all the
time. Ruacana is run as base load when there is enough water available in the Kunene river. As water flow reduces after the
rainy season the plant can no longer run at full output for 24 hours a day. It then follows the load during the day while water is
dammed up in the diversion weir during low demand times (typically at night). As water flow drops even further the water is
dammed in the diversion weir for use only during peak demand time in the evening and early morning.
Page 25 of 222
Ruacana Power Station. The current lack of clear policy interpretation regarding supply mix,
self sufficiency and promotion of renewables makes it very difficult for the regulator and the
trader to take decisions on power purchase agreements for new generators because it is not
clear what premium (if any) should be attached to the above-mentioned policy directives
contained in broad terms in the energy policy paper of 1998. This policy vacuum should be
resolved under the leadership of the Ministry of Mines and Energy through the development
of a national integrated resource plan to be approved by the Ministry which will guide future
investment decisions.
New generators which are small enough to feed into the distribution grids without needing to
access the transmission grid, compete with the selling rate of NamPower Transmission, and
not with the rate of Eskom imports because the REDs buy from Transmission and do not
import directly. This higher small scale in-feed price offers an opportunity for small,
decentralised power plants.
2.3 DEMAND AND SUPPLY

The challenge faced by Namibia is that domestic generation is not adequate to meet current
and future projected demand. To complicate this, supply constraints in South Africa are
resulting in Eskom not being able to meet South African electricity needs let alone those of
Namibia.
Namibia Peak Demand for Electricity

(excluding Skorpion)
500
450
400
350
300
Peak Demand
250 Installed Generator
MW
200 Capacity
150
100
50
0
89
91
93
95
97
99
01
03
05
07
19
19
19
19
19
19
20
20
20
20
Financial Year ending June
Figure 2-7: Namibia Installed Capacity and Peak Demand6
Figure 2-7 shows clearly that Namibia’s installed generation capacity has for some years not
been sufficient to meet system peak demand. What this figure does not show is the fact that
Ruacana power station has an insufficient water supply and cannot run at full capacity
throughout the year.
Figure 2-8 is possibly a better illustration of the contribution that Ruacana makes towards
meeting demand. This figure also shows contributions from planned (2003) new generators.
However none of the new power sources shown in Figure 2-8 (NamPower 2002: 12) have
6
Compiled from NamPower annual reports from 1993 to 2007.
Page 26 of 222
materialised, leaving local generation at the level as shown for 2003 in Figure 2-8. Thus
Namibia is currently dependent on imports to meet its electricity demand.
1600
1400
1200 Ruacana P/ S
240MW
Kunene
1000 360MW
MW
800 High Case
Par at us P/ S Van Eck P/ S Base Case

600 24MW 108MW
Low Case
Kudu CCGT P/ S
400
800MW
Or ange River Hydr o P/ S Divundu P/ S

200
72 MW 30 MW
Figure 2-8: Namibia Demand and Supply Forecast
For the past twenty years Eskom has had surplus generating capacity and has sold
electricity at very low prices by world standards. However, slow and stagnant electricity
supply industry (ESI) reforms and government intervention in South Africa have delayed
Eskom capacity extension plans, resulting in the capacity shortages that are now prevalent.
Figure 2-9: South African Demand and Capacity
The South African electricity predicament is illustrated in Figure 2-9 (Department of Energy
2008: 4). Here it can be seen how demand, shown as the bar part of the figure, has grown
Page 27 of 222
more rapidly than supply, shown as the two lines in the figure. South Africa now has an
average reserve capacity of less than 10%. This means that peak demand becomes
increasingly difficult to meet especially if plant failure occurs, scheduled and unscheduled
maintenance occurs and refuelling of the nuclear power plant is required. To date this has
already led to rolling black-outs throughout South Africa, affecting most sectors, including
industry.
With Eskom’s first priority being to secure supply within South Africa, this development has
jeopardised the electricity supply to neighbouring nations that rely on large scale imports
from Eskom. Namibia is among those countries that now face supply constraints that are
likely to last for at least another five years. If new generation capacity is not brought into
production as planned, or if demand in South Africa keeps rising at higher than expected
rates, then these constraints may well last longer than five years.
Namibia therefore finds itself in a difficult position. Its security of supply is compromised and
power shortages are likely during the coming years because of insufficient local generation
capacity and constrained imports. This is an unprecedented occurrence in a country that has
enjoyed a reliable and stable power supply in the past. Power outages were an uncommon
and rare occurrence in recent Namibian history. At the same time electricity prices are rising
dramatically due to a conglomeration of factors. This places strain on people and business
as well as on the utilities struggling to fund the necessary investments and still provide
quality service.
2.4 ENVIRONMENTAL ISSUES

The three power stations NamPower currently operates in Namibia were commissioned in
the 1970's prior to any required environmental impact assessments. As such, no information
is readily available on the environmental baseline data for these three stations. Further, the
NamPower Annual Reports for 2005, 2006 and 2007 do not report any environmental
statistics for these three power stations. However, in a presentation in 2007 (Carstens 2007)
at a World Energy Council and UPDEA Workshop in Windhoek, it was reported that the Van
Eck coal-fired power station utilized the following resources:
• Total water used 0.32 litres/USO
• Coal burnt (average) 0.57 kg/USO
The Van Eck is a dry-cooled power station due to the water constraints in Windhoek. A
conventional wet cooled power station uses a re-circulating system in which cooling takes
place via evaporation in an open cooling tower. Approximately 85% of the total quantity of
water supplied to a power station evaporates through these open cooling towers. In contrast,
dry-cooling technology does not rely on open evaporative cooling for the functioning of the
main systems. As a result, overall power station water use can be approximately 15 times
lower than a conventional wet-cooled power station (Eskom 2008). Whereas the water
consumption for a conventional wet-cooled power station can be around 2 litres per kWh of
electricity sent out, a dry-cooled plant can be as low as 0.08 litres per kWh of electricity sent
out (Eskom 2008).
The Namibian Ministry of Environment and Tourism (MET) has identified key environmental
issues needing to be addressed in Namibia (MET 2007). The power supply industry impacts
each of these. They include:
• Degradation of ecosystems, desertification, and loss of productivity;
Page 28 of 222
• Decline of water availability;
• Depletion of natural resources;
• Detrimental change in biodiversity;
• Decline of water quality;
• Pollution including toxic chemicals;
• Waste generation and littering;
• Greenhouse effect;
• Ozone layer depletion; and
• Acidification
Climate change is considered one of the most serious threats to Namibia's environment,
human health and well-being as well as its economic development. The arid environment,
recurrent drought and desertification have contributed to make Namibia one of the most
vulnerable countries to the effects of climate change (MET 2008). In 1994, the greenhouse
gas emissions from Namibia were 5,614 gigagrams of carbon dioxide equivalent, excluding
the bush encroachment which is thought to be a net 'sink' of carbon dioxide (MET, 2003).
Namibia is highly dependent on its natural resources such as minerals, fish, agricultural land
and wildlife. The variable rainfall, frequent droughts, and reliance on subsistence agriculture
combine to make Namibia highly vulnerable to climate change (MET, 2003).
As a Party to the United Nations Framework Convention on Climate Change, Namibia is

obliged to prepare an Initial National Communication (INC) on climate change within the
country. The INC was completed in July 2002. Additionally, the MET is designated as the
lead ministry to coordinate climate change activities in Namibia. The national programme
established the Namibian Climate Change Committee (NCCC) in 2001. It is a multi-
stakeholder committee which advises the government on policies and strategies around
climate change.
2.5 POTENTIAL OUTLOOK

This section explores some of the potential future problems that could be faced by the
electricity industry.
2.5.1 Cost
Namibia has in the past had very favourable long-term contracts with Eskom, based on the
excess capacity in the South Africa. A new contract has been negotiated recently, but this no
longer offers firm supply and prices are escalating rapidly due to the change from excess
capacity to supply shortages in South Africa. Figure 2-107 illustrates the nominal cost of
sourcing electricity as reported in NamPower’s annual reports. According to this the average
cost per kWh has almost tripled (in nominal terms) in the last eight years.
7
Graph compiled from NamPower Annual Reports 2000 to 2007 using “cost of electricity” from the income statement and units
sent out” from the key statistics.
Page 29 of 222
Electrity Procurement Cost Trend

10
9
8
7
c/kWh (nominal)
6
5
4
3
2
1
0
1999 2000 2001 2002 2003 2004 2005 2006 2007
Financial Year ending June
Figure 2-10: NamPower Average Electricity Procurement Cost
The supply from Eskom has already been constrained in 2007 and these constraints are
expected to continue until at least 2012 when Eskom is scheduled to bring new generation
capacity online. Since demand in the South Africa will also have grown it is uncertain how
long it will take before imports from Eskom will no longer be constrained. The most optimistic
case used as baseline for the modelling was that in 2012 firm supply from Eskom would
become available again.
Figure 2-11: South African Generation Cost Forecast
Page 30 of 222
Figure 2-11 (NER 2004: 35) illustrates the forecast long run marginal cost8 of power
generation for South Africa as compiled in 2004. Capital and fuel costs have risen
significantly since that time. The modelling of Eskom imports for this report was based on the
above price curve, assuming that the average cost of imports will not be less than the long
run marginal cost.
Figure 2-12: Regional Electricity Prices
Electricity prices in Namibia are threefold higher for commercial users and 70% higher for
domestic users than in South Africa and are among the highest regionally as illustrated in
Figure 2-12 (SAD-ELEC 2006: 19). The differential in electricity price between South Africa
and Namibia is attributed mainly to the much higher transmission and distribution costs in
Namibia. This is a consequence of Namibia’s low population and load density and the higher
taxation of electricity by local authorities. The taxation is used to cross-subsidise other
services provided by these authorities who receive none or very limited fiscal support from
central government.
NamPower has explored other import opportunities and has signed an agreement with the
Zimbabwean utility, ZESA, for 150MW supply with a five-year term, that is scheduled to
commence in 2008 (NamPower, pers. comm., 22 October 2007). Additional supply
agreements with Zambia are also being negotiated, but the status of that agreement is not
known.
2.5.2 Supply options

There are a variety of supply options available to Namibia. Namibia has, for the past fifteen
years at least, considered building new power stations. First, the Epupa and Baynes hydro
schemes were pursued in the mid-1990’s, while the Kudu gas-to-power project is being
8
LRMC = long run marginal cost which includes both capital costs and overheads as well as operating and fuel costs. SRMC =
short run marginal cost which is the additional cost of producing the next kWh and includes only operating and fuel cost and
excludes overheads and capital cost.
Page 31 of 222
pursued more recently. However, none of these (and other smaller) power generation
schemes have resulted in an investment decision.
While the details of the generation and demand management options are explored in section
3, there are a variety of options that are available. On the generation side the ones that have
been included in this study are:
• Baynes Hydro-Electric Power Generation
• Coal-Fired Power Generation
• Natural Gas Power Generation
• Nuclear Power Generation
• Electricity Generation from Biomass
• Concentrating Solar Thermal Power Plants
• Solar Photovoltaic Power Plants
• Wind Energy Power Generation
• Integrated Solar Combined Cycle Plant
On the demand management side the ones that have been included in this study are:
• Compact Fluorescent Lamp roll out to households
• Ripple Control for electric water heaters
• Solar Water Heater roll out to households
2.5.3 Environmental considerations

Electricity generation options must be considered in the light of their environmental
implications. In this regard the WorldWatch Institute, in its State of the World 2008 report,
highlights the seven principles of sustainable global economy. These are (WorldWatch
Institute 2008):
• Adjust the scale of economic activity to fit within the boundaries set by the natural
environment;
• Make prices tell the ecological truth, by incorporating environmental costs into the
price of goods and services;
• Shift the goal of economies from growth and development, by focusing on generating
greater levels of well-being;
• Account for nature's economic contributions through a full assessment and valuation
of nature's services;
• Apply the precautionary principle of economic activity;
Page 32 of 222
• Revitalize ancient principles of commons management to govern resource use in an

increasingly crowded world; and
• Properly value the extensive contributions women make to any economy.
2.5.3.1 Carbon and climate change
Climate change models reported by the Namibian Ministry of Environment and Tourism state
that temperatures could rise in the country by 2 to 6 degrees Celsius by 2100 (MET 2003).
Sea levels could rise anywhere from 0.3m to 1m. Rainfall patterns are uncertain with
reported ranges from an increase of 30mm per year to severe decreases of 200mm below
the annual averages. Changes in rainfall and temperature will affect biodiversity and
ecosystems. The greatest impacts are likely to be felt in the central regions of the country
with evaporation likely to rise 5% per degree of warming (MET 2003).
Water flow in the Kunene River is essential for the operation of the Ruacana hydro power
station. Electricity generation would be affected with increased evaporation and reduced
rainfall. However, the MET (2003) reports that some climate change models suggest that the
rainfall in the catchments of southern Angola could increase thus allowing more runoff in the
Kunene River.
Carbon intensity of the economy (as reflected in CO2 emissions per unit of GDP) will become
increasingly an issue of global economic competitiveness. Additionally, internationally, some
utilities are reporting carbon emissions per customer as they invest in and change their
energy mix towards renewable energy. This figure helps meet their customers' demand for
green energy, but also the utilities' image as going 'green.' The ratio of GDP to carbon
dioxide emissions were calculated using 2004 global figures produced by the International
Monetary Fund (for GDP) and the Carbon Dioxide Information Analysis Centre (CDIAC) for
the United Nations (for CO2) (Wikipedia 2008a). A sample of countries including Namibia
using this ratio is listed in Table 2-1.
Country GDP (USD billions) CO2 Emissions (in GDP per Emissions (in
thousands of metric thousands of USD per
tons) metric ton)
Namibia 5.61 2,471 2.270
Bahamas 5.661 2,009 2.818
Australia 639.356 326,757 1.959
United States of America 11,712.48 6,049,435 1.936
Table 2-1: The carbon intensity of a number of countries
Page 33 of 222
The prevailing international prices for carbon credits (or CERs)9 range from USD 3.50 per ton
CO2 at Chicago Climate Exchange to USD 15.80 per ton in various European markets
(Jindal 2006). As of 2 November 2007, 828 projects have been registered by the clean
development mechanism (CDM) Executive Board as CDM projects. These projects reduce
greenhouse gas emissions by an estimated 171 million ton CO2 equivalent per year
(Wikipedia 2008a). CERs from CDM projects were in 2006 traded on a forward basis
between Euro 5 and Euro 20 per tonne of CO2 equivalent (for further detail, see CDM Gold
Standard <http:// http://www.cdmgoldstandard.org/> and United Nations Framework
Convention on Climate Change <http:// http://unfccc.int/2860.php>) . For example, the
Kuyasa low-cost urban housing energy upgrade project in South Africa forward sold 10,000
CERs at a price of Euro 15 to the UK government to offset the G8 Summit at Gleneagles
(SouthSouthNorth 2005).
There are concerns that international carbon projects may bypass Africa which contributed to
just three percent of the total global trade in carbon offsets in 2003-2004 and a negligible
share in 2004-2005 (Jindal 2006). The global demand for carbon credits will increase steadily
as the first commitment period under the Kyoto Protocol (2008-12) draws near. Countries are
exploring cost-effective measures to reduce carbon emissions and carbon sequestration is a
viable option. The total market for carbon sequestration could be worth $300 million annually
(Jindal 2006).
The environmental and social impact of Export Credit Agency financing has received
increasing attention. However, while much discussion has revolved around setting
environmental guidelines and how to accelerate renewable energy technologies, there are
few (if any) procedures or standards evaluating project impacts on greenhouse gases,
contributions to energy-efficiency improvements or discussions on reporting on project
contributions to greenhouse gas emissions (Maurer 2003). In reforming Export Credit Agency
financing, it is believed that methodology will need to be developed to measure the
emissions to be generated by a project during its construction and development, and the
emissions it is likely to displace or eliminate to improve decision making and strategizing in
the energy sector (Maurer 2003).
While managing and reducing greenhouse gas emissions are important the first step will be
to measure these emissions in Namibia. Without accurate and complete data on the sources
of emissions and the amount they emit climate policies and strategies will be compromised
(Rich 2008).
2.5.3.2 Environmental costs
There can be little doubt that the environmental costs of generation options should be taken
into account. These environmental costs include health impacts, crop losses, acid rain and
global warming, to mention a few. In an exercise in Europe Voss (2000) has quantified these,
9
'Certified Emission Reduction (CERs, commonly known as 'carbon credits')' is a credit equivalent to one tonne of CO2 reduced
under the Kyoto Protocol Clean Development Mechanism (CDM). If a project is registered (approved) and implemented, the
CDM Executive Board (EB) issues credits called Certified Emission Reductions (CERs) (Wikipedia 2008). To avoid giving
credits to projects that would have happened anyway ('free riders'), rules have been set to ensure that projects reduce
emissions more than would have occurred in the absence of the project (additionality). However, there is confusion and various
interpretations of the additionality criterion which will not be discussed in this report.
Further, with costs of emission reduction typically lower in developing countries than in industrialised countries, industrialised
countries can comply with their emission reduction targets at much lower cost by receiving credits for emissions reduced in
developing countries as long as administration costs are low (Wikipedia 2008). While there would always be some cheap
domestic emission reductions available in Europe (for instance), the cost of switching from coal to gas could be in the order of
Euro 40-50 per tonne CO2 equivalent.
Page 34 of 222
the results of which are given in Table 2-2. Here he finds, for example that coal fired power
stations have an environmental cost of € 2.6c per kWh while wind and hydro are € 0.09c and
€ 0.07c respectively.
Quantifiable external costs of energy systems (in Euro-cents/kWh)
Impact
Coal Lignite Gas CC Nuclear PV Wind Hydro
Health effects 0.8 1.0 0.3 0.2 0.4 0.05 0.04
Crop losses -0.03 -0.03 -0.01 0.0008 -0.003 0.0005 0.0004
Material damage 0.02 0.02 0.007 0.002 0.01 0.001 0.0007
Noise nuisance 0.006
Acidification/ 0.2 0.8 0.04 0 0.04 0 0
Eutrophication a)
Global warming b) 1.6 2 0.8 0.03 0.3 0.03 0.03
Sub-total 2.6 3.8 1.1 0.2 0.8 0.09 0.07
Table 2-2: Externality costs for a range of supply options
a) Valuation based on marginal abatement costs required to achieve the EU “50% – Gap Closure” target to reduce acidification
in Europe.
b) Valuation based on marginal CO2-abatement costs required to reduce CO2 emissions in Germany by 25% in 2010 (19
Euro/tCO2).
The challenge with costing in environmental impacts in this study is that, very often, the
environmental cost is site specific. For example a power station located in Windhoek is going
to have very different environmental impacts to one located along an isolated coast. Hence
this study has included the value of carbon credits at a rate of Euro 19/ton as per Table 2-2
and the mass of NOx and SOx emitted. The valuation of other environmental impacts was
beyond the scope of the study.
Page 35 of 222
3 Description of Methodology
This section outlines the methodology that was used in this study. The first major distinction
that needs to be made is the difference between the technical analysis and the economic
analysis both of which have their own methodologies. Each of these methodologies is
presented in sequence below. The analysis from the technical modelling was used as an
input into the economic model to determine the overall economic impact of a variety of
generation options.
3.1 TECHNICAL METHODOLOGY

The technical analysis was conducted in the following sequence. First a demand forecast
was made about the likely increase in demand for electricity in Namibia. Second a range of
generation options and demand management options were identified, assessed and costed.
Finally, these inputs were processed through an electricity dispatch model. This process is
illustrated in Figure 3-1 and discussed in more detail in this section.
TAX
GDP Growth
DX
Price elasticity
Conventional GX 1 TX
Income elasticity
GX
Conventional GX 2 Vision 2030
Conventional GX 3
Price Path
Costing
EE Option 1 Benefits
Costing Carbon
Benefits Conventional GX n Employment
Carbon EE Option 2 Capacity
Employment Supply Iteration Demand Potential / Resource
Capacity Engine Import/Export
Potential / Resource Options in Projections in EE Option 3 Investment
Import/Export Constraints
Investment
Namibia Namibia Building Blocks
Constraints EE Option n Timing
Building Blocks
Timing
Renewable GX 1
Demand Path
Mining
Renewable GX 2 Consumption Numbers
Residential
Import shortage
Renewable GX 3 Other
Export surplus
Renewable GX n
Balance
Demand
and
Supply
Economic Analysis and Interpretation

of selected Scenarios
Figure 3-1: Dispatch Model Overview
3.1.1 Demand Forecast

The first component of the modelling is a demand forecast. This study built a demand
forecast model which differentiates demand into transmission and distribution customers.
Transmission customers are not linearly price sensitive and grow in steps. Distribution
customers are linearly price and income sensitive. Their electricity demand grows in relation
to economic growth and electricity price. The electricity price seen by end users is
determined via the total demand and the mix of generators applied to serve that demand.
The demand in turn is influenced by the price. This requires an iterative model that will solve
for a given set of generators and a given rate of economic growth. This study constructed a
spreadsheet model that was able to solve this iteration process.
Page 36 of 222
The analysis does assume that decisions to build new generating capacity are indeed
influenced primarily by economic (i.e. demand) considerations. We recognise that
experiences in South Africa and some other countries indicate that this might not necessarily
be the case.
The current demand profile is presented in Figure 3-2 (ECB 2006: 124) for the winter and
summer months. This data is based on the calendar year 2005. More recent data has not
been made available by NamPower.
Av e rage We e kday De mand pe r Hour
400
380
360
340
320
Winter
MW
300
Sum m er
280
260
240
220
200
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Hour of Day
Figure 3-2: Average weekday demand profile for winter and summer
The load curve is characterised by a night-time trough, a daytime plateau and a distinct
evening peak. This poses specific challenges for the power supply system, but also offers
significant opportunities for demand side management interventions.
Namibian Demand Forecast including impact of Step Loads & Elasticity

2 000
1 800
High Load forecast (incl all step loads)
1 600 High Load forecast (incl high & med probability step loads)
Medium Load forecast (incl high & med step loads)
Low Load forecast (incl high step loads)
1 400 3.55% Scenario
Vision 2030 scenario
System Demand (MW)
1 200
1 000
800
600
400
200
-
2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034
Year
Figure 3-3: NamPower Demand Forecast

The peak demand forecast in Figure 3-3 (NamPower 2006b: 3) represents scenarios
developed by NamPower for different probabilities of step loads and different economic
Page 37 of 222
growth expectations. The exact underpinning assumptions are not stated explicitly in the
NamPower document. The forecast does not include more recent developments such as the
latest uranium mining and processing ventures, but makes provision for unspecified step
loads anticipated at the time of compilation.
Currently Namibia is experiencing a uranium mining boom. NamPower (NamPower, pers.

comm., October 2007) claims that as much as 200MW of uranium mining and processing
load is being expected by NamPower during 2008 to 2010. Whether these loads materialise
will depend among other on the availability of power. For the purposes of this report it has
been assumed that this load will materialise.
Figure 3-4 depicts the demand forecast produced for this study. The strong growth during the
period 2008 to 2010 reflects the expected 200MW uranium mining and processing load. The
growth of the distribution sector load (i.e. non-mining and processing) has been modelled at
2% (low growth), 3.7% (base growth) and 6% (high growth) per annum and a price elasticity
of -0.35 and income elasticity of 1.27 has been applied.
The transmission (largely mining and processing) load is modelled as increasing at 10MW
per annum after 2011. This type of load is not directly sensitive to local electricity price or
economic growth because it is influenced primarily by global commodity prices and global
commodity demand. No concrete forecasts for this type of load exist, although many projects
have in the past been identified. It is impossible to predict which ones will materialise. For
this reason growth in this sector has been modelled as a flat annual increase in load of
approximately 2.5% (which is an estimate) as a proxy for the expectation that there will be
some growth over the years.
Projected System Peak Demand
1 100
1 050
1 000
950
900
850
800
750
MW
700
650
600
550
500
450
400
350
05/06
06/07
07/08
08/09
09/10
10/11
11/12
12/13
13/14
14/15
15/16
16/17
17/18
18/19
19/20
20/21
21/22
22/23
23/24
24/25
25/26
Years
Without DSM With DSM
Figure 3-4: Current Demand Forecast – System Peak
Later in this report a variety of generation scenarios are developed and discussed. Each of
these has its own demand forecast. For illustrative purposes the demand forecast of the so-
called ‘base case’ scenario is reported on in Figure 3-4. This illustrates the projected system
peak demand with and without implementation of the identified demand side management
Page 38 of 222
(DSM) measures for the ‘base case’ scenario. A steep increase is projected during 2009 to
2012 driven by the uranium mining boom.
3.1.2 Generation and Demand Management Options

This section gives a description of the potential generation and demand management options
that have been used in this study. A more complete description can be found in Appendix C.
3.1.2.1 Generation options
This section describes a range of electricity generation options that are being considered for
Namibia. It does this in the context of the technical, financial, economic, social and
environmental criteria.
During the course of this project a number of additional generation options appeared in
discussions in the Namibian ESI, among these options were a geothermal plant, a plant
based on used shipping oil products, and an additional generator that can be installed at
Ruacana power station. These however came at a time when the modelling had advanced
too far to take them into account, and in most cases the data pertaining to these options was
also regarded as confidential or speculative.
An abbreviated overview of the generating options is given here, stating technology

employed, generation potential, generation capacity considered, costs and risks. Generation
plants are classified into three categories, mid-merit, peaking or base-load or a combination
of the three categories. A mid-merit power plant is typically operated to follow the variable
demand on the system and is therefore able to operate at different power output levels. A
peaking plant only runs during peak times and is not utilised during off-peak times. A base-
load plant runs continuously at close to full output.
3.1.2.1.1 Baynes Hydro-Electric Power Generation
The proposed Baynes hydro dam is located on the Kunene River, which forms the border
between Namibia and Angola, downstream from the Ruacana hydro power station. The
catchment area of the river is pre-dominantly on the Angolan side.
According to NAMANG (1998: 2-1) the Baynes development has a dam capacity of 2.6 billion
cubic metres and an active storage, i.e. usable water capacity for electricity generation of 1.8
billion cubic metres. It has a generation output of 360MW with options for 540MW and
600MW. It has an expected service life of 40 years. The implementation lead time for the
development is estimated at 2 years pre-development time, three years construction time
and 5 years for the dam to fill up10.
The potential generation output for the Baynes dam is estimated at approximately 1,147GWh
per annum (Maritz 2000). Should the Gove dam in Angola be taken into operation again then
the annual generation potential of the Baynes Power Station can reach about up to 1,700
GWh.
The cost of Baynes in 1998 was estimated at USD 555 million (NAMANG 1998: 2-2). The
Team estimated that the cost of the project in today’s USD is estimated at USD 3.9 million
per MW. This high cost is the result the high world wide demand for generation equipment
10
Information provided by Udo Kleyenstüber, NamPower during a meeting on 22 October 2007
Page 39 of 222
and the specific risks associated with building the Baynes dam wall. The operating cost is
estimated at NAD 5 c/kWh and is based on the Ruacana hydro power station.
The project faces a number of risks:
• One of the key risks during the construction of the dam is the challenging dam wall
construction. While this has been identified as a major risk (Maritz 2000) it is likely
that construction methods have developed to such an extent that this risk can be
mitigated to a better degree today.
• A further key risk in operating hydro dams is that there is no water, either due to
extended periods of drought, climate change or upstream water abstraction, e.g.
major irrigation projects can have an impact on the flow of the river and therefore its
power generation capacity.
• Vulnerability to political instability, and terrorism related events.
3.1.2.1.2 Coal-Fired Power Generation
Coal-fired power stations have been around since the early days of electricity generation and
for this reason the option has been included in the generation options. It is however
problematic because of the challenges that a coal-fired power station would face in Namibia.
These are:
• Lack of domestic coal reserves. This means that coal would have to be imported with
the resultant cost implications, transportation implications and the potential lack of
security of supply. Apparently there are coal reserves in Southern Namibia, but these
have not been quantified to be included in this study.
• Lack of water for cooling purposes. A dry-cooled station is not particularly desirable
because of the particle emissions. Locating a coal power station at Walvis Bay is an
option but does have its own problems in using sea water for the purposes of cooling.
The following technical specifications have been used for the purposes of modelling this
option:
• A capacity range of between 175MW and 700MW with step sizes of 175MW.
• The generation output is 1 226GWh per year at 175MW. Own electricity usage is
10%.
• It has been modelled as a dispatchable base-load power plant.
• The capital cost is estimated at USD2.2m per MW United States Department of

Energy (USDOE 2007) with decommissioning costs of 10% and a capital import
component of 80%.
• The fuel cost is expected to be NAD 27c/kWh11 with an import component of 95%
(allowing for a small portion of the transport and handling cost being local cost).
11
Based on coal costs stated verbally by NamPower for van Eck Power Station, adjusted for better plant efficiency of new plant
and lower transport cost due to plant location at Walvis Bay.
Page 40 of 222
3.1.2.1.3 Natural Gas Power Generation
Power generation from natural gas is a well established, mature technology used all over the
world. A combined cycle gas turbine (CCGT) plant is a high efficiency machine available as
off-the-shelf technology and as such has relatively short lead times.
Gas in the Namibian context is to be supplied by the Kudu gas field. This gas field has been
subject to attempts at exploitation for more than 20 years, so far without success. The last
seven or so years have seen concerted efforts by NamPower to develop a power station
based on the Kudu gas field. This has, however, so far not resulted in an investment
decision, and most recent indications are that the project will not go ahead in the short term
(NamPower, personal comm. 2007).
The challenges with developing the Kudu gas field include the relatively high costs of
developing the upstream gas side with the field being located 170 km out at sea (CSIR &
NamPower 2004) and at a depth of 170m and disagreement about the price and currency in
which the gas would have to be paid.
The Kudu gas resource is considered to be proven to sustain an 800MW power plant for 20
years (CSIR & NamPower 2004). It could thus sustain a 400MW plant (which would be much
more suitable for the Namibian system) for 40 years if the gas infrastructure and repayment
period could be arranged accordingly.
Due to confidentiality of the Kudu project the figures used in the modelling have been derived
from public international sources.
The gas turbine power plant has been modelled under the following assumptions:
• It is a base-load plant with a capacity range between 400MW and 800MW in steps of
400MW.
• It has a generation output of 2.98GWh per annum at 400MW using 5% for its own
capacity.
• The project development phase is assumed to take one year and the implementation
phase four years.
• It would have a capital cost of USD 0.66m per MW. Decommissioning costs of 5%
and a capital import component of 80%.
• The operating cost would be USD 0.5c per kWh (US Department of Energy 1999: 5).
Operating expenses, excluding fuel costs, would have a 25% import component.
• The fuel cost has been modelled at USD 4c per kWh (US Department of Energy
1999: 5).
There are a number of risks for such a generation option located in Namibia:
• The most critical risk lies in the significant foreign exchange risks inherent in the gas
pricing, as well as the risk of not achieving an acceptable export price for excess
power that would be produced by a plant of this type.
• There is some risk inherent in a gas field, without proven actual extraction, that
extraction of the gas may prove more difficult than expected.
Page 41 of 222
• The recent increase in demand for gas turbines internationally and related expertise
could result in longer lead times.
3.1.2.1.4 Nuclear Power Generation
Nuclear power plant (NPP) technology has been in use for over 50 years and currently about
30 countries rely to some extent on NPP for their power generation needs. Recent increases
in coal and gas prices as well as climate change concerns associated with fossil fuels have
led to a resurgence of interest in nuclear power.
This supply option has been modelled under the following assumptions:
• Pebble bed modular reactor (PBMR) with a 30 year service life operating as a
dispatchable base load plant.
• A capacity range of 165MW and a generation output of 1 300GWh per annum.
• The lead time is assumed to be 6 years for the project development phase and 4
years for the implementation of the plant. Due to the uncertainty in the technology
developments, it is assumed that such a plant will only become available 5 years after
anticipated implementation in South Africa in 2013.
• A capital cost of USD 4m per MW with decommissioning costs of 30% and a 95%
imported component in the capital cost (PBMR (Pty) Ltd, personal comm. 2007).
• The operating and fuel costs are estimated at USD 1 c/kWh each12. The fuel cost has
been based on the cost of fuel for conventional reactors with an additional margin for
the specialised manufacture of the fuel pebbles.
The potential risks associated with nuclear power include:
• There is little consensus in literature as to the actual worldwide reserves of uranium.

Assuming continued operation of all the existing nuclear plants one source estimates
that the reserves will last for 70 years (OECD Nuclear Energy 2003). With new
nuclear plants under construction it is estimated that by 2050 there will only be
resources for selected countries.
• Long-term nuclear waste storage remains an unresolved issue and costs are not
factored into the current operation as no precedent exists.
• Although the likelihood of major accidents is very low they cannot be excluded. The
potentially devastating consequences could negate the economic benefit of such an
option.
• Nuclear power plants cannot be insured against the consequential damages in the
case of an accident.
12
Fuel cost is quoted as USD 0.5c/kWh in Australian Uranium Association (2007). With the recent increases in the cost of
uranium (while understanding that the uranium portion of enriched uranium is approximately 25% of the cost (Australian
Uranium Association 2007) and the yet uncertain cost of the manufacture of PBMR fuel pebbles, it is assumed that fuel costs
will be in the region of USD 1c/kWh.
Page 42 of 222
3.1.2.1.5 Electricity Generation from Biomass
Biomass is a collective term for organic matter that could be used in energy production.
Diverse fuels derived from wood, agricultural wastes, food processing wastes, animal
manure, solid municipal waste and others embody the term biomass. Biomass can be
burned directly in boilers but the pollution produced and the low energy gain when this
method is utilised have been the driving forces in the development of new methods of energy
production from biomass.
The Desert Research Foundation of Namibia (DRFN) is currently implementing a proof-of-

concept project referred to as CBEND (Combating Bush Encroachment for Namibia’s
Development), which is to demonstrate the technology and focus on the operational and grid
integration issues to guide a future larger scale role-out of biomass to electricity plants within
Namibia. The costing information of this supply option is largely based on the DRFN concept
paper (Hager, Schultz & von Oertzen 2007a).
This generation option has been modelled under the following assumptions:
• A service life of 20 years.
• The technology is highly modular in terms of plant size and can be moved to different
locations, provided that the grid infrastructure is present.
• A capacity range of 0.5MW to 60MW in steps of 0.5MW.
• Generation output of 6.3GWh per annum with dispatch increment of 0.5MW and own
use of capacity of 3%.
• It is a short term13 dispatchable power source modelled as a base load/mid merit

plant.
• Implementation lead times for biomass to electricity plants are of the order of 2 years.
• It has an expected capital cost of NAD 16.7m per MW14 with 80% of the capital being
imported.
• An operating cost of NAD 4.7 c/KWh (Hager, Schultz & von Oertzen 2007b) of which
the import component would be 5%.
• A fuel cost of NAD 5.4 c/kWh (Hager, Schultz & von Oertzen 2007b) with an import
component of 30% due to the liquid fuels used in the bush removal process.
• Job creation of one skilled and two unskilled jobs per MW for construction and four
skilled and 52 unskilled jobs per MW during operation (Hager, Schultz & von Oertzen
2007b).
Some of the potential risks include:
• Resistance from farmers to the project.
13
Short term output for as long as biomass stocks are available and stockpiled.
14
Based on figures from Hager, Schultz & von Oertzen (2007a), with a 10% increase for grid connection and protection
interface.
Page 43 of 222
• Environmental emissions.
• Possible soil erosion of land disturbed for harvesting of biomass.
• Variation in yields due to differences in weather, geology, and soil conditions, but also
the loss of yields due to forest /crop fires.
• Labour related volatilities.
• Uncertainties around costs including possible charges for the invader bush biomass
resource by the land owner and additional costs for grid integration of the generators.
3.1.2.1.6 Concentrating Solar Thermal Power Plants
Concentrating solar power (CSP) plants have been operating in the Mojave Desert in the
USA for more than 20 years. In addition there are numerous such plants developed and in
small-scale operation in other parts of the USA, Australia and Spain.
The principle of operation of concentrating solar power plants is much the same as that of
conventional gas power plants. The difference is that the energy input in CSP plants is
obtained by concentrating solar radiation and converting it to high temperature steam or gas
that drives a turbine or a motor engine. CSP plants consist of two major components. The
first being the solar radiation reflectors with a heat transfer fluid. The second is the
conversion of the solar thermal energy to electric power. The CSP plants are classified in
terms of their solar collector type as parabolic trough, central receiver, and parabolic dish
plants. The availability of CSP can be extended beyond daylight hours by the use of thermal
storage. The annual solar capacity factor in systems without thermal storage is about 25%.
The solar capacity factor can be increased to over 50% if the solar field is increased and
thermal storage is added15.
This generation option has been modelled under the following assumptions:
• A capacity range of 50MW, based on one Andasol unit (European Commission

2007:14-16) to 300MW with steps of 50MW.
• A generation output of 220 GWh16 per annum per 50MW capacity.
• Implementation lead times are in the range of about 3 years. The service life of CSP
plants has been modelled as 25 years.
• Scheduled daytime dispatchability with high probability due to predictability of

Namibian solar resource and with a 7.5 hours storage option.
• A development phase of one year and implementation phase of two years.
• A capital cost of €5.2m per MW (European Commission 2007:14-16).

Decommissioning costs of 1% and an estimated capital import component of 80%.
• Operating costs of USD 1.2c per kWh (World Bank 1999) with an estimated import
component of 15%.
15
Personal communication with Michael Geyer, coordinator of Andasol 1, Spain.
16
Calculated for Namibia based on Andasol.
Page 44 of 222
• Zero fuel costs.
• Construction jobs of five skilled jobs and 1.5 unskilled jobs per 6.5 MW (Jones, J
2007:40). Operation jobs of 5 skilled and 1.5 unskilled per 6.5MW (AGAMA Energy
2003:17&39).
The risks associated with the technology is predominantly in the area of performance as
there has been little development in the last twenty years other than the performance
monitoring of the existing plants.
3.1.2.1.7 Solar Photovoltaic Power Plants
Photovoltaic (PV) technology was initially developed for powering satellites and space
stations in the 1950’s. Since then the technology has matured significantly through improved
efficiency, higher life expectancy and lowering of costs. Today PV is used pre-dominantly in
grid-connected systems in parts of the world where incentives are provided for power
generation through renewable energy. Off-grid applications in regions with good solar
resources are often served through PV where grid extensions or diesel generator systems
are too costly.
PV modules convert sunlight into DC electricity through what is referred to as the

“photovoltaic” effect. The focus of this supply option is on grid-connected PV plants that
consist of PV arrays and grid interactive DC to AC inverters that synchronise to the grid and
thus feed power into the grid. PV systems in conjunction with battery storage are generally
used for stand-alone applications and are not considered here due to the high cost of the
storage and the high maintenance requirements. PV plants without storage have very low
maintenance requirements. Service intervals range between bi-annual and annual visits.
The generation potential of solar PV in Namibia is not limited by the resource or the space
but rather by the amount of PV power that the Namibian grid can accommodate in terms of
network stability and how it fits into the overall generation mix.
This generation option is based on the following assumptions:
• A capacity range of 1 MW peak to 40 MW peak in step sizes of 1 MW peak.
• A generation output of 1.7 GWh per annum per MW peak.17
• A 30 year service life.
• A one year development and one year implementation phase.
• A capital cost of NAD 33 million per MW for plants in excess of 5W peak which is based
on the current price for PV at USD 3.75 per W peak (Bradford and Maycock 2007:65).
The import component of this capital cost is expected to be 95%.
• An operating cost of NAD 2 c/kWh18 with 20% of the operation cost imported.
• Job creation of 12 per MW (10 skilled and 2 unskilled) during construction and 0.1
skilled and 0.5 unskilled per MW during operation (AGAMA Energy 2003:17&39).
17
Calculation based on 72% nominal output from PV array (due to temperature related and power electronic losses) and an
average solar day at place of installation of 6.48kWh/m2/day.
18
Estimate.
Page 45 of 222
The risks associated with solar PV are minor. Quality is a concern and the use of reputable
manufacturer’s are essential in order to ensure that long-term performance is guaranteed.
3.1.2.1.8 Wind Energy Power Generation
The most common utility scale wind energy converters (WEC) are horizontal axis, up-wind
generators, which are based on a three blade design. The WEC, which have been
manufactured in the last ten years are pre-dominantly based on pitch-control which allows for
more refined wind generator control and higher wind power exploitation.
The grid integration of modern WECs is through power electronic converters which
synchronise to the grid and feed-in the generated power. Similar to grid connected solar PV,
the grid voltage needs to be present to allow for power generation. If the grid fails then the
wind generator goes off-line.
The wind resources along the Namibian coast are substantial in particular near Lüderitz and
along parts of the Skeleton Coast (Bicon Namibia 1999). The wind speeds in these areas are
strongest in the summer months19. The most promising sites where more detailed
measurements have been conducted are Lüderitz and Walvis Bay. Focus has also been on
these sites due to existing civil and grid infrastructure.
The modelling of this generation option is based on the following assumptions:
• A capacity range of between 30MW and 60MW with a 30MW step.
• A generation output of 2.8 GWh per annum per MW based on a 32% capacity factor
expected at Lüderitz wind sites.
• A supply type of intermittent and seasonal generation.
• A development period of one year and construction phase of one year.
• Capital investment of NAD 10.5m per MW (Morthorst 2004:99) and an import

component of 85%.
• An operating cost of NAD 6 c/kWh (Morthorst 2004:100) with an import component of

10%.
The risks associated with wind energy are low, the main risk being concerns about resilience
under harsh coastal conditions, integration into the Namibian grid, changes in wind regimes
(climate change), while the remaining risks would impact on the financial performance of the
wind park.
3.1.2.1.9 Integrated Solar Combined Cycle Plant
Integrated solar combined cycle (ISCC) gas turbine power plants are essentially based on a
conventional combined cycle gas plant with an added solar boost. The combined cycle
natural gas systems are popular in areas where natural gas is available. A combined cycle
plant uses gas combustion turbines as the first stage in electricity generation. The hot flue
gases from the combustion pass trough a heat exchanger to generate steam, which drives a
steam turbine at the second stage in the electricity production process. Concentrating solar
power can be integrated in the second stage of the combined cycle plant. The solar heat
19
Personal communication with Bernard Siepker, PolyTechnic of Namibia, October 2007.
Page 46 of 222
increases the capacity of the steam turbine by either generating additional steam in a heat
recovery steam generator or by generating low pressure steam to be injected directly into the
steam turbine. The conventional combined cycle systems have heat-to-electricity efficiencies
of about 55% (Enermodal Engineering and Marbek Resource Consultants 1999). The add-on
peak output of the solar system accounts for 20% to 30% of the combined cycle output. This
means that the system can boost the output of a 100MW combined cycle plant to 130MW.
The ISCC plants’ main advantage is that the peak capacity can be increased at a lower
capital cost and the solar power boost is available during daytime activity. This option is
modelled with:
• A capacity range of between 50MW and 150MW with a 50 MW step.
• A generation output of 6.6 GWh per annum per MW.
• A dispatchable mid-merit plant.
• A project development period of one year and construction phase of three years.
• Capital investment of USD 0.9m per MW and an import component of 80%

(Enermodal Engineering Limited and Marbek Resource Consultants Ltd 1999).
• An operating cost of USD 0.7 c/kWh with an import component of 20% (Enermodal
Engineering and Marbek Resource Consultants 1999).
• A fuel cost of USD 3.6 c/kWh assuming 25% solar plant size contributing for a third of
the time (US DOE 1999).
• Job creation of 0.019 skilled per MW and 0.13 unskilled during construction and
0.023 skilled and 0.11 unskilled per MW during operation20.
The main risk for this option is the financial exposure to foreign exchange risks, assuming
that the contract will be signed in foreign currency. The risk is however already slightly
mitigated through the CSP component, which uses local renewable resources.
A severe risk is the possibility of the gas reserves being lower than anticipated.
3.1.2.1.10 Comparison of the Generation Options
Some of the key aspects of the generation options are compared to each other in this
section. These relate to capital costs, tariffs and construction time lines.
The capital cost for each generation option is given in Figure 3-5 and is shown as the cost
per MW of generating capacity. These costs include the overall capital costs, which consist
of the overnight capital cost, the interest during construction (IDC) and the decommissioning
costs.
The generation option with the lowest capital cost is CCGT, followed by wind, biomass and
clean coal. The most expensive option is concentrating solar, followed by nuclear, Baynes
and solar PV.
20
Calculated as an average between CCGT and CSP plants.
Page 47 of 222
Estimated Capital Cost of Supply Sources
SWH
Ripple Control
CFL
Solar PV
"Clean" Coal
Nuclear
Biomass
Concentrating Solar
Wind
Baynes
CCGT
0 5 10 15 20 25 30 35 40 45 50 55 60 65
Millions
N$ / MW
Capex / MW IDC / MW Decommissioning Cost
Figure 3-5: Full estimated capital cost for generating options NAD million per MW.
The generation tariffs are calculated based on the capital cost21 and taking account of
maintenance and operating costs as well as fuel costs where relevant. The oil/gas fuel costs
are escalated by 3% in real terms while the price for coal is kept constant in real terms.
Cost per kWh for Supply Sources
SWH
CFL
Gas/Solar Hybrid
Zim Import
Solar PV
"Clean" Coal
Nuclear
Fixed cost
Option
Import Eskom
O&M cost
Biomass
Fuel cost
Concentrating Solar
Wind
Baynes
CCGT (Kudu)
Paratus
van Eck
Ruacana
- 20.00 40.00 60.00 80.00 100.00 120.00 140.00 160.00

c/kWh
Figure 3-6: Estimated base tariffs for generating options
21
As example, a power plant costs NAD16 million per MW, including IDC has a life of 30 years and a capacity factor of 85%.
This means per MW it can produce 1MW x 24h x 365 days x 85% = 7 446 000 kWh per annum. Amortising the capital of NAD16
million at a real rate of 6% per annum over 30 years gives an annual amount of NAD1162382 and dividing this into the kWh
output gives a fixed cost of NAD0.1561/kWh, i.e. 15.61 cents per kWh. The fuel cost per kWh and operating cost per kWh
(derived from research looking at typical plants of the type considered) are added to the fixed cost per kWh to arrive at the total
price per kWh that the generator would need to earn to run, repay its investment and give an acceptable return.
Page 48 of 222
Most of the new generation options have tariffs of 40 to 60 NAD cents per kWh. The major
exceptions to this are concentrating solar at around NAD 1 per kWh and solar PV at NAD
1.50 per kWh.
What Could Be Built ...
4500
4000
3500
3000 Gas/Solar Hybrid

Solar PV
"Clean" Coal
2500
Nuclear
MW
Biomass
Concentrating Solar
2000
Wind
Baynes
1500 CCGT (Kudu)
1000
500
0
09/10 10/11 11/12 12/13 13/14 14/15 15/16 16/17 17/18 18/19 19/20 20/21 21/22 22/23 23/24 24/25 25/26
Figure 3-7: Timeline of generators that could be built
Figure 3-7 above provides an illustration of how the various supply options were modelled in
terms of size that could technically be built in Namibia without facing major transmission
constraints or logistical issues. It also shows the earliest year in which power production
could start if an investment decision were taken in 2007. Some of the options are scalable
(i.e. can be built in steps up to the maximum shown above), which has a significant impact
on investment choices taken in the scenario development. Some options, for example
biomass, could possibly be expanded beyond the limit shown, but the concept first needs to
be proven in practice before this can be assessed.
Concentrating
CCGT Baynes Wind Biomass Nuclear "Clean" Coal Solar PV
Solar
Minimum Size MW 400.00 360.00 30.00 50.00 0.50 165.00 175.00 1.00
Step Size MW 400.00 90.00 2.00 50.00 0.50 165.00 175.00 1.00
Probable Limit MW 800.00 540.00 60.00 300.00 60.00 330.00 700.00 100.00
Limiting Factor Gas reserve Water resource Grid stability Cost & Storage Grid capacity Cost Coal Transport Cost & Storage
Table 3-1: Estimated Generator Capacities and Step Sizes
Table 3-1 summarises the parameters used to model the generators. The minimum size is
the smallest unit that is considered feasible from a technical and economic perspective. Step
size is in most cases determined by typical technology sizing (e.g. generator or turbine
standard sizes) or practical step sizes. The probable size limit is set taking into account the
limiting factor(s) listed at the bottom of Table 3-1. The limit was chosen such as to be on the
Page 49 of 222
safe side, i.e. the maximum sizes used in the modelling are not expected to create significant
grid integration or logistical problems.
3.1.2.2 Demand side Management options
This section describes a range of demand side options with their technical details, their
characteristics and their financial implications.
3.1.2.2.1 Options considered
The options considered here rely on the ECB DSM study which was compiled in 2006 and
enumerates and ranks possible options while developing those options considered most
appropriate in more detail (ECB DSM 2006).
Since this present study concerns itself with major grid electricity supply issues, only those
DSM options which offer a major impact on grid electricity consumption and which can also
be quantified with a high degree of confidence have been included in this study. This limits
the options included to three, being:
• Large-scale compact fluorescent lamp roll out to residential consumers,
• Wide-spread implementation of ripple control for electric and electric assisted solar
water heaters,
• Large-scale replacement of electric water heaters with solar water heaters with
electric back up.
The first and last of these are energy efficiency measures with a significant impact on the
national load profile because they are aimed at residential consumers with a rather specific
pattern of using both lights and hot water. The ripple control project is a pure DSM measure
with no energy savings, which only shifts demand from one period to another. Taken
together and implemented fully these three measures have the potential to reduce peak
demand by as much as 20%.
There are many other possible energy efficiency (EE) measures that can be undertaken in
Namibia such as the introduction of time of use tariffs, the increase of consumer awareness,
commercial and building energy efficiency measures as well as energy efficiency promotion
in energy intensive industries. The fact that they are not included in the modelling of this
study does by no means imply that they are not worth pursuing. They have been excluded
from this study primarily because they are difficult to quantify and even the DSM study only
produced estimates of what they may cost and what they may achieve. Such vague numbers
were considered inappropriate for inclusion in the modelling for this study, and developing
any of them to a more concrete level is outside the scope of this study.
3.1.2.2.2 Compact Fluorescent Lamp Roll Out
The distribution of compact fluorescent lamps (CFL) is a widely used measure of energy
efficiency with a high impact on networks with an evening peak. It can be rolled out in a short
timeframe, almost to the extent that it can be used as an emergency measure for unforeseen
power supply shortages.
The programme proposed in the DSM study (and since implemented differently by
NamPower) recommends the purchase of around one million CFLs and distribution of these
to residential consumers in exchange for used incandescent bulbs.
Page 50 of 222
This intervention is based on the following assumptions:
• The implementation of this programme is expected to result in a maximum demand

reduction of approximately 20MW during peak demand times and an annual energy
saving of 22 GWh (ECB 2006).
• Investment of NAD 625,000 per MW (ECB 2006).
• It is assumed that this programme can be implemented over a two-year period.
• The calculations in the DSM study are based on a total number of 106,770 grid
connected households. Assumptions have been made regarding the level of
penetration for converting from incandescent to compact fluorescent lamps at
household level. This was guided by the perceived accessibility within the various
centres.
• The total investment cost of this energy efficiency drive is estimated at NAD 13
million. This translates to a cost of NAD 625,000 per MW.
The main risks of this programme are:
• Low quality CFLs exist and may enter the market after the initial dissemination drive.
This would lead to premature failure and discrediting of the programme.
• The dissemination programme may find it difficult to reach a high percentage of

households, resulting either in lower penetration or higher dissemination costs. This
would lead to either a lowering of the expected savings or to cost increases.
3.1.2.2.3 Ripple Control
Ripple control of electric water heaters is a well-established means for utilities to defer
electricity consumption from peak to off-peak times. Most of Namibia’s larger towns have the
potential to beneficially implement ripple control to control demand peaks, as well as national
demand peaks and the timing of peak consumption.
Ripple control is a peak shifting facility and does not save energy. Therefore its impact is only
on maximum demand. The time of impact is related to the use of the appliance. Since
electric water heaters are most active during the night (significant contribution to evening
peak) as well as during the morning hours, those are the periods that ripple control can
impact on.
The following assumptions have been made:
• It is estimated that ripple control can provide at least an additional demand shifting
capacity of 24 MW over and above the existing installed ripple control capacity. This
is based on the practical experience of the City of Windhoek where an average drop
of 1kW is achieved per electric water heater switched off. The expected impact is
therefore based on 1kW per electric water heater switched (ECB 2006).
• The programme can be implemented over a four-year period in the major centres of
Namibia.
• The estimated cost for ripple control is NAD 2.5 million per MW. This is based on an
estimated cost of NAD 1,000 per receiver (including installation) and costs of around
Page 51 of 222
NAD 7 million for a 66kV injection station and NAD 2.25 million for an 11kV injection
station (ECB 2006).
• For around 27,000 receivers, one 66kV injector and eleven 11kV injectors the total
project budget is expected to be in the region of NAD 60 million (ECB 2006).
• Direct jobs during construction are 1.06 per MW for skilled and unskilled staff.
The risks that affect this DSM feature most significantly are:
• NamPower and the REDs cannot reach agreement on implementation and operation
of this DSM option.
• Significant consumer resistance may prevent the installation of receivers in more

homes than anticipated, thereby reducing the benefits of the system.
• A parallel solar water heater (SWH) implementation roll-out may lead to over-
investment in ripple control equipment since the back-up element of SWH can also be
controlled through timers.
• Utilities use the ripple system to improve their load profile vis-à-vis NamPower without
passing the benefits on to consumers.
• Technology costs are escalating, thus jeopardising the continued viability of the ripple
system.
3.1.2.2.4 Solar Water Heater Roll Out
Solar water heaters (SWH) are one of the technologies that are financially competitive with
electrical water heaters. The breakeven between domestic solar heaters and domestic
electric heaters is dependent on the tariff and on the metering and varies between 3 to 9
years (Emcon Consulting Group 2005).
With an estimated 97,000 electrical water heaters in Namibia, the electrical load is
significant, especially during the evening peak. According to the Emcon study an estimated
106MW are contributed to the evening peak through electric water heater (Emcon Consulting
Group 2005).
The following assumptions were made in the modelling:
• SWH programme is rolled out in addition to the ripple control programme. This is
done because a) the ripple control can be rolled out much faster than SWH and
brings relief for the critical evening peak and b) SWH will most likely still have electric
back up elements, that one would want to ripple control to avoid excessive peaks in
adverse weather conditions.
• The estimated impact on the evening peak for the purposes of this study is
approximately 80MW. Assuming an average daily consumption of 8.5kWh per electric
water heater an annual savings of 250GWh could be achieved.
• It is assumed that this programme can be rolled-out over a ten-year period.
• The investment cost is based on the average cost of one SWH costing NAD 15 000.
Assuming a 50% diversity factor a 2kW geyser will displace 1kW of maximum
Page 52 of 222
demand. Therefore 1 000 geysers displace 1MW. The investment cost is therefore
NAD 15 million per MW.
• Direct jobs during construction: 3.33 skilled and unskilled staff per MW.
The major risks of this energy efficiency drive are:
• Poor quality SWH equipment and installation practise leading to a poor reputation
and low uptake of the technology.
• Availability of second-hand electric water heaters on the local market as a result of

retrofit activity. Low-income households might be lured into adopting electric water
heaters for their homes as a result of the low initial cost.
• Serious exchange rate fluctuations may drive up the costs of a SWH if they are being
imported and not produced in Namibia.
• There are challenges in terms of existing installation capacity in Namibia as well as to

build further capacity to support the implementation of such an extensive roll-out.
3.1.3 Dispatch Modelling

Following the determination of likely future demand and identification of potential generation
and demand management options an electricity dispatch model was developed. The model
takes a specified constellation of generators, demand side management options and load
growth and solves the dispatch equation until supply and demand balance. If it cannot match
demand with existing supply then there is a residual called unserved energy.
The dispatch of generators and the revenue requirement for transmission and distribution
define the cost to be recovered through electricity sold and thus determine the end consumer
price. As distribution consumers are price sensitive higher prices result in reduced
consumption, and a different dispatch pattern for the generators. The model is designed to
iterate this sequence and solve for the lowest cost and least unserved energy.
Page 53 of 222
3.1.3.1 Technical Model Characteristics
The dispatch model allows for 15 different generators and five different demand side
management options. The characteristics of the model are described in Table 3-2.
Parameter Quantity Notes
Model timeframe 20 years
Generators Max 15 Generators are dispatched by hour for three “seasonal”

load profiles.
Generators have characteristic availability profiles for

low season and high/max season, which describes how
much of the installed capacity can be dispatched per
hour.
Generators have annual pricing divided into fixed, fuel

and operating rates. The fuel rate can be varied per hour
for low season and high/max season. The fixed rate is
calculated on rated annual output. Unrecovered fixed
cost due to lower than 100% dispatch is reported and
added to consumer price.
DSM Options Max 5 DSM options modify the load curve
Load Curve Max 3 The 2005 Namibian load curve was used and separated
into three “seasons”:
Low season = all days Aug – Apr
High season = all days May – Jul
Max days = 15 highest peak days
Unserved demand is reported per hour per season per

year.
Dispatch Hourly per Generators dispatched on price. “Must run” generators

“season” are dispatched first on the basis of their relative price
and thereafter the other sources are dispatched on the
basis of their relative price. Prices are defined hourly.
Page 54 of 222
Load Growth Annual Split between “Transmission” and “Distribution”.
Distribution growth subject to economic growth at 1.27

income elasticity and -0.35 price elasticity (ECB 2007a:
6).
Transmission growth taken as 75MW p.a. for 2008/9/10

(driven by uranium mining load as expected by
NamPower) and thereafter at 10MW p.a. (which
translates to an annual growth of around 2.5% of the
expected transmission load).
Generator “Build Annual New generators are entered into the model assisted by a
Decisions” calculation of unserved demand and price preference of
generators which could be brought on stream if an
investment decision is made earliest in 2007. Price
preference is weighted according to the contribution that
the generator types can make towards meeting the daily
load curve (e.g. a solar PV generator without storage
does not contribute to meeting the evening peak and
thus has less “value” than another generator that can
contribute).
Electricity Price Cost reflective The consumer electricity price is calculated on basic
cost reflective principles, i.e. generators, transmission
and distribution recover all their revenue requirement via
the kWh sold. In addition allowance is made for
• A return on investment of 6% real.
• Recovering generator fixed costs lost due to

generators not being dispatched 100%
• Recovering the fixed costs of DSM options
• Local Authority surcharge starting at the current

level and diminishing with inflation down to a real
2007 10c/kWh and kept constant thereafter in
real terms.
Environmental Environmental costs are given only in the physical volume of emissions
costs and and not in a monetary value except for CO2 emission. The remaining
benefits emissions have not been added into the cost of generation or electricity
tariffs because in the current electricity pricing regime there is no
provision for this.
Page 55 of 222
Carbon A baseline is calculated from the current CO2 emissions based on the
emissions emissions of van Eck, Paratus, as well as Eskom imports. This is divided
by the total energy consumption of Namibia. This results in a baseline of
approximately 0.6kg of CO2 per kWh. This baseline is calculated for each
year as a function of the generation mix. When a renewable energy
generator is built, it will generate carbon credits at the baseline emission
rate for the year within which it is brought online. The new RE generator
will then lower the baseline for the next RE generator. Addition of a CO2
emitting generation supply will increase the baseline as from your of
operation etc.
Social costs and Social costs and benefits have been itemised as contribution to GDP
benefits (because that is income); jobs – skilled and unskilled; contribution to
small, medium and micro enterprises (SMME)
Table 3-2: Technical Model Key Characteristics
3.1.3.2 Technical Model Assumptions
The assumptions on which the dispatch model is based are presented in Table 3-3.
Parameter Notes
Economic Economic growth has been assumed at 3.7% per annum as base case,
Growth based on the average actual growth reported by GRN. Sensitivity
analysis was performed at 2% and 6%.
Income elasticity Income elasticity for electricity was taken as 1.27 (ECB 2007a:6)
(distribution
customers)
Price elasticity Electricity price elasticity for distribution customers was taken as -0.35,
(distribution (ECB 2007a:6). Transmission customers were modelled as insensitive to
customers) price since they do not generally have a linear price-demand relationship
but rather a step relationship (i.e. they either operate or they shut down).
The modelling assumes that a “shut-down” price level for transmission
customers is not reached.
Tariffs For the purpose of this study the consumer price is modelled as
generation cost plus transmission cost plus distribution costs plus Local
Authority surcharge. What is added in this study, which is not yet
transparently reflected in consumer tariffs, are the cost of DSM initiatives
and the un-recovered fixed cost of generators.
The cost of demand side management initiatives undertaken and paid for
by NamPower is recovered through the tariff but is not accounted for
separately. The model calculates the cost of the proposed demand side
management initiatives on a per kWh basis and adds this to the tariff
seen by end consumers.
Page 56 of 222
Parameter Notes
Unserved The model assumes that demand which is not served in one year still
demand exists in the future years, i.e. no major energy carrier switching takes
place and no significant slump in electricity demand occurs due to power
outages.
Demand Growth The model assumes transmission load to grow by 75MW per year for
Steps three years initially (due to new uranium mining and treatment facilities)
(NamPower, pers. comm., October 2007 ) and then by 10MW per year
(around 2.5% of projected transmission load in 2012) to cater for
unspecified future transmission load additions
Load curve The model assumes that the transmission customers have a flat load
curve over the 24-hour cycle, i.e. all variation in the load curve is
projected onto the distribution customers. The majority of transmission
load is based on continuous operations with little variation over the 24
hour cycle and thus this is considered to be a reasonable approximation.
Eskom supply As a base case it has been assumed that Eskom supply will be severely
restricted until 2013, especially during peak times.
Zimbabwe The imports from Zimbabwe have been included in all scenarios as per
imports the information provided by NamPower on the basis of a five year
agreement, i.e. commencing in 2008, going to full capacity in 2009 and
ending in 2014/15.
Zambia, Angola Imports from and exports to Zambia for supply to the Caprivi Region, and
& Botswana exports on a small scale to Angola and Botswana have not been included
imports and in the modelling due to their limited impact.
exports
Eskom price The import price from Eskom is modelled to be time and season
path differentiated to reflect the actual contract closely. However it has been
assumed that over a ten-year term the energy base price will rise to
reflect the estimated coal station fuel cost for a new efficient and clean
coal plant built in South Africa at a coal mine. This fuel price estimate has
been derived using the current coal price seen by NamPower at van Eck
and adjusting this for a more efficient plant and stripping out the transport
costs. This is viewed as a conservative approach since Eskom is likely to
get a better price on a long term contract than achieved by NamPower
currently on short-term purchases. The resulting projected average cost
exceeds the South African long run marginal cost (LRMC) (NER 2004)
(inflation adjusted to 2007) by around 20%.
Local coal power stations will not be able to compete with Eskom coal
stations in the long run because of the high transport cost for coal which
for van Eck doubles the coal price from the mine. This is taken to imply
that coal stations built in Namibia should be limited in size, and should be
designed for mid merit operation. The only exception to this would be
considered under the assumption the imports from Eskom will not be
available for a rather long time, basically to the end of the model horizon
of 20 years.
Page 57 of 222
Parameter Notes
Van Eck It has been assumed that van Eck should be retired as soon as sufficient
replacement capacity is available because of its age and low efficiency.
In most scenarios analysed in this study this does not happen earlier than
2012.
Paratus It has been assumed that this power station will continue operating as
peaking plant.
Baynes It has been modelled as a base load station with a 360MW output in this
study. The capacity factor is assumed at approximately 90%. In some
scenarios it is expanded by additional generators, which would drive it
more into the mid merit range of operation. The project team takes
cognisance that the more recent thinking about this power station tends
towards a larger capacity plant with a mid merit to peaking run profile.
However this information reached the team too late to be fully
implemented in the model
Ruacana Ruacana has been modelled as mid-merit and peaking plant according
the way it is utilised for the majority of the year.
The fourth turbine at Ruacana hydro-electric power station has not been
factored into the technical model because the announcement of it
impeding commissioning was made after the analysis was completed.
Exchange rate All costs stated in foreign currency are adjusted for inflation and then
converted to Namibian dollars at current exchange rates (7 for USD and
10 for Euro).
Oil Price The base assumption escalates the oil price by 3% per annum from the
2007 level. The high oil price sensitivity escalates oil to 200% of 2007
level over 5 years and escalates at 3% p.a. thereafter. The low oil price
sensitivity reduces oil prices by 50% of 2007 level over five years and
keeps it constant thereafter. All computations are in real terms.
Coal Price The base assumption keeps the coal price constant. The high coal cost
sensitivity escalates coal price to 200% of 2007 level over 5 years and
keeps it constant thereafter. The low coal price sensitivity keeps the coal
price constant. All computations are in real terms.
Table 3-3: Technical Model Key Assumptions
3.1.3.3 Limitations
Table 3-4 lists the known limitations of the dispatch model. Most models seeking to simulate
complex systems make certain assumptions and simplifications of reality. Simplifications are
also made to reflect the lack of accurate data to underpin certain aspects of reality. This
model is no exception and makes some important simplifications that need to be noted to
understand in which way the model results may not reflect reality closely. The most important
of these are listed below.
Page 58 of 222
Parameter Notes
Generator The generator availability curve is modelled as static over the years, i.e.
availability curve a generator’s hourly production curve does not change over the years.
This is not significant for generators that will remain the same over the
years. It does mean that Ruacana is modelled on mid merit to peaking
for all years while its operations may change in future, especially if a
fourth generator is installed.
Load curve The load curve is modelled in three “seasons” instead of following the
more common approach of having weekday, Saturday and Sunday
curves for a low and high season (i.e. three curves instead of six
curves). The effect hereof is that the dispatch resolution is reduced
because statistical simplification is made. Since the primary purpose of
the study is to compare different generation mixes with each other (as
opposed to accurately forecasting the cost of any particular generation
mix) and the simplification affects all mixes equally it is not deemed
significant for the purposes of this study. This potentially reduces the
accuracy of the model’s outputs for purposes other than this study.
Generator data The generator cost data (both capital and operating) was derived
mostly from public sources, and has not been adjusted in detail for
Namibian conditions and very recent international cost developments.
In general the numbers used are considered conservative, i.e. costs
used are at the upper end of data found during research to compensate
for recent price increases due to rising international demand for
generators.
No economies of scale was reflected in the pricing of small scale

renewable generators such as wind, biomass to electricity and
concentrating solar plants. Costing of these generators is thus
conservative.
Price smoothing In reality the regulator would ensure that large changes in end
consumer price are phased in over time. In the model this smoothing
functionality has not been built in, and prices are reflected as they are
calculated. This results in some large step price increase that would
normally be smoothed over time. The impact of this is not significant
since the study considers the entire generator supply timeframe and
does not consider individual year-to-year changes.
Table 3-4: Technical Model Limitations
It is believed that while the above simplifications are significant they do not invalidate the
modelling and are not likely to distort the key outcomes.
3.1.3.4 Unserved Energy
As mentioned above the model follows the following sequence. The model takes a specified
mix of generators, demand side management options and load growth and solves the
dispatch equation until supply and demand balance. If it cannot match demand with existing
supply then there is a residual called unserved energy.
Page 59 of 222
The model deals with unserved demand as the sum of two calculations. The first part of the
calculation is through a generator called “load shedding” with a capacity of 25MW22 and a
price of NAD 5 per kWh that is dispatched like any other generator in all scenarios. The
second way is to report on demand that could not be met by the available generators on an
hourly basis per year. By adding the two numbers together the model computes the total
unserved demand per year.
3.2 DESCRIPTION OF GENERATION SCENARIOS

This section describes the various generation and demand side management options, and
explains how they were combined into a variety of scenarios. Essentially each scenario
seeks to describe a policy driver option. The scenarios modelled are listed in Table 3-5.
CCGT/Solar Concentrating
Import Zimbabwe (Coal)
Import Eskom (Coal)
Concentrating Solar
Nuclear (PMBR)
Ripple Control
Invader Bush
CCGT Gas
New Coal
Ruacana
Solar PV
van Eck
Paratus
Baines
SWH
Wind
CFL
1 Base Case 1 1 1 1 1 1 1 1 1
2 No new local generation 1 1 1 1 1
3 Price minimised 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
4 Maximum supply diversity 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
5 Generation Self sufficiency 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
6 Energy Imports minimised 1 1 1 1 1 1 1 1 1 1 1 1 1 1
7 Maximum renewable 1 1 1 1 1 1 1 1 1 1 1
8 All options considered 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
Table 3-5: Scenario Definitions
The scenarios depict how the generation mix for Namibia could be constructed under various
policy imperatives. These scenarios and policy imperatives are described in detail below.
Deciding which generation options to consider and which to give priority depends on what
energy policy is being pursued. It is a key purpose of this project to demonstrate that various
policy options could be pursued and to show what results might be achieved through such
pursuit. The results are both technical and economic in nature.
For all scenarios the primary objective is to meet electricity demand by building generators or
procuring capacity within the policy imperative. The secondary objective is to minimise the
overall average generation price while still meeting the first objective. Furthermore,
generators are dispatched according to their lead time constraints and within reasonable
capacity limits for intermittent generators like wind power.
22
Windhoek and Walvis Bay have ripple control systems for electric water heaters which together can shed around 25MW. Co-
ordination of these systems to meet national load shedding needs has not been formally established. The cost of NAD5/kWh for
this “generator” was arbitrarily chosen to exceed the cost of all other generator options to ensure that this option is only
dispatched if all other sources are exhausted. It demonstrates what could be achieved through co-ordinated use of the existing
ripple systems.
Page 60 of 222
3.2.1 Scenario 1: Base Case

This scenario depicts a policy of using proven and conventional, large scale, centralised
generation options. It was modelled as follows:
Generation Capacity Installed
1600
1400
1200
1000
MW
800
600
400
200
0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
Gas/Solar Hybrid - - - - - - - - - - - - - - - - - - - - -
Zim Import - - - 40 150 150 150 150 150 - - - - - - - - - - - -
Solar PV - - - - - - - - - - - - - - - - - - - - -
"Clean" Coal - - - - - - 350 350 350 350 350 350 350 350 350 350 350 350 350 350 350
Nuclear - - - - - - - - - - - - - - - - - - - - -
Import Eskom 200 200 200 150 100 100 100 150 250 450 450 450 450 450 450 450 450 450 450 450 450
Biomass - - - - - - - - - - - - - - - - - - - - -
Concentrating Solar - - - - - - - - - - - - - - - - - - - - -
Wind - - - - - - - - - - - - - - - - - - - - -
Baynes - - - - - - - - - - - - 360 360 360 360 360 360 360 360 360
CCGT (Kudu) - - - - - - - - - - - - - - - - - - - - -
Paratus 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24
van Eck 120 120 120 120 120 120 120 120 120 - - - - - - - - - - - -
Ruacana 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243
Years
Figure 3-8: ‘Base case’ generation mix
In this scenario only two power plants are built, namely a 350MW coal plant commissioned in
2011 and 360MW of additional hydro on the Kunene River in the form of Baynes in 2017.
This is augmented by increasing imports from Eskom, based on the assumption that Eskom
will be able to supply increasing capacity again starting from 2012. Van Eck power station is
retired as soon as replacement capacity can be obtained from Eskom. This scenario does
not implement demand side management except the CFL campaign.
Page 61 of 222
3.2.2 Scenario 2: No New Local Generation

This scenario centres on implementing no new generation within Namibia and relying
primarily on electricity imports. It could also be seen as depicting a scenario where no
investment decision is made even though there are possible options. It is made up of the
generation options illustrated in Figure 3-9.
1600
1400
1200
1000
MW
800
600
400
200
0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
Gas/Solar Hybrid - - - - - - - - - - - - - - - - - - - - -
Zim Import - - - 40 150 150 150 150 150 - - - - - - - - - - - -
Solar PV - - - - - - - - - - - - - - - - - - - - -
"Clean" Coal - - - - - - - - - - - - - - - - - - - - -
Nuclear - - - - - - - - - - - - - - - - - - - - -
Import Eskom 200 200 200 150 100 100 100 150 250 750 850 850 850 950 950 950 1 050 1 050 1 150 1 150 1 150
Biomass - - - - - - - - - - - - - - - - - - - - -
Concentrating Solar - - - - - - - - - - - - - - - - - - - - -
Wind - - - - - - - - - - - - - - - - - - - - -
Baynes - - - - - - - - - - - - - - - - - - - - -
CCGT (Kudu) - - - - - - - - - - - - - - - - - - - - -
Paratus 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24
van Eck 120 120 120 120 120 120 120 120 120 - - - - - - - - - - - -
Ruacana 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243
Years
Figure 3-9: ‘No new local generation’ generation mix
This scenario relies increasingly on imports over time. During the time before Eskom imports
become available again it produces severe amounts of unserved demand, meaning that wide
spread load shedding cannot be avoided. Imports from the Southern African Power Pool
(SAPP) may become possible by that time, but at this stage it is not clear whether and when
the planned investments in SAPP member countries will actually come into production. It is
also not clear what portion of new generators might in future be available to Namibia for
imports. In the consultants’ assessment it is quite possible that no additional import sources
may be available due to delays in new generation projects and continued demand growth in
the region.
This scenario also does not implement demand side management measures to reduce
demand, except the CFL campaign.
Page 62 of 222
3.2.3 Scenario 3: Price minimised

This scenario centred on providing the cheapest electricity possible while still avoiding large
scale load shedding: It is illustrated in Figure 3-10.
1600
1400
1200
1000
MW
800
600
400
200
0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
Gas/Solar Hybrid - - - - - - - - - - - - - - - - - - - - -
Zim Import - - - 40 150 150 150 150 150 - - - - - - - - - - - -
Solar PV - - - - - - - - - - - - - - - - - - - - -
"Clean" Coal - - - - - - 350 350 350 350 350 350 350 350 350 350 350 350 350 350 350
Nuclear - - - - - - - - - - - - - - - - - - - - -
Import Eskom 200 200 200 150 100 100 100 150 250 350 350 350 350 350 350 350 350 350 350 350 350
Biomass - - - - 20 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60
Concentrating Solar - - - - - 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50
Wind - - - - - - - - - - - - - - - - - - - - -
Baynes - - - - - - - - - - - - 360 360 360 360 360 360 360 360 360
CCGT (Kudu) - - - - - - - - - - - - - - - - - - - - -
Paratus 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24
van Eck 120 120 120 120 120 120 - - - - - - - - - - - - - - -
Years
Figure 3-10: ‘Price minimised’ generation mix
A biomass power plant is built in 2009 and expanded in 2010. A small coal plant in 2011 and
a small concentrating solar plant is built in 2010. This happens to cover the time before
Eskom capacity again becomes available. Baynes hydro opens in 2017 to harness additional
hydro potential of the Kunene River.
Page 63 of 222
3.2.4 Scenario 4: Maximum Supply Diversity

This scenario centred on the analysis of diversifying the generation options as much as
possible. This is intended to minimise the risk exposure to any one particular energy carrier
or plant technology by building more generators of different types and smaller sizes. The
composition of this scenario is illustrated in Figure 3-11.
1800
1600
1400
1200
1000
MW
800
600
400
200
0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
Gas/Solar Hybrid - - - - - - - - - - - - - - - - - - - - -
Zim Import - - - 40 150 150 150 150 150 - - - - - - - - - - - -
Solar PV - - - - - - - - - - - - - - - - - - - - -
"Clean" Coal - - - - - - 175 175 175 350 350 350 350 350 350 350 350 350 350 350 350
Nuclear - - - - - - - - - - - - - - - - - - - - -
Import Eskom 200 200 200 150 100 100 100 150 250 250 250 250 250 250 250 250 250 250 250 250 250
Biomass - - - - 20 40 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60
Wind - - - - 30 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60
Baynes - - - - - - - - - - - - - - - - - - - - -
CCGT (Kudu) - - - - - - - - - - - - 400 400 400 400 400 400 400 400 400
Paratus 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24
van Eck 120 120 120 120 120 120 120 120 - - - - - - - - - - - - -
Ruacana 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243
Years
Figure 3-11: ‘Maximum supply diversity’ generation mix
In this scenario a coal fired station opens in 2011; biomass production starts in 2009 and is
scaled up over the next three years; concentration solar starts in 2009 and is doubled in
2013; wind power takes off in 2009; and CCGT from Kudu commences in 2017. Notably this
is the only scenario where Baynes hydro is not built. This is based on avoiding large risk
exposure to water availability in the Kunene River to which Ruacana is already exposed.
Solar PV is not build because of cheaper alternatives nor is nuclear because of the very long
lead time.
This is the first of the scenarios to include demand side management. The three demand
side management measures discussed in section 3.1.2.2 are implemented commencing in
2008.
Page 64 of 222
3.2.5 Scenario 5: Generation Self Sufficiency

This scenario centres on generating as much of the electricity within Namibia as possible and
to minimise the dependence on importing electricity. Imported fuels are part of the mix as are
the imported components of capital and operating expenditure on both the supply and
demand side. The composition of this scenario is illustrated in Figure 3-12
1600
1400
1200
1000
MW
800
600
400
200
0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
Gas/Solar Hybrid - - - - - - - - - - - - - - - - - - - - -
Zim Import - - - 40 150 150 150 150 150 - - - - - - - - - - - -
Solar PV - - - - - - - - - - - - - - - - - - - - -
"Clean" Coal - - - - - - 175 175 175 175 175 175 175 175 175 175 175 175 175 175 175
Nuclear - - - - - - - - - - - - - - - - - - - - -
Import Eskom 200 200 200 150 100 100 100 100 50 50 50 50 50 50 50 50 50 50 50 50 50
Biomass - - - - 20 40 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60
Wind - - - - - 30 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60
Baynes - - - - - - - - - - - - 360 360 360 360 360 360 360 360 360
CCGT (Kudu) - - - - - - - 400 400 400 400 400 400 400 400 400 400 400 400 400 400
Paratus 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24
van Eck 120 120 120 120 120 120 120 - - - - - - - - - - - - - -
Ruacana 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243
Years
Figure 3-12: ‘Generation self sufficiency’ generation mix
This scenario calls for the construction of a small coal plant which, because of the short lead
time opens in 2011; a gas power plant which opens in 2012 as well as the Baynes hydro
plant that would come on line in 2017. These are augmented with smaller renewable plant,
which offer short implementation times and contribute to long term supply stability and
diversity. The demand side management options are implemented since they contribute
significantly to reducing the need for imported electricity and at the same time delaying the
need for investment in local power plant.
Page 65 of 222
3.2.6 Scenario 6: Energy Imports Minimised

This scenario is similar to scenario 5 but apart from just limiting electricity imports it aims at
minimising all imports. The composition of this scenario is illustrated in Figure 3-13
1600
1400
1200
1000
MW
800
600
400
200
0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
Gas/Solar Hybrid - - - - - - - - - - - - - - - - - - - - -
Zim Import - - - 40 150 150 150 150 150 - - - - - - - - - - - -
Solar PV - - - - - - - - - - - - - - - - - - - - -
"Clean" Coal - - - - - - 175 175 175 175 175 175 175 175 175 175 175 175 175 175 175
Nuclear - - - - - - - - - - - - - - - - - - - - -
Import Eskom 200 200 200 150 100 100 100 100 50 50 50 50 50 50 50 50 50 50 50 50 50
Biomass - - - - 20 40 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60
Wind - - - - - 30 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60
Baynes - - - - - - - - - - - - 360 360 360 360 360 360 360 360 360
CCGT (Kudu) - - - - - - - 400 400 400 400 400 400 400 400 400 400 400 400 400 400
Paratus 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24
van Eck 120 120 120 120 120 120 120 - - - - - - - - - - - - - -
Ruacana 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243
Years
Figure 3-13: ‘Energy imports minimised’ generation mix
This generation mix consists of biomass starting in 2009 and scaling up over the next two
years; concentrating solar and wind both coming on line in 2010 and Kudu gas coming on
line in 2012. However after analysis the modelling process revealed that it would be very
difficult to meet electricity demand without some imports. Thus it is necessary to build a small
coal station as base load support until the gas power station can be completed. The result is
a scenario that differs from the previous one only in some relatively minor timing details. As
for the previous scenario demand side management options are fully implemented.
Page 66 of 222
3.2.7 Scenario 7: Maximum Renewable Energy Options

This scenario centred on using as many renewable energy generation options as possible
and seeks, as far as possible, to avoid fossil fuel generation. The composition of this
scenario is illustrated in Figure 3-12.
1600
1400
1200
1000
MW
800
600
400
200
0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
Gas/Solar Hybrid - - - - - - - - - - - - - - - - - - - - -
Zim Import - - - 40 150 150 150 150 150 - - - - - - - - - - - -
Solar PV - - - - - 20 40 40 40 40 40 40 40 40 40 40 40 40 40 40 40
"Clean" Coal - - - - - - - - - - - - - - - - - - - - -
Nuclear - - - - - - - - - - - - - - - - - - - - -
Import Eskom 200 200 200 150 100 100 100 100 100 250 250 250 250 250 250 250 250 250 250 250 250
Biomass - - - - 20 60 120 120 120 120 120 120 120 120 120 120 120 120 120 120 120
Wind - - - - 30 60 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90
Baynes - - - - - - - - - - - - 360 360 360 360 360 360 360 450 450
CCGT (Kudu) - - - - - - - - - - - - - - - - - - - - -
Paratus 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24
van Eck 120 120 120 120 120 120 120 120 120 120 120 120 - - - - - - - - -
Ruacana 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243
Years
Figure 3-14: ‘Maximum renewable’ generation mix
The resulting generation portfolio relies heavily on concentrating solar (with storage) as well
as biomass and wind, while also building Baynes hydro and continuing reliance on imports
from Eskom. This is one of the few scenarios where even a solar PV plant is built despite its
high price. Van Eck is retired later than in the other scenarios. Demand side management is
implemented since it would form part of a policy based on renewable resources. No new
local fossil fuel plants are built.
Page 67 of 222
3.2.8 Scenario 8: ‘All options considered’

This scenario was compiled because it mixes a strong renewable focus with a strong
concern for supply diversity to minimise risk exposure to energy carriers and technology
types. It is intended as the main counterpoint to the ‘base case’ scenario, and also seeks to
bring in strong decentralised elements.
The generation mix is illustrated in Figure 3-15.
1400
1200
1000
800
MW
600
400
200
0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
Gas/Solar Hybrid - - - - - - - - - - - - - - - - - - - - -
Zim Import - - - 40 150 150 150 150 150 - - - - - - - - - - - -
Solar PV - - - - - 5 10 20 20 20 20 20 20 20 20 20 20 20 20 20 20
"Clean" Coal - - - - - - 175 175 175 175 175 175 175 175 175 175 175 175 175 175 175
Nuclear - - - - - - - - - - - - - - - - - - - - -
Import Eskom 200 200 200 150 100 100 100 100 150 350 350 350 100 100 100 100 100 100 100 100 100
Biomass - - - - 20 40 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60
Wind - - - - 30 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60 60
Baynes - - - - - - - - - - - - 360 360 360 360 450 450 450 450 450
CCGT (Kudu) - - - - - - - - - - - - - - - - - - - - -
Paratus 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24
van Eck 120 120 120 120 120 120 120 120 120 - - - - - - - - - - - -
Ruacana 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243 243
Years
Figure 3-15: ‘All options considered’ generation mix
The generation mix relies on a small coal plant, biomass and Baynes hydro as the backbone
for 24 hour electricity supply. This is complemented by an array of diverse renewable
generators, each limited in size to ensure grid stability and cost effective grid integration by
avoiding grid issues that would arise for larger versions of the same plant type. This scenario
also implements demand side management.
The scenario has excluded the building of nuclear or gas plants.
Page 68 of 222
3.3 ECONOMIC METHODOLOGY

This section describes the economic modelling that was undertaken to determine the
potential economic impact of the various electricity generation options. There are a variety of
different types of economic analyses. These include cost benefit analysis, micro-economic
costs and benefits, and macro-economic analysis.
An economic analysis that is used to make decisions that affect the entire society of a
country requires a cost benefit analysis. The final form of analysis, the micro-economic
analysis, is used to determine the impacts at a more localised level. This was not called for in
the Terms of Reference and was also not deemed necessary for this study.
The section starts with a general description of the methodological approaches used to
perform the cost benefit analysis and determine the macro-economic impact. This is followed
by a description of the data used in the model.
3.3.1 Cost Benefit Analysis

Cost Benefit Analysis (CBA) treats the national economy as an entity in and of itself. It
assumes that what is demonstrably good for the economy as a whole is a reasonable
approximation of what would be good for the majority of the people living and working in that
area.
When large infrastructure investments are contemplated, decision makers need to know
what impact the new, or improved, infrastructure would have on the economy as a whole and
hence how much benefit can be assumed to accrue to the people of the country. The main
linkage is via the changes in electricity costs, emissions and the amount of unserved
electricity demand that occur when the different generation options are compared to one
another.
The economic cost benefit analysis has taken a number of costs and benefits into account.
These are:
• the capital cost of constructing and/or upgrading alternative generation options,
• the costs of operating and maintaining the power plants,
• fuel costs (which could include the cost of importing electricity),
• productivity changes due to different electricity prices,
• the cost of unserved demand,
• carbon credits.
The outcome of this analysis is the reporting of a benefit cost ratio (BCR). All costs and
benefits are reduced to a present day value by using a social discount rate of 8% as
specified by the South African National Treasury (Conningarth 2007). The BCR measures
the changes in benefits and costs that would result from an investment. BCRs are typically
used when there are many competing alternatives and projects need to be funded from a
limited set of resources.
Page 69 of 222
If the evaluated benefits of a project are indeed greater than the overall project costs then the
BCR ratio would be greater than 1. A BCR greater than 1 indicates that the completed
project would constitute an economic asset; a BCR less than 1 implies that the project would
be an economic liability. The higher the BCR the less risk there is that the proposed
investment could turn out to be less than beneficial economically. Low BCRs, even if greater
than 1, provide a warning that a project could be risky and may turn out to be an economic
liability instead of an asset.
A high BCR is usually a good indicator that it would be possible to raise finance to implement
a project. In the case of a private sector investment a high BCR would be an important
motivator of the business case presented to funders. If it is a public infrastructure project, a
high BCR should give confidence that it is worth funding the project directly from its Treasury
or, alternatively, to make suitable institutional arrangements for the involvement of the private
sector in the project’s funding.
The cost benefit analysis of the electricity generation options is to inform policy makers in
Namibia of the scale of economic benefits arising from the different generation options and to
provide a means of comparison between them.
The cost benefit analysis focuses purely on direct costs and benefits and does not take any
indirect costs and benefits into account. Indirect costs and benefits would include those costs
and benefits obtained through multiplier effects. For example, the construction of a new
power plant would have spin off effects for the construction industry and the building
materials supply industry. These, in turn, would have backward linkages with other
commodity suppliers and retail industries. This is part of the macroeconomic analysis and is
described in detail in section 3.3.2.1.
The CBA was developed based on best practice and in consultation with the guidelines of
the Manual for Cost Benefit Analysis in South Africa (Conningarth, 2007). The CBA makes
use of the NPV method of discounting all costs and benefits as a means for comparing the
various options. The analysis has been conducted from a country wide, i.e. Namibian,
perspective.
To explore the economic value of any investment programme, the analyst must identify the
costs of and the benefits to investment in the project when compared to the situation that
would have prevailed if no such investment had been made. This latter situation is commonly
referred to as the ‘base case’, do nothing or do minimum situation. In this analysis the
various generation scenarios are compared to the “No New Local Generation” scenario.
Such an approach allows the determination of the least-cost solution, as well as the
identification of the benefits in such a way that they can be compared across the economy to
aid the rational and efficient allocation of resources.
3.3.2 Macro-Economic Analysis

The size of a national or regional economy is measured in terms of the sum total of all
economic activities taking place within the area concerned, both in the public and private
sectors. For countries like Namibia, this necessarily includes measures of informal sector
activity as well. The name given to the measure of the size of the economy is Gross
Domestic Product (GDP). Sometimes this is referred to as Gross Value Added (GVA) which
is a slight variation on GDP. The unit of measurement is the national currency.
Underlying the measurement of GDP is the understanding that all economic activity is
dependent on the physical and institutional support systems that enable an economy to
operate effectively. These include the various levels of governmental structure, the legal
system, and the administrative, financial and educational infrastructure in the country. In
Page 70 of 222
terms of physical infrastructure, all economic activity depends on water and energy supply,
telecommunication, and transport infrastructure. Without all of these systems being in place
the economy could not operate.
3.3.2.1 Macro-Economic Modelling
While there are a number of different types of macro-economic effects, the two most
important are contribution to GDP and the creation of jobs. The importance of job creation is
obvious. Increases in GDP are synonymous with increases in peoples’ economic standards
of living. Increased GDP – i.e. increased production – is experienced in the form of more
jobs, higher wages and reduced economic hardship.
The actual task of calculating the macro-economic impact of the electricity generation options
demands a detailed and multifaceted approach, not least because of the so-called multiplier
effects. It is well recognised that the simple act of spending – building a new power station,
for example, - leads to other economic effects. Demand for steel and cement can lead to
increased production in those industries. Increased demand for steel and cement, in turn,
leads to increased demand for mining output which uses wood, water, electricity and so on.
These are the so-called multiplier effects.
While this process unfolds, each industry employs people and pays wages. Employees, in
turn, spend their wages and cause further multiplier effects through the economy. Measuring
this is further complicated by the fact that different industries demand different types of skills.
This leads to different wage structures across the various industries. People earning different
wages have different spending patterns. Thus, the change in overall spending patterns is
dependent on which industries are affected.
Therefore, the macro-economic estimates that are made in this report relate directly to the
actual cost of developing the infrastructure, as well as the operating cost and productivity
gains that business would experience. So included in the macro-economic calculations are
all the backward economic linkages for construction and operations and the forward
economic linkages where construction workers and others spend their salaries.
3.3.2.2 Namibian Social Accounting Matrix
The macro-economic modelling is based on a social accounting matrix (SAM) of the

Namibian economy that was developed by the Namibian Economic Policy Research Unit
(NEPRU) in association with the University of Columbia (NEPRU 2004). A SAM is an
extension of an input/ output table. An input/output table gives the relationship between the
various sectors in an economy from a demand and supply perspective. For example, if there
is an increased demand for agricultural output then the agricultural sector would use
additional inputs in the form of seed, fertiliser, fuel, etc. Hence input/output tables are used to
track how changes in demand filter through the economy and determine the overall impact
on gross domestic product (GDP), job creation, etc. A SAM takes this further by also showing
which groups in society receive the increased wages, profits, interest and rents.
Typically one would compare a SAM to an input/output table. In the Namibian case the latter
is not available and therefore the macro-economic analysis is based fully on the SAM. The
SAM has therefore been interrogated to determine the degree to which it can be relied on to
give correct answers to the analysis that is being imposed on it.
In their description of the SAM, NEPRU states that the 1993/94 Namibian Household Income
and Expenditure Survey were used for the development of the matrix and that it would be
updated as soon as the results of the 2003/04 survey are available. This does not appear to
have yet been done and our efforts to secure the updated version were unproductive.
Page 71 of 222
Consequently a number of tests were performed on the SAM to determine if it is suitable for
the purposes of this analysis.
The first test is taken from NEPRU’s discussion document on the SAM. In table 4 of the
document (pg 31) a comparison is made between the sizes of the sub-sectors in the
Namibian economy as published in the National Accounts and as determined from the SAM.
These are NAD 30 101 million from the National Accounts and NAD 31 447 million from the
SAM for 2002. The difference of 4.5% is considered to be accurate enough for the purposes
of this analysis.
The second test was to compare the GDP multipliers for South Africa against those of
Namibia. Both sets of multipliers were calculated from the respective country’s SAM’s. A
comparison of selected economic sector multipliers is shown in Table 3-6.
Namibia South Africa

Mining 2.162 Gold mining 2.699
Other mining 1.912
Meat Processing 1.856 Food processing 1.940
Fish Processing 2.555
Bev&Other Food Prod 1.242 Beverages 1.952
Textiles 0.392 Textiles 1.696
Light Manufacturing 0.402 Other manufacturing 1.733
Petroleum Products 0.150 Chemicals 1.593
Heavy Manufacturing 0.360 Machinery 0.947
Electricity 2.071 Electricity 2.276
Water 2.285
Construction 1.014 Building construction 2.235
Civil engineering 2.287
Trade, Repairs 2.461 Trade 2.596
Hotels and Restaurants 2.174 Accommodation & catering 1.669
Transport 1.135 Transport 2.297
Communication 1.712 Communication 2.469
Finance and Insurance 2.095 Financial services 2.329
MktRealEst + Bus Services 1.912
Other Private Services 2.269
Government Services 2.420 Community services 3.116
Average 1.642 Average 1.929
Table 3-6: Comparison of GDP multipliers
The first comparison that is made is for specific sectors. The mining sector for Namibia has a
multiplier of 2.162, which compares well with the South African multipliers of 2.699 for gold
mining, and 1.912 for other types of mining. The food processing sector also compares well,
with the Namibian multipliers for meat and fish processing at 1.856 and 2.555 respectively,
compared to a multiplier of 1.940 for South African food processing.
Textiles, light manufacturing, petroleum products and heavy manufacturing all have GDP
multipliers of less than 1 for the Namibian case, because these sectors are heavily
dependent on imports. In the South African case these multipliers are all higher than their
Namibian counterparts. It is interesting to note that the Chemicals sector in South Africa has
a multiplier of 1.593, which is much higher than the Namibian multiplier of 0.150. This would
be due to the fact that South Africa refines its own crude oil and also due to SASOL and
PetroSA. For South Africa the machinery sector has a multiplier of less than 1 due to its
dependence on imported components.
Page 72 of 222
The electricity, trade & repairs and financial services sectors in Namibia have similar GDP
multipliers to their respective South African sectors, albeit slightly lower. On the other hand
the hotels and restaurants sector in Namibia is higher than the South African multiplier.
Finally, the transport, communications and government services sectors are considerably
lower than the respective South African multipliers, although the pattern of the government
services multiplier generally being higher than most other sectors is still observed.
In general the Namibian multipliers are lower than the South African multipliers, as can be
seen by the value of the average multiplier at the bottom of Table 3-6. The average multiplier
for Namibia is 1.642 compared to 1.929 for South Africa. This can probably be explained by
Namibia being more import dependant than South Africa in certain sectors.
The third check that was performed was to determine the number of jobs that are projected
from the SAM. The SAM gives the total salaries and wages payment for each sub-sector.
These salaries and wages are further disaggregated into skilled, semi-skilled and mixed
income23. By using average annual salaries for these three categories of NAD 125 000 for
skilled workers, NAD 24 000 for unskilled workers and NAD 75 000 for mixed incomes, a
total of 334 287 formal jobs is calculated. This number compares well with the 340 480
estimated in the 2001 Population and Housing Census (National Planning Commission,
Namibia) the 368 461 in the 2004 Labour Force Survey (Namibia Ministry of Labour).
Based on these arguments we are confident that the SAM is accurate enough for the
purposes of this analysis. This is particularly the case because the intention is not to decimal
point the analysis but rather to understand the likely trends that can be expected from
specific interventions. To the extent that there are potential errors in the SAM these errors
will be carried forward to all the scenarios and will not unduly undermine the relative
differences between the various scenarios.
3.3.2.3 Data Description and Macro-Economic Methodology
The macro economic model uses four data sets. These are:
1. Capital costs
2. Operating costs
3. Fuel costs
4. Changes in the price of electricity
All the analysis is done in real terms, i.e. inflation is eliminated.
3.3.2.3.1 Capital Costs
The analysis of the impact of capital expenditure on the economy tracks the actual capital
construction costs and where those funds flow through the economy. It also monitors the
degree to which capital is imported and the impact on the balance of trade.
23
The earnings of the self-employed, such as farmers, are called ‘mixed income’ because the surplus of sales revenue over
input costs includes both a payment for their own labour as well as a payment for capital inputs. It is difficult to impute the labour
cost, so the national accounts simply leaves the surplus as mixed income (NEPRU Working Paper No. 97, pg 8).
Page 73 of 222
Three steps are required to measure the overall macro-economic impact of the construction
costs:
• First, to determine the relative proportions of plant, material and imports for the
capital expenditure for each generation option.
• Second, to assign each item of material and plant to the appropriate economic sector
in the Namibian SAM.
• Finally, all the SAM coded items are brought together. The total multiplier effect of the
capital expenditure is calculated as the aggregate product of the SAM coded
spending on plant and material multiplied through the national multipliers. The
national multipliers are determined from the Namibian SAM.
3.3.2.3.2 Operating Costs
The analysis of operating costs monitors where these funds go through into the economy
and the overall impact on the balance of trade. This follows a similar logic as used for the
analysis of capital costs. This analysis excludes the cost of fuel, which is treated separately
and discussed in the next section.
Four steps are required to measure the overall macro economic impact of the operating
costs of each generation option.
• First, to determine the relative proportions of profit, labour, plant, material and imports
for each generation option.
• Second, to assign each item of material and plant to an appropriate economic sector
in the Namibian SAM.
• Third, to decompose labour and profit into skilled, unskilled and mixed income
categories and apportion the total wages and profits to each income category.
Following this, estimates of expenditure patterns by income category are used to
determine total spending patterns24.
• Finally, all the items in the SAM coded list are brought together. The total multiplier
effect of the operating costs is calculated as the aggregate product of the SAM coded
spending on plant and material, as well as the SAM coded spending by workers
multiplied through the national multipliers. The national multipliers are determined
from the Namibian SAM.
3.3.2.3.3 Fuel Costs
The analysis of fuel costs of the various supply options monitors where these funds flow
through the economy and the overall impact on the balance of payments.
Three steps are required to measure the overall economic impact of the fuel costs of each
generation option.
24
Use is made of the spending patterns in South Africa for corresponding income categories. This was done because we did
not have access to a Namibian income and expenditure survey. Anecdotal evidence suggests that there are not major
differences between Namibian and South African spending patterns.
Page 74 of 222
• First, to determine the relative proportions of material and imports, if any, for each
generation option.
• Second, to assign each item of material to an appropriate economic sector in the

Namibian SAM.
• Third, to bring all the items in the SAM coded list together. The total multiplier effect
of the fuel costs is calculated as the aggregate product of the SAM coded spending
on plant and material. The national multipliers are determined from the Namibian
SAM.
3.3.2.3.4 Changes in the Price of Electricity and Power Outages
The economic impact of changes in the cost of electricity is a key component of the macro-
economic analysis. The economic model takes the electricity price from the dispatch model
and tracks its impact on the various sectors of the economy. A number of variables are taken
into account in the analysis. These are:
• The degree to which sectors are price takers, i.e. sectors for which the price of their
output is fixed by world markets and they must accept that price. There are three
such sectors in the Namibian economy. The most obvious of these is the mining
sector where commodity prices are set by world markets and not by the mines. To a
slightly lesser extent the fishing and agricultural sectors are also price takers. What
this means therefore is that if there is any fall in the price of electricity then profits will
increase and any price increases will result in lower profits or even losses.
• Price elasticity of demand. The price elasticity of demand is a measure of the degree
to which people respond to a price change. Therefore for those sectors that can pass
on changes in electricity prices, the change in the amount that people will now be
willing to buy depends on price elasticity. An example of a good that is relatively
inelastic is petrol whereas a relatively elastic good would be a restaurant meal. The
elasticities that have been used in the analysis are given in Table 3-7.
• Projected economic growth rates. There was a need to have a baseline potential
economic growth for the various sectors of the Namibian economy. This was in order
to project the degree to which a change in the price of electricity would increase or
decrease that potential growth. We were unable to find any public domain projections
for the Namibian economy. As a result we made use of projections that have been
made for the South African economy by ABSA – a large South African bank. It is
recognised that this is a second best solution but is a move in the right direction. The
growth rates that have been used are given in Table 3-7. In order to address potential
problems with these growth estimates they are subject to a variety of sensitivity tests
in section 4.2.5.2.
• The degree to which electricity outages affect the performance of various sectors. It
will be recognised that some sectors, like fishing for example (as apposed to fish
processing), will be little, if at all, affected by any electricity outages. Others, like
manufacturing will be more severely affected. At the other end of the spectrum the
mining industry could be very severely affected by a power outage. This occurs not
because the mine is shut down but because the mineral processing plants are highly
electricity dependent. The assumptions that have been used in this regard are given
in Table 3-7 as a percentage of change in output so 100% means that there is no
impact, for example, and -200% means downtime that is twice as great as the power
outage.
Page 75 of 222
Elasticity of Affected
Economic Sector Growth
Demand Production
ComCereal 1.30% 0.00 100%
ComOtherCrops 1.30% 0.00 95%
Commercial animal prods 1.30% 0.00 90%
FoodForOwnCons 1.30% 0.00 100%
Fishing 1.00% 0.00 100%
Mining 2.10% 0.18 -200%
Meat Processing 0.50% 0.39 25%
Fish Processing 1.00% 0.39 50%
Grain Milling 1.30% 0.39 0%
Bev&Other Food Prod 0.50% 0.39 0%
Textiles -1.30% 0.69 0%
Light Manufacturing 2.00% 1.01 0%
Petroleum Products 3.60% 0.10 -100%
Heavy Manufacturing 3.90% 1.00 -100%
Electricity 1.90% 0.20 -100%
Water 1.90% 0.17 -100%
Construction 3.90% 0.92 50%
Trade, Repairs 2.10% 1.01 25%
Hotels and Restaurants 2.10% 1.23 50%
Transport 1.30% 0.89 75%
Communication 1.30% 0.89 10%
Finance and Insurance 5.30% 0.30 25%
Real estate, own 5.30% 0.30 75%
MktRealEst + Bus Services 5.30% 0.30 50%
Other Private Services 0.80% 0.30 50%
Government Services 2.80% 0.30 50%
Direct purch abroad by res 2.14% 0.00 100%
Tourism 4.00% 1.23 85%
Table 3-7: Growth rates, elasticities of demand and production effects from electricity outages
per economic sector
The economic model handles changes in electricity prices using the following approach. The
objective is to determine the percentage change in output per economic sector as a result of
a change in the price of electricity. To calculate the percentage change in output the model
first calculates the change in electricity costs (compared to a base case scenario) and the
effect that this would have on the overall price of a product25. The degree to which the price
change is passed onto the consumer or absorbed by the producer is then calculated from the
price elasticity of demand as is the change in output. If electricity prices are higher than the
‘base case’ scenario then this will detract from economic output, whereas if electricity prices
are lower than the ‘base case’ scenario then this will add to economic output.
25
The Namibian SAM provides the proportion of electricity costs as a percentage of overall input costs for each economic
sector.
Page 76 of 222
4 Analytical Results, Sensitivity Analysis and

Interpretation
This section reports the results to the analysis and presents the key conclusions. There are
three subsections to this part of the report:
• Section 4.1 gives the results to the analysis. In most instances this is done purely as
a presentation of facts. Where possible some context is also given. The interpretation
is left for later because it is important to first present all the results before attempting
to interpret them in any coherent and holistic manner.
• Section 4.2 presents a detailed comparison of the economic, social and

environmental implications of two of the generation scenarios, namely, the ‘base
case’ and ‘all options considered’ scenarios. It also provides a sensitivity analysis
around key assumptions.
• Finally section 4.3 summarises the overall results and interprets the findings.
4.1 ANALYTICAL RESULTS

The analytical results of this study are presented in eight parts:
• The first part gives the results of the economic cost benefit analysis.
• The second analyses various price projections and the implications of these.
• The third analyses the impact of economic growth on electricity demand, on the one
hand, and the impact of the generation scenarios on the economy, on the other.
• The fourth shows the likely social impacts in the form of job creation and contribution
to small, medium and micro enterprises.
• The fifth indicates the economic consequence of supply interruptions, on the one
hand, and potential unserved electricity demand under the various generation
options, on the other.
• The sixth looks at the potential contribution that renewable resources make in each of
the generation scenarios.
• The seventh analyses the potential electricity grid costs of centralised and
decentralised electricity generation.
• Finally, the eighth shows the environmental impacts of each of the generation
scenarios in the form of projected emissions.
4.1.1 Economic Cost Benefit Analysis

This section reports on the results of the cost benefit analysis for each generation scenario.
From an economic perspective it is on the benefit cost ratio particularly, and the net present
Page 77 of 222
value, to a lesser extent that a decision should be taken about which is the best generation
scenario for Namibia. It will be recognised, of course, that this decision criteria does not take
into account political and social considerations.
In addition to reporting on the results to the cost benefit analysis this section also reports on
a sensitivity analysis. This has been conducted to determine whether the main conclusions
that emerge from the CBA show any changes in the overall conclusions. For the purposes of
this analysis only the ‘base case’ and ‘all options considered’ scenarios are compared. This
sensitivity analysis is repeated in section 4.2.5 when it is done for the macro economic
analysis as opposed to the cost benefit analysis.
The sensitivities that are tested are:
1. The demand side management interventions.
2. Namibian economic growth of 6% and 2%.
3. The price of oil, coal and gas doubling or halving over the next five years.
4. The price of Eskom imports increasing by double or half the base assumption over
the next five years.
5. Assuming Eskom shortages to continue to 2017 (instead of 2012).
6. The cost of unserved demand varying between NAD 10.00 per kWh and NAD 40.00
per kWh, compared to the baseline scenarios which used NAD 20.00 per kWh.
7. The value of carbon credits varying between NAD 100 per ton and NAD 400 per ton,
compared to the baseline scenarios which used NAD 200 per ton.
As discussed in the methodology section, international best practise requires that each
option is compared to the do minimum option and that in this case the do minimum option is
the ‘no new local generation’ scenario. The options are not compared to the ‘base case’
because in this specific option it has been assumed that Namibia will continue going along as
business as usual, which means that investment will be made in the more mature
technologies (such as coal-fired power stations) and importing electricity from South Africa.
In this case the correct do minimum option is therefore the ‘no new local generation’
scenario.

NPV PV PV B:C
Comparison Table
No New Local Generation -13,325.1 13,362 37 0.00
Price Minimised 20,750.1 14,804 35,554 2.40
Maximum Supply Diversity 20,578.0 17,225 37,803 2.19
Generation Self-Sufficiency 18,800.0 18,183 36,983 2.03
Minimum Energy Imports 18,114.0 18,098 36,212 2.00
Renewables Maximised 20,463.3 15,859 36,322 2.29
Table 4-1: NPVs, incremental costs, incremental benefits and BCRs
The results of the economic cost benefit analysis are presented in Table 4-1. This gives the
overall net present value (NPV), the present value (PV) of the costs and benefits,
Page 78 of 222
respectively and the benefit cost ratio (BCR). The analysis includes the construction,
operating and fuel costs, impact of price changes on economic productivity, the cost of
unserved demand and carbon credits for each generation scenario.
A cost benefit analysis is a comparative exercise which is why the ‘no new local generation’
scenario has a BCR of zero - the benefits of all the other scenarios are compared to this one.
The scenario with the highest BCR is the ‘all options considered’ (2.69) followed by ‘price
minimised’ (2.40), ‘renewables maximised’ (2.29), ‘maximum supply diversity’ (2.19) and the
‘base case’ (2.09). The ‘minimum energy imports’ has the lowest BCR (2.00). It should be
noted that from an economic perspective this scenario is still more desirable than the ‘no new
local generation’ scenario because the BCR is greater than 1.
The conclusions that are drawn are that from an economic perspective:
• The ‘all options considered’ scenario, with a BCR of 2.69 is the most desirable
scenario.
4.1.1.1 Demand Side Management Sensitivity
In the baseline set of scenarios the ‘base case’ was implemented without demand side
management (also referred to as energy efficiency options) interventions while the ‘all
options considered’ scenario was implemented with demand side management interventions.
We now examine the impact if the ‘base case’ is implemented with demand side
management interventions as well as the impact of the ‘all options considered’ scenario
without demand side management interventions. The results are presented in Table 4-2.

NPV PV PV B:C
Comparison Table
Base Case With EE 17,312.2 13,277 30,589 2.30
All Options Considered Without EE 18,669.3 14,491 33,161 2.29
Table 4-2: Demand Side Management Sensitivity
For the ‘base case’ without DSM/EE, the BCR is 2.09. When implementing DSM for this
scenario the BCR increases to 2.30. The main reason for the increase in this ratio is that the
increased benefit from implementing DSM in the form of less unserved demand is more than
the incremental cost of implementing the DSM.
For the ‘all options considered’ scenario with DSM, the BCR is 2.69. If these DSM
interventions are excluded the BCR decreases to 2.29. The reason for the drop in the ratio is
similar to that for the ‘base case’, i.e. the saving in capital costs from not having to implement
the DSM is less than the lost benefit of the unserved demand.
The conclusion that is drawn is that CBA results remain consistent with the inclusion of
demand side management.
Page 79 of 222
4.1.1.2 Economic Growth Sensitivity
In the baseline set of scenarios it is assumed that the Namibian economy on average would
grow at 3.7% per annum. We now examine the impact if this growth is increased to 6%
(simulating the minimum growth needed to attain the goals of Vision 2030) and reduced to
2% per annum respectively. The results are presented in Table 4-3.

NPV PV PV B:C
Comparison Table
Base Case - 3.7% Growth 15,385.3 14,086 29,471 2.09
All Options Considered - 3.7% Growth 23,168.8 13,697 36,866 2.69
Base Case - 6% Growth 46,733.8 16,291 63,025 3.87
All Options Considered - 6% Growth 49,610.1 15,661 65,271 4.17
Base Case - 2% Growth 7,328.4 12,746 20,074 1.57
All Options Considered - 2% Growth 13,116.6 12,582 25,699 2.04
Table 4-3: Economic Growth Sensitivity
The BCRs increase when increasing the economic growth rate from 3.7% to 6% per annum.
For the ‘base case’ the BCR increases from 2.09 for a growth rate of 3.7% to 3.87 for a
growth rate of 6.0%, while for the ‘all options considered’ scenario the BCR increases from
2.69 to 4.17 respectively. The main reason for the increase in the BCR is the increase in the
change in unserved demand when compared to the ‘no new local generation’ scenario for
the higher growth rate.
Similarly, when the economic growth rate reduces to 2% p.a. the BCRs also decrease. For
the ‘base case’ the BCR decreases from 2.09 for a growth rate of 3.7% to 1.57 for a growth
rate of 2.0%, while for the ‘all options considered’ scenario the BCR decreases from 2.69 to
2.04 respectively. Once again the reason for the decrease in the BCR is the drop in the
change in unserved demand when compared to the ‘no new local generation’ scenario for
the lower growth rate.
The conclusion that is drawn is that the ‘all options considered’ scenario has a higher BCR
than the ‘base case’ for all growth rates considered, as well the fact that both scenarios have
BCRs greater than one and are desirable under all the economic growth rates analysed.
4.1.1.3 Oil, Coal and Gas Price Sensitivity
In the baseline set of scenarios it is assumed that the real price of oil would increase by
about 3% per annum going forward, while the real price of coal and gas would stay constant.
We now examine the impact if the price of all three commodities doubles and halves over the
next five years. The results are presented in Table 4-4.

NPV PV PV B:C
Comparison Table
Base Case - Baseline Commodity Prices 15,385.3 14,086 29,471 2.09
All Options Considered - Baseline Commodity Prices 23,168.8 13,697 36,866 2.69
Base Case - Double Commodity Prices 7,325.9 18,759 26,085 1.39
All Options Considered - Double Commodity Prices 16,753.7 16,525 33,279 2.01
Base Case - Half Commodity Prices 18,329.8 11,856 30,186 2.55
All Options Considered - Half Commodity Prices 25,366.7 12,322 37,689 3.06
Table 4-4: Commodity Price Sensitivity
Page 80 of 222
As one would expect, the BCRs reduce if the real price of oil, coal and gas increase. The fuel
cost increases, therefore making those generation options dependent on these commodities
for fuel more expensive and lowering the overall BCR. For the ‘base case’ the BCR
decreases from 2.09 to 1.39, while for the ‘all options considered’ scenario the BCR reduces
from 2.69 to 2.01. The ‘base case’, which is more dependent on these commodities for fuel,
is more adversely affected than the ‘all options considered’ scenario.
If the price of these commodities halves over the next five years then the BCR increases.
The BCR for the ‘base case’ increases from 2.09 to 2.55 while that for the ‘all options
considered’ scenario increases from 2.69 to 3.06.
The conclusion that is drawn is that the ‘all options considered’ scenario has a higher BCR
than the ‘base case’ for the fuel price changes that have been analysed. Both the ‘base case’
and the ‘all options considered’ scenario are desirable under the full range of commodity
prices examined.
4.1.1.4 Eskom Price Increase Sensitivity
In the baseline set of scenarios it is assumed that Eskom would increase their rates along a
standard set of prices where the base energy rate would reach the estimated cost of coal
over a period of ten years. We now examine the impact if these increases are both double
and half the baseline increase assumptions. The results are presented in Table 4-5.

NPV PV PV B:C
Comparison Table
Base Case - Baseline Eskom Increases 15,385.3 14,086 29,471 2.09
All Options Considered - Baseline Eskom Increases 23,168.8 13,697 36,866 2.69
Base Case - Double Eskom Increases 13,427.6 15,322 28,750 1.88
All Options Considered - Double Eskom Increases 21,790.8 14,345 36,136 2.52
Base Case - Half Eskom Increases 17,424.9 12,716 30,141 2.37
All Options Considered - Half Eskom Increases 24,400.9 13,104 37,505 2.86
Table 4-5: Eskom Price Sensitivity
The BCRs reduce when the price of the Eskom imports increase, in line with expectations.
For the ‘base case’ the BCR reduces from 2.09 to 1.88 when the Eskom price increase
doubles, and it increases to 2.37 when the price increase halves. For the ‘all options
considered’ scenario the BCR reduces from 2.69 to 2.52 when the price doubles, and it
increases to 2.86 when the price halves. The ‘base case’ is much more adversely affected by
changes in the price of Eskom imports.
The conclusion that is drawn is that the ‘all options considered’ scenario is more desirable
than the ‘base case’ for the full range of Eskom prices analysed. However, both scenarios
are economically justified for the range of Eskom price increases considered.
4.1.1.5 Eskom Electricity Shortages Sensitivity
In the baseline set of scenarios it is assumed that Eskom shortages will continue up until
2012. We now examine the impact if these shortages continue until 2017 (for the next ten
years). The results are presented in Table 4-6.
Page 81 of 222

NPV PV PV B:C
Comparison Table
Base Case - Baseline Eskom Shortages 15,385.3 14,086 29,471 2.09
All Options Considered - Baseline Eskom Shortages 23,168.8 13,697 36,866 2.69
Base Case - Eskom Shortages to 2017 1,896.3 15,746 17,642 1.12
All Options Considered - Eskom Shortages to 2017 10,212.4 15,015 25,227 1.68
Table 4-6: Eskom Shortages Sensitivity
The BCRs reduce if the Eskom shortages were to continue to 2017. For the ‘base case’ the
BCR decreases from 2.09 for the baseline shortages to 1.12 for when the shortages continue
to 2017, while for the ‘all options considered’ scenario the BCR decreases from 2.69 to 1.68
respectively.
The conclusion that is drawn is that the ‘all options considered’ scenario has the higher BCR
under both sets of Eskom shortages, and it is the ‘base case’, which is more dependent on
Eskom imports, that is the more adversely affected of the two cases.
4.1.1.6 Sensitivity to the Value of Unserved Demand
The generation scenarios assumed a cost of NAD 20.00 per kWh of unserved demand. The
effect of varying this cost between NAD10.00 per kWh and NAD 40.00 per kWh on the BCR
is investigated here and presented in Table 4-7. The BCR is the only variable in this analysis
that is impacted on by the cost of unserved demand and therefore is the only set of results
subjected to the sensitivity analysis.
Cost of Unserved Demand

Comparison Table
N$20/kWh N$10/kWh N$40/kWh
Base Case Without EE 2.09 1.05 4.18
No New Local Generation 0.00 0.00 0.00
Price Minimised 2.40 1.22 4.76
Maximum Supply Diversity 2.19 1.14 4.31
Generation Self-Sufficiency 2.03 1.04 4.01
Minimum Energy Imports 2.00 1.03 3.95
Renewables Maximised 2.29 1.21 4.45
All Options Considered With EE 2.69 1.39 5.30
Table 4-7: Cost of Unserved Demand
The conclusion that is drawn is that while varying the cost of the unserved demand does
have an effect on the value of the BCR, it does not change the relative order of the ‘all
options considered’ scenario in comparison to the rest of the scenarios. At a cost of
NAD 10.00 per kWh the ‘all options considered’ scenario BCR reduces from 2.69 to 1.39,
indicating that it is still beneficial to society and is the most desirable of all the options. At a
cost of NAD 40.00 per kWh the BCR for the ‘all options considered’ scenario increases to
5.30.
4.1.1.7 Sensitivity to the Value of Carbon Credits
The generation scenarios assumed a value of NAD 200 per ton of CO2 emissions saved. The
effect of varying this cost between NAD 100 per ton and NAD 400 per ton on the BCR is
investigated here and presented in Table 4-7. The BCR is the only variable in this analysis
Page 82 of 222
that is impacted on by the cost of unserved demand and therefore is the only set of results
subjected to the sensitivity analysis.
Value of Carbon Credits

Comparison Table
N$200 / t N$100 / t N$404 / t
Base Case Without EE 2.09 2.09 2.09
No New Local Generation 0.00 0.00 0.00
Price Minimised 2.40 2.38 2.45
Maximum Supply Diversity 2.19 2.15 2.29
Generation Self-Sufficiency 2.03 2.00 2.10
Minimum Energy Imports 2.00 1.97 2.07
Renewables Maximised 2.29 2.21 2.44
All Options Considered With EE 2.69 2.65 2.78
Table 4-8: Value of Carbon Credits
The BCR for the ‘base case’ is not affected by the change in value of carbon credits because
there is no reduction in CO2 for this particular scenario. For the ‘all options considered’
scenario the BCR reduces from 2.69 to 2.65 when the value of the carbon credits is changed
from NAD 200 per ton to NAD 100 per ton. Similarly, the BCR increases to 2.78 when the
value of the carbon credits is NAD 400 per ton.
As would be expected, the ‘renewables maximised’ scenario is the most sensitive to changes
in the value of carbon credits. The BCR reduces from 2.29 to 2.21 when the value of the
carbon credits is changed from NAD 200 per ton to NAD 100 per ton. Similarly, the BCR
increased to 2.44 when the value of the carbon credits is NAD 400 per ton
The conclusion that is drawn is that the ranking of the scenarios relative to one another is
only slightly affected within the range of the value of carbon credits tested, but in all
instances the ‘all options considered’ scenario still remains the best option.
4.1.2 Electricity Price Projections and Sensitivities

One of the key issues for this study is the likely future electricity prices that people and
business could face. This section presents the results of this analysis for the generation
scenarios. It also presents the sensitivity of the electricity price for changes in key
assumptions. These are sensitivity of the electricity price to economic growth, to fuel costs; to
demand side management measures and to electricity import constraints. In all cases the
sensitivity analysis is presented as a comparison of the ‘base case’ to the ‘all options
considered’ scenario.
Two general points need to be made before presenting the results to this section.
• First, electricity prices for each scenario are shown both as series over time and as
an average over time. The reason for this is that there is some variation in the
electricity price over time for some of the scenarios. In some scenarios prices, for
example, initially increase and then decrease. Hence it can be difficult to compare the
prices of the different scenarios. That is why average prices are also given. However
it needs to be borne in mind that average prices do not show the price variation that
could take place. Therefore in interpreting prices both average prices and the price
path need to be taken into account.
• Second, the cost of unserved energy is not included in the price of electricity.
Therefore section 4.1.5 needs to also be considered in interpreting the projected price
of electricity. (This has been however already been incorporated into the economic
Page 83 of 222
cost benefit analysis where a cost has been attached to unserved energy and the
overall impact on the various scenarios estimated).
4.1.2.1 Price projections
The average consumer price for each of the scenarios is illustrated in Figure 4-1. This is
shown in NAD in real 2007 terms for the entire period under study. The most expensive
electricity option is that maximum renewable at 106.9 c/kWh. This is followed by maximum
supply diversity at 106.8 c/kWh. The lowest electricity price is from the ‘no new local
generation’ at 94 c/kWh, followed by the ‘base case’ at 96 c/kWh and ‘price minimised’ at
97.5 c/kWh. It will be recognised of course that the ‘no new local generation’ scenario as well
as the ‘base case’ both have the highest level of unserved energy. The price minimised
scenario comes in only as the third cheapest scenario because it is centred on providing the
cheapest electricity possible while still avoiding large scale load shedding. The ‘all options
considered’ has the fourth lowest price of electricity.
Average End Consumer Price (real)
All options considered
Maximum renewable
Energy Imports minimised

Scenario
Generation Self sufficiency
Maximum supply diversity
Price minimised
No new local generation
Base Case
85 90 95 100 105 110

c/kWh
Figure 4-1: Projected Average Electricity Price
Figure 4-2 shows consumer price path for the scenarios. What is interesting here is that
while both the ‘base case’ and ‘no new local generation’ scenarios might have the lowest
average price they both see fairly large increases in prices up until 2011 before the price
starts to drop. The ‘maximum supply diversity’ scenario ends up having the highest electricity
price by the end of the forecast period. This is followed by ‘maximum renewable’, ‘generation
self sufficiency’, ‘energy imports minimised’, ‘all options considered’, and ‘price minimised’.
As before the ‘price minimised’ scenario comes in as the third cheapest scenario because it
is centred on providing the cheapest electricity while still avoiding large scale load shedding.
Page 84 of 222
Projected Electricity Price
130
125
120 Base Case
115 No new local generation
c/kWh (real 2007)
110 Price minimised

105
100
95
90 Energy Imports minimised
85 Maximum renewable
80 All options considered
75
70
05/06
06/07
07/08
08/09
09/10
10/11
11/12
12/13
13/14
14/15
15/16
16/17
17/18
18/19
19/20
20/21
21/22
22/23
23/24
24/25
25/26
Years
Figure 4-2: Projected Electricity Price Development
greatest price volatility.
• The ‘maximum supply diversity’ and ‘maximum renewable’ scenarios have the
highest electricity prices although they are less volatile than the lowest price
scenarios.
• The ‘all options considered’ scenario strikes a balance between the low prices of the
‘base case’ and ‘no new local generation’ scenarios and the high prices of the other
scenarios. One can therefore argue that from a price perspective a balanced policy
approach similar to the ‘all options considered’ scenario should yield an acceptable
price level while producing many economic benefits.
than the ‘price minimised’ and ‘base case’ scenarios. This demonstrates that it should
be possible to reduce supply risk, increase self-sufficiency, reduce emissions and
strongly embrace renewable energy without driving the price of electricity out of
reach.
4.1.2.2 Sensitivity to Economic Growth
As can be expected economic growth will impact on the demand for, and therefore price of,
electricity. As before the initial analysis has been done on a 3.7% economic growth rate and
the sensitivity on 2% and 6% growth rates. The average electricity price results are illustrated
Figure 4-3 in while the time series results are given in Figure 4-4.
On the whole the ‘all options considered’ scenario has slightly higher price sensitivity to
economic growth than the ‘base case’ scenario. In the ‘all options considered’ scenario the
electricity price decreases from 100 c/kWh to 97 c/kWh for high economic growth and
increases to 112 c/kWh for low economic growth. The reason this happens is that on the
generation side it does not necessarily make a big difference, but the transmission and
Page 85 of 222
distribution and Local Authorities surcharge costs are spread across a larger number of kWh,
thus reducing the average cost / kWh. This holds true while there is excess network capacity
which is largely the case in Namibia.
Average Electricity Price - Economic Growth Sensitivity
All options considered - Low Grow th
All options considered - High Grow th
Base Case - Low Grow th
Base Case - High Grow th
Base Case
0 10 20 30 40 50 60 70 80 90 100 110 120

c/kWh (real, average)
Figure 4-3: Average Electricity Price Sensitivity to Economic Growth
Figure 4-3 illustrates that the ‘base case’ has a low sensitivity to economic growth, while ‘all
options considered’ is sensitive to low growth.
Electricity Price - Economic Growth Sensitivity
135
130
125
120
115
110 Base Case
Base Case - High Growth
c/kWh (real)
105
100 Base Case - Low Growth
95 All options considered
90 All options considered - High Growth
85 All options considered - Low Growth
80
75
70
65
60
05/06
06/07
07/08
08/09
09/10
10/11
11/12
12/13
13/14
14/15
15/16
16/17
17/18
18/19
19/20
20/21
21/22
22/23
23/24
24/25
25/26
Years
Figure 4-4: Electricity Price Sensitivity to Economic Growth
Figure 4-4 shows a fairly low sensitivity of electricity price trends to high versus low economic
growth. This is mainly because, as mentioned above, the transmission and distribution cost
grows much slower than demand as long as there is excess grid capacity (which is generally
Page 86 of 222
true for Namibia) and thus these costs are distributed over a larger number of kWh, thus
reducing the average c/kWh. The only notable exception is the high price resulting from the
‘all options considered’ scenario with low growth. This results in reduced demand due to
cumulative effects of high price increases and slow economic growth which in turn increases
the transmission and distribution cost per kWh because the same cost is divided into less
kWh.
4.1.2.3 Sensitivity to Resource Costs
This sensitivity examines the impact on the price of electricity of fluctuations in the price of
coal, oil and electricity imports. The base assumption on these parameters is that the oil
price rises by 3% per annum in real terms in NAD, while coal stays constant in real terms26.
Eskom imports have been modelled in such a way that the marginal energy rate rises to
meet the fuel costs calculated for new coal plant in the model (assuming high efficiency and
no transport costs since South African coal stations are normally built at the mine).
All future Namibian power stations have been modelled with set sensitivity to oil and coal
price. So with changing oil and coal prices the fuel cost of the generators changes according
to their exposure to these prices. The Eskom and Zimbabwe imports are also modelled as
being exposed to world oil and coal prices. The Eskom imports have been modelled as being
80% exposed to coal price trends, while the Zimbabwe imports have been modelled as being
insensitive to coal price because they are understood to have a fixed price. The assumptions
in this regard have been given in Table 3-3.
Electricity Price Sensitivity to Resource Costs
All options considered - Low Eskom

price
All options considered - High Eskom

price
All options considered - Low Oil/Coal

price
All options considered - High Oil/Coal

Price
Base Case - Low Eskom price
Base Case - High Eskom price
Base Case - Low Oil/Coal price
Base Case - High Oil/Coal Price
-6.0% -4.0% -2.0% 0.0% 2.0% 4.0% 6.0% 8.0% 10.0% 12.0%
Average Price Change
Figure 4-5: Average Electricity Price Sensitivity to Resource
26
Fluctuations in the value of the Namibian dollar have not been taken into account in this analysis. There are two reasons for
this. First, such analysis would be purely speculative and, as is well known in economics, it is not possible to forecast the value
of a currency because all knowledge of the future value of a currency are already embodied in the spot price of a currency.
Second, and more importantly, because the Namibian dollar is fixed to the South African Rand it can be shown that the real
effective Rand exchange rate (as opposed to the nominal effective rate) has been largely unchanged since the 1980’s. What
that means is that any currency depreciation would be offset by an increase in inflation and the effective real exchange rate
remains unchanged.
Page 87 of 222
Figure 4-5 shows the percentage variation in average electricity price over the model
timeframe for the various sensitivities. The following conclusions are drawn:
• For the ‘all options considered’ a low Eskom price decreases prices by 1.2% while
higher prices increase electricity prices by 1.3%.
• A low coal or oil price lowers the ‘all options considered’ electricity price by 3% while
high coal and oil prices increase prices by over 6%.
• For the ‘base case’ scenario a low Eskom price decreases prices by 3% while higher
prices increase electricity prices by 2.5%.
• A low coal or oil price lowers the ‘base case’ electricity price by 4.5% while high coal
and oil prices increase electricity prices by nearly 10%.
• The ‘all options considered’ has a significantly lower exposure to variations in

resource costs than the ‘base case’.
Electricity Price - Resource Cost Sensitivities
130
125
120
115
Base Case
110
Base Case - High Oil/Coal Price
105 Base Case - Low Oil price
c/kWh (real)
100 Base Case - High Eskom price

Base Case - Low Eskom price
95
90 All options considered - High Oil/Coal Price
All options considered - Low Oil price
85
All options considered - High Eskom price
80 All options considered - Low Eskom price
75
70
65
60
05/06
06/07
07/08
08/09
09/10
10/11
11/12
12/13
13/14
14/15
15/16
16/17
17/18
18/19
19/20
20/21
21/22
22/23
23/24
24/25
25/26
Years
Figure 4-6: Electricity Price Sensitivity to Resource Costs
Figure 4-6 shows significant sensitivity to resource prices during the first half of the model
timeframe when all scenarios rely heavily on indexed resources. Once Baynes hydro comes
in the generation mix the price variations become much narrower.
4.1.2.4 Sensitivity to Continuing Electricity Import Constraints
As has been discussed, Namibia imports about half its electricity, mostly from Eskom. These
imports are currently constrained by shortages in South Africa and are projected to last for at
least five years. This sensitivity analysis seeks to understand what would happen to the price
of electricity if these constraints should last for ten years instead of five.
Page 88 of 222
Electricity Price Sensitivity to Import Constraints
All options considered -

Import Constraints
Base Case - Import

Constraints
0.0% 1.0% 2.0% 3.0% 4.0% 5.0% 6.0%
Average Price change
Figure 4-7: Average Price Sensitivity to Import Constraints
Figure 4-7 however demonstrates that for the “all options considered” there would be an
overall increase in price of electricity of nearly four percent. This stands in contrast to the
‘base case’ where there would be an electricity price increase of over five percent.
The conclusion that can be drawn is that:
• the ‘all options considered’ scenario has a lower sensitivity to South African import
constraints than the ‘base case’ because of its lower overall reliance on imports.
Electricity Price - Import Constraint Sensitivity
120
115
110
105
100
c/kWh (real)
95 Base Case
Base Case - Import Constraints
90
85 All options considered - Import Constraints
80
75
70
65
60
05/06
06/07
07/08
08/09
09/10
10/11
11/12
12/13
13/14
14/15
15/16
16/17
17/18
18/19
19/20
20/21
21/22
22/23
23/24
24/25
25/26
Years
Figure 4-8: Electricity Price Sensitivity to Import Constraints
Page 89 of 222
Figure 4-8, indicates the price trend over time of higher import constraints does however
shows distinct impacts on price trends. This demonstrates that both scenarios rely
significantly on imports. Because of the projected continuing low cost of South African
imports, relative to any generation options within Namibia, any scenario that seeks to contain
price is very likely to include a strong component of Eskom imports.
4.1.3 Economic Impacts

There are two categories of pure economic impact. The first is the impact of economic
growth on the demand for electricity and the implications that this has for future planning for
electricity generation. The second is the economic impact that each generation scenario will
have on the economy itself. This is measured as contribution to GDP. Each of these is
discussed in turn below.
4.1.3.1 Demand Changes due to Economic Growth
The demand development takes its starting point from a projected economic growth of 3.7%
per annum as ‘base case’. Sensitivity analysis is performed at 2% and 6%. The methodology
of demand forecasting is described fully in section 3.1.1.
Projected System Peak Demand
1 100
1 050
1 000
950
900
850
800
750
MW
700
650
600
550
500
450
400
350
05/06
06/07
07/08
08/09
09/10
10/11
11/12
12/13
13/14
14/15
15/16
16/17
17/18
18/19
19/20
20/21
21/22
22/23
23/24
24/25
25/26
Years
Figure 4-9: Projected Peak Demand at an economic growth rate of 3.7%
Figure 4-9 shows very clearly two possible demand developments differentiated by the
implementation of demand side management. The conclusion that is drawn is that with the
economy growing at 3.7% peak demand will grow from just under 450MW in 2007/8 to
1100MW by 2025/6. This growth in demand could be limited to 950MW with the
implementation of demand side management.
Figure 4-10 illustrates a similar trend for the energy supply demanded, with demand side
management reducing the demand in 2025/6 by as much as 900GWh per annum. There is a
12% lower demand because of demand side management.
Page 90 of 222
Projected Demand for Electricity
8 000
7 000
GWh per annum
6 000
5 000
4 000
3 000
2 000
05/06
06/07
07/08
08/09
09/10
10/11
11/12
12/13
13/14
14/15
15/16
16/17
17/18
18/19
19/20
20/21
21/22
22/23
23/24
24/25
25/26
Years
Figure 4-10: Projected Electricity Demand at an economic growth rate of 3.7%
Clearly one of the key factors that will influence the future demand for electricity is economic
growth. Variations in electricity demand due to economic growth are illustrated in Figure 4-11
for a high growth of 6% and a low growth of 2%. In both cases the electricity demand has
been tempered by the introduction of demand side management. In the ‘base case’ a high
economic growth increases the demand for electricity in 2025/6 to 9 600 GWh per annum
compared to 7 700 GWh for a 3.7% growth. For an economic growth of 2% the electricity
demand would be 6 700 GWh per annum. Similarly in the ‘all options considered’ scenario a
high economic growth would result in electricity demand of 8 400 GWh per annum at the end
of the forecast period and a low growth a demand of 5 900 GWh per annum.
Demand - Economic Growth Sensitivity
12000
10000
8000
GWh per annum

Base Case - Low Growth
6000
All options considered - High Growth
All options considered - Low Growth
4000
2000
0
05/06
06/07
07/08
08/09
09/10
10/11
11/12
12/13
13/14
14/15
15/16
16/17
17/18
18/19
19/20
20/21
21/22
22/23
23/24
24/25
25/26
Years
Figure 4-11: Projected Demand for Electricity at an economic growth rate of 2% and 6%.
Just as future economic growth will affect GWh so it also affect peak load MW. Variations in
peak load due to economic growth are illustrated in Figure 4-12 for a high growth of 6% and
a low growth of 2%. In both cases the electricity demand has been tempered by the
Page 91 of 222
introduction of demand side management in the ‘all options considered’ scenario. In the
‘base case’ a high economic growth increases peak load in 2025/6 to 1 450 MW (from 1 100
MW for a 3.7% growth). For an economic growth of 2% the electricity demand would be 900
MW. Similarly in the ‘all options considered’ scenario a high economic growth would result in
electricity demand of 1 250 MW at the end of the forecast period and a low growth a demand
of 780 MW.
System Peak - Economic Growth Sensitivity
1600
1400
1200
1000
Base Case - Low Growth
MW
800
All options considered - High Growth
All options considered - Low Growth
600
400
200
0
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
Financial Ye ar ending June
Figure 4-12: Projected System Peak Demand at economic growth of 2% and 6%
4.1.3.2 Generation Scenarios Contribution to GDP
One of the important aspects of a policy decision is the potential contribution of competing
projects to GDP. Figure 4-13 illustrates the cumulative impact on GDP between 2007/08 and
2025/26. It is given in NAD billions in 2007 values.
The conclusions that are drawn from the economic analysis are that:

constructing additional capacity at the Baynes hydroelectric power station (more than
in any of the other scenarios), as well as requiring significant output from
concentrating solar power stations.
‘generation self sufficiency’ scenarios, both at NAD 13.1bn.
GDP.
Page 92 of 222
Cumulative Impact on GDP
All Options Considered With EE
Renewables Maximised
Minimum Energy Imports
Generation Self-Sufficiency
Maximum Supply Diversity
Price Minimised
No New Local Generation
Base Case Without EE
0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0

NAD billion
Figure 4-13: Cumulative impact on GDP under different scenarios, NAD billion (2007 values).
4.1.4 Social Impacts

Attention was paid to the social impacts of electricity generation options and scenarios. The
social impacts that were focussed on are the direct impacts of the various scenarios rather
than the broader social consequences of increased electricity generation like land use
changes because of new hydro stations, for example. These broader impacts are discussed
in the context of each generation option and can be found in Appendix C.
There are two direct impacts that are analysed: job creation (skilled, unskilled, direct and
indirect) and potential contribution to small, medium and micro enterprises (SMME).
4.1.4.1 Job Creation
Job creation is clearly an important social consequence of policy decisions. This section
reports on total job creation for each generation scenario. As discussed in the methodology
section (3.3) we report both direct jobs and indirect jobs. In addition the analysis also
distinguishes between skilled positions and unskilled positions.
Figure 4-14 indicates the average number of jobs created for each of the scenarios. The
following conclusions are drawn:
followed by the ‘minimum energy imports’ and the ‘generation self-sufficiency
scenarios at 5 851.
• The main reason for the high job creation for these three scenarios is the
implementation of the biomass (invader ) generation option. This is a labour intensive
generation option, particularly at the unskilled level.
Page 93 of 222
Total Job Creation

Unskilled Jobs Skilled Jobs
All Options Considered With EE
Renewables Maximised
Minimum Energy Imports
Generation Self-Sufficiency
Maximum Supply Diversity
Price Minimised
No New Local Generation
Base Case Without EE
0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000

Job Numbers
Figure 4-14: Total jobs within each scenario.
4.1.4.2 Contribution to small, medium and micro enterprises
One potential positive contribution of changes in electricity generation is that they could allow
easier access for SMMEs. The promotion of SMMEs is important in a country that faces
unemployment because it empowers people directly, stimulates economic growth and
creates jobs. Because this was not the direct focus of the study the likely contribution of each
generation options to the promotion of SMMEs was discussed within the study group and
ranked on a scale of zero to ten where zero was a low contribution and ten a high
contribution. Each of the generation options was then brought into the generation scenarios
and weighted according to its contribution to the overall generation mix. This resulted in a
final SMME score for each generation option. These results are reported in Table 4-9.
Page 94 of 222
Weighted Scores (GWhr Produced x Score)

Generation Option Score
Scen 1 Scen 2 Scen 3 Scen 4 Scen 5 Scen 6 Scen 7 Scen 8
Ruacana 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
van Eck 2 0.0 0.1 0.0 0.0 0.0 0.0 0.1 0.0
Paratus 1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
CCGT (Kudu) 3 0.0 0.0 0.0 0.8 1.1 1.1 0.0 0.0
Baines 4 0.9 0.0 0.9 0.0 0.8 0.8 1.0 1.1
Wind 2 0.0 0.0 0.0 0.1 0.0 0.0 0.1 0.1
Concentrating Solar 2 0.0 0.0 0.1 0.3 0.1 0.1 0.4 0.1
Biomass (Inv Bush) 5 0.0 0.0 0.3 0.3 0.3 0.3 0.6 0.3
Import Eskom 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Nuclear 1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
"Clean" Coal 2 0.2 0.0 0.2 0.1 0.0 0.0 0.0 0.1
Solar PV 1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Loadshed 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Zim Import 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Gas/Solar Hybrid 3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Weighted Score 1.2 0.1 1.5 1.5 2.4 2.4 2.1 1.8
Table 4-9: SMME contribution of each generation scenario
• Many of the generation options will not promote much SMME development, as is
evidenced by the zeros and ones on their scorecards. These generation options
include the Ruacana power station and the import and load shedding options (all
zeros), the Paratus power plant and any nuclear or solar photovoltaic power stations
(all ones).
• Scenarios five and six (the generation self-sufficiency and minimum energy imports
scenario) score the highest as far as potential SMME development is concerned. This
is largely due to the Kudu, Baynes and biomass (invader bush) generation options
contribution.
• As would be expected the ‘no new local generation’ scenario (mainly imports) scores
4.1.5 Supply Interruptions and Unserved Demand

This section deals with two related issues. The first is the potential impact of electricity supply
interruptions and the second that of unserved electricity demand. Each of these is dealt with
in turn below.
4.1.5.1 Impact of Supply Interruptions
It is clearly important to understand the likely economic impacts of interruptions in the supply
of electricity. This was done by analysing the economic impact of a 10% and a 20% drop in
peak electricity supply in the form of an unannounced power outage. What was found, and
will be demonstrated below, is that a 10% drop in peak supply will have little economic
impact but a 20% drop will. What could not however be established with any degree of
certainty is the economic impact of this. In consequence an alternative approach was settled
Page 95 of 222
on which was to estimate the economic impact not of drop in peak supply but rather of a total
power outage.
The starting point in addressing this task was to profile the maximum electricity demand.
Figure 4-15 illustrates the expected hourly electricity demand profile for Namibia for 2007.
The figure indicates that the maximum demand is 441.4MW and occurs at 8pm at night.
Three demand profiles are displayed, namely a maximum demand profile, the weekly high
and the weekly low profile. The maximum demand level occurs on 15 days of the year, the
weekly maximum on 55 days of the year (generally in winter) and the weekly minimum on the
remainder 295 days. Also shown on the graph are two lines that indicate a reduction of 10%
and 20% respectively in maximum demand. In other words it shows at what times a power
outage would occur.
The 10% drop in maximum demand corresponds to a demand level of 397.3MW. What this
indicates is that if load shedding had to occur to reduce maximum demand by 10% (or
44.1MW) then this load shedding would only occur between the hours of 7:00pm and
10:00pm on the 15 days of maximum demand. At this time of the day it can be expected that
households would probably be targeted first to reduce this demand. This would then be
followed by night time entertainment such as restaurants and hotels. It can therefore be
expected that this level of power outage will have a very limited effect on economic growth
although it would impact adversely on entertainment and restaurant businesses.
Similarly, a 20% drop in maximum demand corresponds to a demand level of 353.1MW.

What this indicates is that if load shedding had to occur to reduce maximum demand by 20%
(or 88.3MW) then this load shedding would occur between the hours of 7:30am and 10:30pm
on the 15 days of maximum demand. It would also have an effect on the days of weekly high
demand between the hours of 9:00am and 2:30pm and 6:30pm and 10:00pm, while the days
of weekly low demand would be affected between the hours of 8:30am and 3:00pm and
again between 7:00pm and 10:00pm. Households would probably be most affected during
the effecting blackouts, but businesses would start to feel the effects during the daytime
blackouts. In consequence such a power outage would start to have negative economic
impacts. As this would eat directly into the days most productive times and businesses would
be adversely affected.
Hourly Electricity Demand - 2007

500.0
450.0
10% Reduction
400.0
20% Reduction
350.0
Demand (MW)
300.0 Week Low

Week High
Max
250.0 10% Reduction
20% Reduction
200.0
150.0
100.0
H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13 H14 H15 H16 H17 H18 H19 H20 H21 H22 H23 H24
Time of Day
Figure 4-15: Electricity Demand Profile
Page 96 of 222
• For a 10% reduction in maximum demand in the evening hours the impact on the
economy would be minimal This, however, does not take into account the
inconvenience of the blackouts, the social consequences and the negative image that
the country would portray to potential foreign investors.
• For a 20% reduction in maximum demand the impact would be greater and
businesses would now be affected.
The constraining factor in estimating the economic impact of this drop in peak supply is that
we do not know which industries would be affected by the drop. In consequence an
alternative, illustrative approach was taken to attempt to address this issue and identify those
industries that could be most adversely affected.
In doing this we looked at the economic impact of a 24-hour one day a month power
blackout. To do this some assumptions were made about the degree to which a blackout
would affect different sectors. This has already been introduced in section 3.3.2.3.4 but is
repeated here. It will be recognised that some sectors, like fishing for example (as apposed
to fish processing), will be little, if at all, affected by any electricity outages. Others, like
manufacturing will be more severely affected. At the other end of the spectrum the mining
industry could be very severely affected by a power outage. This occurs not because the
mine is shut down but because the mineral processing plants are highly electricity
dependent. The assumptions that have been used in this regard are given in Table 4-10 as a
percentage of change in output so 100% means that there is no impact, for example, and -
200% means downtime that is twice as great as the power outage.
GVA % output if GVA if %

Sector / Stakeholder
(N$m) no electr. interrupt. Change
ComCereal 66 100% 66 0%
ComOtherCrops 318 95% 317 0%
Commercial animal prods 1,694 90% 1,688 0%
FoodForOwnCons 493 100% 493 0%
Fishing 2,276 100% 2,276 0%
Mining 6,898 -200% 6,217 -10%
Meat Processing 328 5% 313 -4%
Fish Processing 1,017 5% 973 -4%
Grain Milling 610 0% 582 -5%
Bev&Other Food Prod 2,140 0% 2,041 -5%
Textiles 400 0% 382 -5%
Light Manufacturing 926 0% 883 -5%
Petroleum Products 382 100% 382 0%
Heavy Manufacturing 1,359 -100% 1,270 -7%
Electricity 808 0% 781 -3%
Water 429 0% 415 -3%
Construction 895 50% 880 -2%
Trade, Repairs 4,418 25% 4,265 -3%
Hotels and Restaurants 1,329 50% 1,307 -2%
Transport 1,365 95% 1,362 0%
Communication 1,047 50% 1,030 -2%
Finance and Insurance 1,551 5% 1,483 -4%
Real estate, own 2,096 75% 2,071 -1%
MktRealEst + Bus Services 2,242 50% 2,191 -2%
Other Private Services 1,141 50% 1,115 -2%
Government Services 8,580 50% 8,382 -2%
Total 44,806 43,165
1,641 -3.7%
Table 4-10: Effect of supply interruptions on GVA
Page 97 of 222
The following conclusion is drawn:
• Based on these assumptions it is calculated that a 24 hour blackout for one day a
month – the equivalent of a 3.29% drop in electricity supply would reduce Gross
Value Added by up to 3.7%. As mentioned earlier this impact is purely illustrative and
based on the assumption that there is only one 24-hour interruption per month. What
has not been factored into this analysis is the real possibility that many firms would
respond to the power outages by installing their own generators. Under such
circumstances the conclusions drawn above would be at the outer end of the
spectrum and the actual economic harm may not be as great.
4.1.5.2 Unserved Electricity Demand
As mentioned before the dispatch model takes a specified constellation of generators,

demand side management options and load growth and solves the dispatch equation until
supply and demand balance. If it cannot match demand with existing supply then there is a
residual called unserved energy.
Figure 4-17 shows the potential unserved energy for the various scenarios.
By far the largest amount of unserved demand occurs in the “no new local generation”
scenario and is driven by the Eskom shortages expected in the coming years. Under this
scenario unserved energy could increase to a cumulative 500 MWh in 2010/11 before
accelerating to cumulative 2 750 MWh in 2012/13 and then gradually increasing to nearly
3 000 MWh by the end of the forecast period. This amounts to a peak of 10.8% of cumulative
unserved demand by 2012/13 which implies dramatic impacts on the Namibian economy.
This is a dramatic and frightening prospect.
2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
Base Case 0.7% 1.6% 4.6% 3.2% 2.4% 1.9% 1.6% 1.3% 1.2% 1.0% 0.9% 0.8% 0.7% 0.7% 0.6% 0.6% 0.5% 0.5%
No new local generation 0.7% 1.6% 4.6% 9.5% 10.8% 10.1% 8.4% 7.1% 6.1% 5.4% 4.8% 4.4% 4.0% 3.6% 3.4% 3.1% 2.9% 2.7%
Price minimised 0.7% 1.0% 1.4% 0.9% 0.7% 0.6% 0.5% 0.4% 0.4% 0.3% 0.3% 0.2% 0.2% 0.2% 0.2% 0.2% 0.2% 0.1%
Maximum supply diversity 0.7% 0.6% 0.5% 0.3% 0.3% 0.2% 0.2% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1%
Generation Self sufficiency 0.7% 0.8% 0.7% 0.5% 0.4% 0.3% 0.3% 0.2% 0.2% 0.2% 0.2% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1%
Energy Imports minimised 0.7% 0.8% 1.1% 0.8% 0.6% 0.5% 0.4% 0.3% 0.3% 0.3% 0.2% 0.2% 0.2% 0.2% 0.2% 0.2% 0.2% 0.1%
Maximum renewable 0.7% 0.6% 0.5% 0.7% 0.7% 0.6% 0.6% 0.6% 0.6% 0.5% 0.5% 0.4% 0.4% 0.3% 0.3% 0.3% 0.3% 0.3%
All options considered 0.7% 0.6% 0.7% 0.5% 0.4% 0.3% 0.3% 0.2% 0.2% 0.2% 0.2% 0.2% 0.2% 0.1% 0.1% 0.1% 0.1% 0.1%
Figure 4-16: Potential Cumulative Unserved Electricity Demand
The scenario with the next highest level of unserved energy is that of the ‘base case’. For
this scenario unserved energy could increase dramatically to 500 MWh (cumulative) by
2010/11 before levelling off for the rest of the forecast period because in the ‘base case’ no
significant shortages are expected thereafter.
The remaining scenarios have a very limited amount of unserved energy. The maximum
renewable has unserved energy of 0.3% by 2025 while the remainder have unserved energy
of 0.1%. The reason for these relatively low levels of unserved demand is that these
scenarios all seek to minimise unserved demand.
Page 98 of 222
Cumulative Unserved Demand %
12.0%
% of total cumulative demand
10.0%
8.0%
6.0%
4.0%
2.0%
0.0%
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
Years
Base Case No new local generation All options considered
Figure 4-17: Potential Unserved Electricity Demand
4.1.6 Contribution of Renewable Resources to Electricity

Generation
As one of the major thrusts of the terms of reference was to determine the contribution that
renewable energy resources could play in the Namibian electricity generation this section
documents the potential contribution from renewable resources. For the purposes of clarity
this contribution is documented both with and without the contribution made by hydro
generators. The reason for this is two fold. First, hydro generators are large relative to the
other renewable options and therefore tend to crowd out the other results. Second, hydro,
despite being renewable, is large-scale, centralised and conventional and therefore in a
slightly different league to the other renewable options.
The detailed generation mixes for each scenario are described in section 3.2. Figure 4-18
and Figure 4-19 show the contribution to total GWh made by renewable generators. The first
figure includes hydro power while the latter excludes this contribution.
Page 99 of 222
Renewable Part of Generation Mix (including Hydro)
100%
90%
80%
% annual MWh Renewable
Base Case
70% No new local generation
60% Price minimised

50%
40% Energy Imports minimised
30% Maximum renewable
20% All options considered
10%
0%
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
Years
Figure 4-18: Projected Renewable (including hydro) Generation Contribution
It is of interest to note that at the moment nearly one half of all Namibia’s electricity
generation comes from renewable generators in the form of hydro power. In time the
‘maximum renewable’ scenario could contribute between 90% and 100% of all of Namibia’s
electricity needs. In contrast, and as could be expected, the ‘no new local generation’
scenario has the lowest renewable component. The sudden jump in the contribution of
renewable energy in 2017 is the commissioning of Baynes hydro power plant.
Renewable Part of Generation Mix (excluding Hydro)
50%
40%
% annual MWh Renewable
Base Case
30% No new local generation
Price minimised
20%
10% Maximum renewable
0%
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
-10%
Years
Figure 4-19: Projected Renewable Generation Contribution excluding hydro
If one excludes hydro generation then the non hydro renewable part of the ‘maximum
renewable’ scenario stands at 46% in 2014 to 2017 before dropping to 25% as Baynes
comes into the generation mix.
Page 100 of 222

It should be noted that the “maximum renewable” scenario would achieve a renewable
contribution of over 90% in the long run, but as can be seen from the previous sections this
would come at a price both in terms of unserved demand as well as higher electricity price.
4.1.7 Grid Cost: Centralised vs. Decentralised

The project team has explored the topic of grid cost impact of centralised versus
decentralised generators both within the team as well as through an extended discussion
with NamPower grid planning and operations specialists.
One main outcome of these deliberations is that it is very difficult to predict the effect on grid
revenue requirement in general, and even for a set of scenarios like those developed in this
study since the determinants of grid costs are complex and not in all cases easily linked to
generation options in general. The consensus was that decentralised smaller scale
generation should reduce grid costs in the long term, provided that technologies employed
for generators and generator grid connection are such that they do not require major grid
strengthening or major investment in grid control and protection equipment.
The other main outcome of these deliberations was that for a number of renewable, small
scale and decentralised generator options the grid connection presents significant
restrictions in terms of plant sizing, connection technology, plant site and maximum capacity
of specific plant types that can easily be accommodated on the grid.
Biomass plants are limited in terms of individual plant size and total summated plant capacity
due to grid constraints in the target area for this type of plant. Wind generators also face
restrictions in terms of ability of the grid to handle their input in the areas with the best-known
wind resources (Lüderitz). This limits the application of these generator types in our
modelling.
Specifically for biomass generators it will be easiest to accommodate small units (500kW or
1MW each) on the distribution networks in northern Namibia. If inverter technology is used
for the connection between the generator and the grid then it should be possible to
accommodate between 2MW to 5MW on a typical distribution network. This depends on the
transformer rating of the source substation (2.5MVA or 5MVA in the case of most rural
networks), the loading on the distribution line and the distance from the generator to the
source substation. The total relevant transformer capacity of substations in the area in
question amounts to approximately 60MW to 70MW. On this basis the maximum capacity for
biomass generation has been modelled as 60MW in most cases.
Specifically for wind generators at Lüderitz this means a limit of around 30MW since this is
capacity of the transmission infrastructure to the Lüderitz area. A larger wind park at Lüderitz
may cause system instability even though the sum of local load and the transmission line
together exceeds 30MW. NamPower is also concerned that a larger wind park’s intermittent
output would make it significantly more difficult to balance demand and supply of the country
since it becomes too large a portion of the total supply capacity. The maximum wind
generation has therefore been modelled as 60MW, assuming 30MW at Lüderitz and 30MW
at Walvis Bay.
4.1.8 Emissions
Total emissions of carbon dioxide (CO2), nitrogen oxide (NOx) and sulphur oxide (SOx) are
reported for the various generation scenarios. This is done first by reporting total cumulative
emissions over the entire forecast period and then the changes in emissions over time. In the
latter case only CO2 emissions are reported because NOx and SOx emissions follow similar
patterns to CO2 emissions.
Page 101 of 222

The broader environmental impacts are discussed in the context of each generation option
and can be found in Appendix C.
Figure 4-20 indicates the total cumulative emissions for each generation scenario for the full
forecast period. These emissions have been measured as total emissions at source and not
just emissions in Namibia. In other words emissions in South Africa for power used in
Namibia have also been included.
Cumulative Environmental Emmissions (SO2 & NO2 - kt, CO2 - Mt)
All Options
Considered
Renewables
Maximised
Minimum Energy
Imports
Generation Self-
Sufficiency
Maximum Supply
Diversity
Price Minimised
No New Local
Generation
Base Case
0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0 400.0
NO2 Emmissions SO2 Emmissions CO2 Emmissions
Figure 4-20: Cumulative environmental emissions
The following conclusions are drawn about total cumulative emissions:
• The ‘maximum renewables’ scenario has the lowest emissions, followed by the
‘generation self-sufficiency’ and ‘minimum energy imports’ scenarios.
• The ‘no new local generation’ scenario has the most emissions, followed by the ‘base
case’ scenario.
Page 102 of 222

Projected CO2 Emissions
7 000
6 000
Base Case
tons (1000's) per annum
5 000 No new local generation

Price minimised
4 000
3 000
2 000 Maximum renewable
1 000
-
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
Years
Figure 4-21: Projected CO2 Emissions
The changes in emissions over time are illustrated in Figure 4-21, Figure 4-22 and Figure
4-23. The following conclusions about the levels, and changes over time, of emissions:
of emissions.
• The ‘all options considered’ scenario achieves the second best level of emissions in
the latter part of the model horizon behind the ‘maximum renewables’ scenario. This
shows that a balanced policy approach with strong renewables preference can
achieve better emissions results than most of the other policies at a moderate price
premium.
emission even though these do take place in South Africa and not in Namibia.
Page 103 of 222

Projected NOx Emissions
12 000
10 000
Base Case

8 000
Price minimised
6 000
4 000
Maximum renewable
2 000
-
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
Years
Figure 4-22: Projected NOx Emissions
Projected SOx Emissions
10 000
9 000
Base Case
8 000
7 000 Price minimised

6 000 Maximum supply diversity
5 000 Generation Self sufficiency
4 000 Energy Imports minimised
3 000 Maximum renew able
2 000 All options considered
1 000
-
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
Years
Figure 4-23: Projected SOx Emissions
Page 104 of 222

4.2 COMPARISON OF THE ‘BASE CASE’ AND THE ‘ALL

OPTIONS CONSIDERED’ SCENARIOS
This section gives a detailed comparison of the economic differences of the ‘base case’
compared to the ‘all options considered’ scenario. This is done for contribution to GDP, job
creation, balance on current account and the electricity price. It also conducts a sensitivity
analysis on the key variables to see if there is any significant change in the results.
4.2.1 Contribution to GDP

Gross Domestic Product is the total value of all final goods and services produced in the
country. It is clearly fundamental to the economic quality of life of people in the country. It is
also the most important and all encompassing measure of the macro-economic effect of the
electricity generation options.
4.2.1.1 Base case
Table 4-11 reports on the ‘base case’ contribution to GDP and the composition of this
contribution.
Contribution to Namibia Gross Domestic Product

N$ million, 2007 prices
2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2017/18 2025/26
Capital Expenditure 2.1 677.8 743.0 743.0 743.0 675.7 0.0 0.0
Operational Expenditure 137.6 150.1 164.0 196.7 261.1 244.9 341.0 390.4
Fuel Expenditure 10.1 13.7 17.9 27.7 18.3 15.9 0.8 7.7
Economic Productivity
Total Contribution 149.7 841.6 924.9 967.4 1,022.4 936.4 341.9 398.1
Cumulative Contribution 149.7 991.3 1,916.2 2,883.6 3,905.9 4,842.4 6,060.8 8,982.0
Table 4-11: Contribution of the ‘base case’ to Gross Domestic Product
It can be seen from Table 4-11 that operating expenditure is the largest consistent
contributor to gross domestic product, followed by sporadic contributions from capital
expenditure for maintenance purposes. Capital expenditure includes the capital cost
associated with demand management systems. The contribution from fuel expenditure varies
over time as the different generation options come on line. There is no contribution from
changes in economic productivity because the ‘base case’ is the scenario against which all
other scenarios are compared for this category.
Operating expenditure is set to contribute NAD 137.6m to GDP in 2007/08, increasing to

NAD 390.4m in 2025/26. The total contribution to GDP from all forms of expenditure is
expected to increase from NAD 149.7m in 2007/08 to NAD341.9m in 2017/18 and NAD
398.1m in 2025/26.
GDP is important not just because it is income but also because income has the capacity to
add to wealth. Based on these projections, the ‘base case’ generation option is expected to
make a cumulative contribution to GDP of over NAD 8.9bn by 2025/26.
4.2.1.2 All options considered
The contribution to GDP for the ‘all options considered’ scenario is reported in Table 4-12.
Page 105 of 222

Contribution to Namibia Gross Domestic Product

N$ million, 2007 prices
2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2017/18 2025/26
Capital Expenditure 2.1 691.1 930.6 1,056.7 917.7 891.9 39.4 0.0
Operational Expenditure 137.6 149.9 178.8 236.9 285.7 290.9 452.7 534.2
Fuel Expenditure 10.1 13.6 17.5 26.4 21.5 22.8 14.8 21.4
Economic Productivity 0.0 0.2 29.8 31.2 -26.1 -36.5 -30.2 -41.7
Total Contribution 149.7 854.8 1,156.7 1,351.3 1,198.8 1,169.1 476.7 513.9
Cumulative Contribution 149.7 1,004.5 2,161.2 3,512.5 4,711.3 5,880.4 8,182.5 12,053.3
Table 4-12: Contribution of ‘all options considered’ to Gross Domestic Product
Operating expenditure is set to contribute NAD 137.6m to GDP in 2007/08, increasing to

NAD 534.2m in 2025/26. The total contribution to GDP from all forms of expenditure is
expected to increase from NAD 149.7m in 2007/08 to NAD 476.7m in 2017/18 and NAD
513.9m in 2025/26.
Based on these projections, the ‘all options considered’ scenario is expected to make a
cumulative contribution to GDP of over NAD 12.0bn by 2025/26.
4.2.1.3 Comparison
This section compares the total cumulative contribution to GDP of the two options and is
illustrated in Figure 4-24. The ‘all options considered’ scenario has a cumulative contribution
to GDP of NAD 12.05bn, compared to the NAD 8.98bn of the ‘base case’. The greatest
difference in the two cases lies in the operating and capital expenditure categories, while the
change in economic productivity has a slight tempering effect on the ‘all options considered’
scenario. The negative impacts on economic productivity in the later years outweigh the
positive impacts of the early years and the net effect over the time frame considered is a
relative reduction in economic output.
Cumulative Contribution to GDP

14,000.0
12,000.0 Fuel Expenditure
Operational Expenditure
Capital Expenditure
10,000.0
8,000.0
GDP (N$ millions)
6,000.0
4,000.0
2,000.0
0.0
Base Case All Options Considered
-2,000.0
Figure 4-24: Comparative Cumulative Contribution to GDP
Page 106 of 222

4.2.2 Job Creation

As explained in the methodology section there are three types of jobs that are affected. The
first are the direct jobs in the new power plants. The second are the so-called indirect jobs
that are due to multiplier effects of both the capital, operational and fuel expenditure costs.
The third type of jobs results from the structural economic changes attributable to the
changes in electricity prices.
4.2.2.1 Base case
It can be seen from Table 4-13 that 166 operating jobs are expected to be sustained in
2007/08 from the ‘base case’ scenario. These direct jobs are expected to increase to 249 in
2009/10 and 2010/11 when a new coal power station comes on line. The total number of
direct jobs is then expected to vary between 236 and 197 thereafter.
There are no direct jobs created from the fuel expenditure (they are included in the
operational expenditure jobs). There are no changes in jobs due to changes in economic
productivity because this is the ‘base case’ scenario and all forms of productivity gains in
other scenarios are based on the relative performance of the economy in this scenario.
Contribution to Direct Jobs by Type of Expenditure
2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2017/18 2025/26

Capital Expenditure 0 0 83 83 0 0 0 0
Operational Expenditure 166 166 166 166 236 236 197 197
Fuel Expenditure
Total Contribution 166 166 249 249 236 236 197 197
Table 4-13: ‘Base case’ Contribution to Direct Jobs
The number of indirect jobs, as indicated in Table 4-14, is set to increase from 747 in
2007/08 to 2 009 in 2025/26. Operating expenditure is the most consistent contributor to
indirect job creation, with capital expenditure creating periods of high employment.
Contribution to Indirect Jobs by Type of Expenditure
2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2017/18 2025/26

Capital Expenditure 11 2,389 2,378 2,378 2,378 2,378 0 0
Operational Expenditure 697 761 831 997 1,324 1,241 1,729 1,979
Fuel Expenditure 38 51 66 100 70 61 3 30
Total Contribution 747 3,201 3,275 3,475 3,772 3,681 1,732 2,009
Table 4-14: ‘Base case’ Contribution to Indirect Jobs
Total jobs (direct plus indirect jobs) are reported in Table 4-15. In this table it can be seen
that total jobs are set to increase from 913 in 2007/08 to 2 206 in 2025/26.
Page 107 of 222

Contribution to Total Jobs
2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2017/18 2025/26

Operational Expenditure 863 927 997 1,163 1,560 1,477 1,926 2,176
Table 4-15: ‘Base case’ Contribution to Total Jobs
4.2.2.2 All options considered
The direct jobs resulting from the ‘all options considered’ scenario are given in Table 4-16.
Here 166 operating jobs are expected in 2007/08. These direct operating jobs are expected
to increase to 3 607 in 2011/12 from the new coal power station, Biomass (invader bush) and
concentrating. As with the ‘base case’ there are no direct jobs created from the fuel
expenditure (they are included in the operational expenditure jobs).
The electricity price in this scenario is initially lower than in the ‘base case’ and then, by
2011/12 goes above the ‘base case’ price. The consequence is that there would be
economic productivity gains, relative to the ‘base case’ in the earlier years and then
productivity losses. These gains and losses are seen as job increase and then job losses
relative to the ‘base case’. The job gains increase from 1 to 209 before becoming job losses
from 2011/12 onwards as more expensive electricity has an effect on productivity in the
Namibian economy.
Contribution to Direct Jobs by Type of Expenditure
2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2017/18 2025/26

Capital Expenditure 0 51 153 145 69 53 53 0
Operational Expenditure 166 169 1,303 2,441 3,607 3,606 3,567 3,574
Fuel Expenditure
Economic Productivity 0 1 200 209 -187 -261 -216 -298
Total Contribution 166 221 1,657 2,795 3,489 3,398 3,404 3,276
Table 4-16: ‘All options considered’ Contribution to Direct Jobs
The number of indirect jobs, as indicated in Table 4-17, is set to increase from 747 in
2007/08 to 2 550 in 2025/26. Operating expenditure is the most consistent contributor to
indirect job creation, with capital expenditure creating periods of high employment.
Contribution to Indirect Jobs
2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2017/18 2025/26

Operational Expenditure 697 760 908 1,209 1,461 1,487 2,308 2,721
Table 4-17: ‘All options considered’ Contribution to Indirect Jobs
Total jobs (direct plus indirect jobs) are reported in Table 4-18. In this table it can be seen
that total jobs are set to increase from 913 in 2007/08 to 5 826 in 2025/26.
Page 108 of 222

Contribution to Total Jobs
2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2017/18 2025/26

Operational Expenditure 863 929 2,212 3,651 5,068 5,093 5,875 6,295
Table 4-18: ‘All options considered’ Contribution to Total Jobs
4.2.2.3 Comparison
The difference in average total jobs is illustrated in Figure 4-25. Here it can be seen that the
‘all options considered’ scenario creates many more jobs than the ‘base case’ scenario,
particularly at the unskilled job level. This is explained by the use of the Biomass (invader
bush) generation option in the ‘all options considered’ scenario, which is labour intensive and
uses predominately unskilled labour.
Average Total Job Creation

6,000
Mixed Income Job

Unskilled Job
5,000 Skilled Job
4,000
Job Numbers
3,000
2,000
1,000
0
Figure 4-25: Comparison between Average Total Job Creation by Skill Level
4.2.3 Balance on Current Account

The generation or use of foreign exchange is an important policy consideration. In all the
generation scenarios there is a drain on foreign exchange. What this section does is illustrate
the differences between the ‘base case’ and ‘all options considered’ scenarios.
The balance of current account (i.e. the difference between imports and exports) for the
‘base case’ and the ‘all options considered’ scenario is illustrated in Figure 4-26.
Page 109 of 222

Balance of Current Account for Base Case and All Options Considered Scenarios
0.0
6
/0
/0
/1
/1
/1
/1
/1
/1
/1
/1
/1
/1
/2
/2
/2
/2
/2
/2
/2
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
-1,000.0
-2,000.0
-3,000.0
NAD millions
-4,000.0
-5,000.0
-6,000.0
-7,000.0
-8,000.0
Figure 4-26: Balance of Current Account for the ‘base case’ and the ‘all options considered’
Scenario
A number of observations can be made:
• For both the ‘base case’ and the all options considered case the value of imports
exceeds the value of exports for all years. This is evidenced by the negative values in
the graph.
• In the earlier years (until 2016/17) the ‘base case’ has a smaller negative balance of
current account than the all options considered. What this means is that the
difference between imports and exports is lower for the ‘base case’ than the all
options considered case. The reason for this is explained by the higher capital costs
of the all options considered scenario, which in turn have a higher import component.
This is particularly so because of the greater generating capacity requirement of the
Baynes hydroelectric power station (about 25% more for the all options considered
scenario than the ‘base case’) and the concentrating solar power stations.
• In the later years (from 2017/18 onwards) the situation reverses and the all options
considered case has a smaller negative balance of current account than the ‘base
case’. The reason for this is that once the power plants in the all options considered
scenario are built Namibia is less dependent on imports for generating electricity, be it
electricity itself or fuel and coal to power the power stations.
• Cumulatively, the ‘base case’ has a worse negative balance of current account than
the all options considered scenario. The ‘base case’ has a cumulative balance of
current account of –NAD 35.6bn, whereas the all options considered scenario has a
cumulative balance of –NAD 34.1bn. What this means is that the higher upfront
import costs more than compensate later on by reducing Namibia’s dependence on
imports in the operations phase.
Page 110 of 222

4.2.4 Electricity Price

The electricity price has already been discussed in the context of all generation scenarios
has already been discussed in section 4.1.2.1. However it is repeated here for convenience
for the two scenarios under discussion.
The comparison between the ‘base case’ and the ‘all options considered’ scenario
distribution customer electricity price is presented in Figure 4-27. It can be seen that in the
early years (2007/08 to 2010/11) the ‘all options considered’ scenario has a lower electricity
price than the ‘base case’. This effect impacts positively on economic productivity and
contributes to lower production costs. However, from 2011/12 onwards the electricity price in
the ‘all options considered’ scenario is more expensive than for the ‘base case’, and this in
turn impacts negatively on economic productivity.
Distribution Customer Price Charges

120.00
110.00
100.00
cents/kWh
90.00
80.00
70.00
60.00
07/08 08/09 09/10 10/11 11/12 12/13 13/14 14/15 15/16 16/17 17/18 18/19 19/20 20/21 21/22 22/23 23/24 24/25 25/26
Figure 4-27: Electricity Price Comparison
4.2.5 Macro Economic Sensitivity Analysis

As with the cost benefit analysis, a sensitivity analysis has been undertaken for the macro
economic comparison between the ‘base case’ and ‘all options considered’ scenarios.
The sensitivities that are tested are:
1. The demand side management interventions. .
2. Namibian economic growth of 6% and 2%.
3. The price of oil, coal and gas doubling or halving over the next five years.
4. The price of Eskom imports doubling or halving over the next five years.
5. Assuming Eskom shortages to continue to 2017.
Page 111 of 222

4.2.5.1 Demand Side Management Sensitivity
In the baseline set of scenarios the ‘base case’ was implemented without demand side
management interventions while the ‘all options considered’ scenario was implemented with
demand side management interventions. We now examine the impact if the ‘base case’ is
implemented with demand side management interventions as well as the impact of the ‘all
options considered’ scenario without demand side management interventions.
Figure 4-28 and Figure 4-29 show the cumulative contribution to GDP and the average total
job creation for both scenarios with and without demand side management. The charts show
that for both the ‘base case’ and the ‘all options considered’ scenarios the cumulative
contribution to GDP and the average total job creation is higher without implementing
demand side management interventions. What this translates into is that there is more
overall spending on additional generation options when the demand side management
interventions are not implemented.
Effect of Energy Efficiency on Contribution to GDP

14.0
12.0
Cumulative Contribution to GDP (N$ millions)
10.0
Without Energy Efficiency

With Energy Efficiency
8.0
6.0
4.0
2.0
0.0
Figure 4-28: Cumulative Contribution to GDP
Page 112 of 222

Effect of Energy Efficiency on Average Total Job Creation

7,000
6,000
5,000
Job Numbers
4,000

3,000
2,000
Without Energy
With Energy
Efficiency
Efficiency
1,000
0
Figure 4-29: Average Total Job Creation
4.2.5.2 Economic Growth Sensitivity
In the baseline set of scenarios it was assumed that the Namibian economy would grow on
average at 3.7% per annum. We now examine the impact if this growth is increased to 6%
and reduced to 2% per annum respectively.
job creation for the three economic growth rate scenarios. The charts show that for both the
’base case’ and the ‘all options considered’ scenarios the cumulative contribution to GDP
and the average total job creation is higher with a higher economic growth rate, in line with
expectations.
Effect of Economic Growth On Gross Domestic Product

14,000.0
3.7% Growth
12,000.0 6.0% Growth
2.0% Growth
Cumulative GDP (N$ millions)
10,000.0
8,000.0
6,000.0
4,000.0
2,000.0
0.0
Figure 4-30: Cumulative Contribution to GDP for Different Economic Growth Rates
Page 113 of 222

Effect of Economic Growth On Average Total Job Creation

7,000
3.7% Growth
6,000 6.0% Growth
2.0% Growth
5,000
Job Numbers
4,000
3,000
2,000
1,000
0
Figure 4-31: Average Total Job Creation for Different Economic Growth Rates
4.2.5.3 Oil, Coal and Gas Price Sensitivity
In the baseline set of scenarios it is assumed that the real price of oil would increase by
about 3% per annum going forward, while the real price of coal and gas would stay constant.
We now examine the impact if the price of all three commodities doubles and halves over the
next five years.
job creation for the three commodity price scenarios. The charts show that for both the ’base
case’ and the ‘all options considered’ scenarios the cumulative contribution to GDP and the
average total job creation is higher for the lower commodity prices. The reason for this is that
with lower commodity prices cheaper electricity can be generated, which would eventually
feed through the economy and result in a lower price of goods and services, thereby
stimulating productivity.
Page 114 of 222

Effect of Commodity Prices On Gross Domestic Product

14,000.0
Baseline Commodity Prices

12,000.0 Commodity Prices Doubling
Commodity Prices Halving
10,000.0
8,000.0
6,000.0
4,000.0
2,000.0
0.0
Figure 4-32: Cumulative Contribution to GDP for Different Commodity Prices
Effect of Commodity Prices On Average Total Job Creation

7,000
Baseline Commodity Prices

6,000 Commodity Prices Doubling
Commodity Prices Halving
5,000
Job Numbers
4,000
3,000
2,000
1,000
0
Figure 4-33: Average Total Job Creation for Different Commodity Prices
4.2.5.4 Eskom Price Increase Sensitivity
In the baseline set of scenarios it is assumed that Eskom would increase their rates along a
standard set of prices where the base energy rate would reach the estimated cost of coal
over a period of ten years. We now examine the impact if these increases are both double
and half the baseline assumptions.
job creation for the three Eskom price scenarios. The charts show that for both the ’base
case’ and the ‘all options considered’ scenarios the cumulative contribution to GDP and the
average total job creation is higher with a lower Eskom price. The reason for this is that the
Page 115 of 222

lower Eskom prices result in cheaper electricity. This eventually feeds through the economy
and results in a lower price of goods and services, thereby stimulating productivity.
Effect of Eskom Prices On Gross Domestic Product

14,000.0
Baseline Eskom Price Increases

12,000.0 Double Baseline Eskom Prices
Half Baseline Eskom Prices
10,000.0
8,000.0
6,000.0
4,000.0
2,000.0
0.0
Figure 4-34: Cumulative Contribution to GDP for Different Eskom Prices
Effect of Eskom Prices On Average Total Job Creation

7,000
Baseline Eskom Price Increases

6,000 Double Baseline Eskom Prices
Half Baseline Eskom Prices
5,000
Job Numbers
4,000
3,000
2,000
1,000
0
Figure 4-35: Average Total Job Creation for Different Eskom Prices
Page 116 of 222

4.2.5.5 Eskom Shortages Sensitivity
In the baseline set of scenarios it is assumed that Eskom shortages will continue up until
2012. We now examine the impact if these shortages continue until 2017 (for the next ten
years).
job creation for the two Eskom shortages scenarios. The charts show that for the ’base case’
the Eskom shortages to 2017 has the higher contribution to GDP than the baseline Eskom
shortages, due to the fact that more generation plants would need to be built. The opposite is
true for the ‘all options considered’ scenario. The reason for this phenomenon is that the
generation requirements for the Eskom shortages to 2017 are more import dependent than
for the baseline shortages case for the ‘all options considered’ scenario. However, as far as
job creation is concerned both the ’base case’ and the ‘all options considered’ scenarios
show higher average total job creation for the longer Eskom shortages, due to the higher
local generation requirements and hence higher operating jobs sustained.
Effect of Eskom Shortages On Gross Domestic Product

14,000.0
Baseline Eskom Shortages

12,000.0
Eskom Shortages to 2017
10,000.0
8,000.0
6,000.0
4,000.0
2,000.0
0.0
Figure 4-36: Cumulative Contribution to GDP for Different Economic Growth Rates
Page 117 of 222

Effect of Eskom Shortages On Average Total Job Creation

7,000
Baseline Eskom Shortages

6,000
Eskom Shortages to 2017
5,000
Job Numbers
4,000
3,000
2,000
1,000
0
Figure 4-37: Average Total Job Creation for Different Economic Growth Rates
4.2.6 Environmental assessment

All supply options will carry various environmental aspects and impacts during the
construction, operation and decommissioning of the plants. How these aspects and impacts
are addressed will be guided by the Namibian Environmental Assessment Policy (1994) and
the Draft Environmental Management Bill of 1998 (Tarr 2003).
In brief, in addition to the air emissions highlighted in this report there are other
environmental issues that will require more in-depth study and management for each of the
options proposed. However, these are not the subject of this report and will not be discussed
in great detail. These issues might include those relating to:
• Other greenhouse gas emissions (other than CO2),
• Mercury emissions,
• Water quality and quantity,
• Land-use management (both at a power station site, but also the land under
transmission lines),
• The use of natural resources,
• Waste management,
• The impacts on flora and fauna species, and
• Social considerations (health and safety)
Page 118 of 222

All scenarios that include the Baynes hydro display a sharp drop in direct emissions when
that power plant comes on-line. However, the Baynes hydro-electric supply option has
various environmental and social aspects that would require attention in both the ’base case’
and 'all options considered.' These include, but are not limited to:
• Ensuring some natural flow of water,
• Evaporation losses due to any damming of water,
• Reclaimed land,
• Loss of livelihood by the semi-nomadic Himba tribe,
• Loss of biomass,
• Minimizing the impacts to endemic flora and fauna species including the loss of
habitat required to sustain these species,
• Population displacement,
• Political and cross-border considerations,
• Loss of cultural, heritage and archaeological sites,
• Flooding of grave sites unless relocated,
• International border agreements and shared water resources, and
• Potential loss and/or benefits to eco-tourism
(Sources: Maritz, P 2000, 'The Baynes Alternative’, Lang 2007; and NAMANG Consortium
1998 'Baynes Environmental Assessment Summary').
4.2.6.1 Base case
When comparing the ‘base case’ and the ‘all options considered’ scenarios, it is apparent
that the ’base case’ relies more heavily on fossil fuels both within Namibia and outside its
borders through Eskom imports. As such, the emissions of CO2, NOx and SOx are reportedly
higher as reported in section 4.1.
Although the Van Eck coal-fired power station is likely to retire after 2012, until that time the
power station is likely to release high levels of air pollution into the atmosphere. Due to the
age of this station it is likely that the plant has some degree of asbestos materials and PCBs
(polychlorinated biphenols) which will have to be accounted for during decommissioning for
the appropriate handling and disposal of these materials.
Additional power could be available in South Africa after 2012 and available for import to
Namibia. However, there are additional environmental aspects associated with the
transmission of power that will need attention both in South Africa and Namibia. These
include:
• The transmission line losses and thus, unwise use of natural resources from the
generation source to the point of use,
Page 119 of 222

• The cost of operation and maintenance of land beneath the long-distance

transmission lines,
• The risks of fire under these lines, the loss of power to Namibia, and
• Wildlife and bird inter-actions of power lines and the loss of power or electricity supply
interruptions to Namibia.
The lack of demand-side management and energy efficiency alternatives in the ‘base case’
scenario avoids the role that the public and private business alike should play in the wise use
of electricity. Without such a programme in place the general public does not take ownership
for their electricity use nor understand the relationship that the generation of electricity has in
its impacts upon the environment.
4.2.6.1.1 Walvis Bay Coal-fired Power Station
The coal for a proposed new coal-fired power station in Walvis Bay would likely be shipped
from Richards Bay in South Africa. Although not addressed in this study it should be noted
that the transportation of the fuel (coal) will increase the overall calculation of Namibia's
carbon foot-print.
This new power station would probably use seawater for cooling as opposed to dry-cooled
currently in place at Van Eck. The proposed seawater flue gas desulphurisation plant would
significantly remove the sulphur content of the flue gas without using lime stone and without
producing gypsum (ECB 2007b). Of possible environmental concern is the suggestion that
the desulphurization plant be dropped from the proposal not only to save costs, but because
'no international environmental standard requires desulphurization for this site' (ECB 2007b,
Attachment 1:11).
Although the detail is unclear from the application submitted to the ECB in 2007, there is a
proposal for low-NOx burners and 'high efficiency ash filters.' The exact values for the
efficiencies proposed for these burners and ash filters are unknown to the authors.
Generally, low-NOx burners can achieve 30-60% NOx reductions while bag filters are more
effective than electrostatic precipitators in achieving collection efficiencies up to 99.5% (IEA
Coal Research 2001). These efficiencies will be influenced by the coal quality burnt at the
station, e.g. sulphur, carbon, ash, and moisture content.
The proposed Walvis Bay coal-fired power station would be designed to collect and treat
various effluents from the site including (ECB 2007b):
• Oil contaminated effluents,
• Chemically contaminated effluents,
• Rain water effluents, and
• Other effluents
Sea water discharge will need to be closely monitored and managed. It is estimated that the
(ECB 2007b):
• Cooling water temperatures discharged could increase + 8 to 8.2 degrees Celsius,
• Sulphates could increase by a maximum of 3%,
Page 120 of 222

• Dissolved oxygen may reach a maximum saturation of 70%, and
• pH could drop to a minimum of 6.
4.2.6.2 All Options considered
As reported in section 4.1 the 'all options considered' scenario achieves the second best
level of emissions. Additionally, see the above discussion at the beginning of section 4.2.6 on
Baynes hydro power supply environmental considerations.
Due to the mix and diversity of supply options proposed under this scenario, the
environmental and social impacts should be less than those of a larger, centralised power
station. Under this scenario, the general public will take greater ownership of the electricity
they use through the use of diverse renewable energy options and DSM.
Included in this option is that of biomass burning for energy. This is not without its share of
environmental concerns. Biomass is assumed to yield no net emissions of CO2 because of
the sequestration of biomass during the planting cycle. However, the entire process
(harvesting, transportation, and the conversion to electricity) can be considered to be a small,
positive net emitter of CO2 (USDOE 2007a, and Power Scorecard 2007, web accessed).
For Namibia, the location of the biomass will not necessarily be where the need for electricity
is on the system. However, more localized, distributed power supply options could be
considered to avoid the high carbon footprint and cost of transporting the biomass great
distances to a centralized power station. Also, smaller, localized systems could support
communities in the area of the station by buying fuel and/or employing people from the
surrounding areas. If planned properly, land for biomass-for-energy production in Namibia
should not have to conflict or compete with land for food production for the country. However,
this would have to be verified and is outside the scope of this study.
In addition to land use management, the carbon footprint of transporting biomass as fuel
from source to power station, and water consumption of irrigated crops, a biomass-to-energy
plant in Namibia will have to address:
• Air emissions,
• Resulting ash production, its storage, handling and disposal,
• Review and possible change of crops produced for fuel, considering water demands
of the crop in question, the harvesting methods, transportation, and its conversion to
electricity),
• Agricultural residues, energy crops, forestry residues, urban wood waste / mill
residues.
4.3 SUMMARY AND DISCUSSION

The preceding analysis is very wide ranging and has considered the electricity generation
scenarios from a range of perspectives. The objective of this section is to summarise and
interpret these findings. As a starting point the key analytical findings are summarised below.
This is then followed by discussion and interpretation.
Page 121 of 222

4.3.1 Summary of results

The key conclusions from the cost benefit analysis are:
• The ‘all options considered’ scenario is the most desirable scenario from an economic
efficiency perspective.
• The ‘all options considered’ scenario remains the most desirable scenario even when
testing sensitivity to the inclusion of demand side management; changes in economic
growth rates analysed; changes in Eskom prices; increased or decreased Eskom
shortages, the value of unserved energy; and the value of carbon credits.
The key conclusions from an electricity price perspective are:
greatest price volatility.
• The ‘maximum supply diversity’ and ‘maximum supply diversity’ scenarios have the
highest electricity prices but are less volatile than the ‘no new local generation’
scenarios.
than the ‘price minimised’ and ’base case’ scenarios.
• From a sensitivity perspective the ‘all options considered’ scenario has a significantly
lower exposure to variations in resource costs than the ‘base case’; and has a lower
sensitivity to South African import constraints than the ‘base case’.
As far as contribution to GDP is concerned:

constructing additional generation plant at the Baynes hydroelectric power station
(more than in any of the other scenarios), and concentrating solar power stations.
‘generation self sufficiency’ scenario at NAD 13.1bn.
• The ‘all options considered’ scenario makes the fourth largest contribution to GDP at
NAD 12bn.
GDP.
From a social perspective two issues are quantifiable – job creation and promotion of
small, medium and micro enterprises (SMME).
For jobs the following conclusions are drawn:
followed by the ‘minimum energy imports’, the ‘generation self-sufficiency scenarios
at 5 851 and the ‘all options considered’ scenario at 5 800.
Page 122 of 222

• The main reason for the high job creation for these four scenarios is the
implementation of the Biomass (invader bush) generation option. This is a labour
intensive generation option, particularly at the unskilled level.
The following conclusions are drawn on the degree of SMME promotion:
• Many of the generation options will not promote much SMME development. These
generation options include the Ruacana power station and the import options, the
Paratus power plant and any nuclear or solar photovoltaic power stations.
• The ‘generation self-sufficiency’ and ‘minimum energy imports’ scenarios score the
highest as far as potential SMME development is concerned. This is largely due to
the contribution from Kudu, Baynes and biomass generation.
• As would be expected the no new local generation scenario (mainly imports) scores
From a renewable energy perspective:
• At the moment nearly one half of all Namibia’s electricity generation comes from
renewable generators in the form of hydro power. In time the ‘maximum renewable’
scenario could contribute between 90% and 100% of all of Namibia’s electricity
needs. In contrast, and as could be expected, the ‘no new local generation’ scenario
has the lowest renewable component.
• If one excludes hydro generation then the non hydro renewable part of the ‘maximum
renewable’ scenario stands at 46% in 2014 to 2017 before dropping to 25% as
Baynes comes into the generation grid.
As far as emissions are concerned:
• On average the ‘maximum renewables’ scenario has the lowest emissions, followed
by the ‘generation self-sufficiency’ and ‘minimum energy imports’ scenarios.
• On average the ‘no new local generation’ scenario has the most emissions, followed
by the 'base case' scenario.
of emissions.
• Considered over time the 'all options considered' scenario achieves the second best
level of emissions in the latter part of the model horizon behind the “maximum
renewables’” scenario.
emission even though these occur in South Africa and not Namibia.
The final set of conclusions relates to a comparison of the ‘base case’ to the ‘all options
considered’ scenario.
Page 123 of 222

• The ‘all options considered’ scenario has a cumulative contribution to GDP of

NAD 12.05bn, compared to the NAD 8.98bn of the ‘base case’ although the ‘all
options considered’ scenario does reduce economic productivity because the
electricity price is higher than that of the ‘base case’.
• At 5 600 average annual jobs the ‘all options considered’ scenario creates many
more jobs than the ‘base case’ at an annual average of 2 200. This is particularly the
case for unskilled job because of the inclusion of the Biomass (invader bush)
generation option in the ‘all options considered’ scenario.
• As far as the price of electricity is concerned the ‘all options considered’ scenario has
a lower electricity price than the ‘base case’ in the early years. However, from
2011/12 onwards the electricity price in the ‘all options considered’ scenario is more
expensive than for the ‘base case’.
4.3.2 Comparative Discussion

The previous section has compared the results for each type of analysis. What this section
does is take each of the generation scenarios and consider its merits and demerits. In order
to assist this discussion a summary table has been prepared in the form of Table 4-19. This
table summaries the analysis for each generation scenario. It also introduces two addition
variables. These are
• The average price of electricity but adjusted for potential carbon credits.
• Unserved energy over the next 20 years, over the next 7 years and the unserved
energy as a percentage of total energy demand over the next 20 years.
For convenience the table also includes a colour coded ranking of each variable. The dark
green shading is for the best ranking of each variable, the light green shading for the second
best and the pink shading is for the worst ranking. For example, looking at the first indicators
of cumulative contribution to GDP, the ‘renewables maximised’ scenario contributes the most
to GDP, followed by the ‘minimum energy imports’ scenario. These two scenarios are coded
dark green and light green under the contribution to GDP column respectively. The ‘no new
local generation’ scenario contributes the least to GDP and is therefore coded pink.
The ‘no new local generation’ scenario is unequivocally the worst policy option. It makes the
lowest contribution to GDP; has very few jobs; does not promote SMMEs; has massive
environmental emissions; and has an unserved energy equal to nearly 10% of electricity
demand. It is the most beneficial for the current account because it does not require the
importation of capital equipment; and has a relatively low average price.
In contrast to the worst policy option there is no clear winner. The ‘base case’ has a good
electricity price but scores poorly in contribution to GDP, job creation, current account; and
with unserved energy. It has a very low benefit cost ratio, meaning it is not the most desirable
economically. It attracts no carbon credits and is the second worst polluter.
The ‘price minimised’ scenario is potentially a good choice because it has a good benefit cost
ratio; creates many jobs; has the lowest price after ‘base case’ and ‘no new local generation;
it has the second lowest price if carbon credits are included. On the other hand it is the third
worst polluter and does not make the most of renewable generation options.
The ‘maximum supply diversity’ option comes top in ensuring that there is as little unserved
energy as possible. However, in contrast, it also results in the second most expensive way to
generate electricity and is the most expensive if carbon credits are factored in.
Page 124 of 222

The ‘generation self sufficiency’ option creates many jobs; promotes SMMEs; has low
environmental emissions; and low levels of unserved energy. However from an economic
perspective it has the second lowest benefit cost ratio. This implies that economic criteria
would weigh heavily against this option.
The ‘minimum energy imports’ scenario is also economically inefficient with a benefit cost
ratio the same as that of ‘generation self sufficiency’. Surprisingly it is also the worst culprit
on the current account because, although very little electricity is being imported, the capital
equipment and necessary fuel are being imported. It is also an expensive option which is
why is makes a good contribution to GDP. On the other hand it would promote SMMEs and
has low levels of emissions.
The ‘renewables maximised’ has, apart from SOx, the lowest levels of emissions and, as can
be expected, the highest carbon credits. It creates the most jobs but at the cost of the highest
priced electricity. It is reasonably economically efficient with a benefit cost ratio of 2.3 but has
the third highest level of unserved energy. It is also expensive which is why it makes such a
large contribution to GDP.
As the above results became evident during the research process the conclusion that was
reached was that a balanced approach to electricity generation and demand side
management is probably the most desirable option. In the study this became known as the
‘all options considered’ scenario. This scenario is the most efficient economic choice with the
highest cost benefit ratio. It makes a good contribution to the economy; creates a handsome
number of jobs; has the second lowest impact on the current account; promotes SMMEs; has
a reasonable electricity price and is not an excessive polluter having the second lowest
emissions of CO2. All this is achieved for very low levels of unserved energy.
Expend Direct Jobs
Account (N$ billion)

Balance of Current
Benefit Cost Ratio
Price (incl Carbon
Environment CO2
Price (cents/kWh)
Environment NOx
Environment SOx
Unserved Energy
Unserved Energy
SMME Ranking
Impact on GDP
Carbon Credits
(GWh) - 20 yrs
% Unserved to
Ave Electricity
Ave Electricity
Total Demand
Total Direct &
(GWh) - 7 yrs
Indirect Jobs
(Mega-tons)
Operational
(N$ billion)
(N$ billion)
(kilo-tons)
(kilo-tons)
credits)
Num Scenario Description

1 Base Case Without EE 9.0 186 2,259 -35.6 2.1 1.2 98.5 98.5 98.2 88.6 56.6 0.0 528 524 1.8%
2 No New Local Generation 3.0 98 1,025 -29.8 0.1 96.0 96.0 145.3 134.2 78.5 0.0 2,897 2,717 9.7%
3 Price Minimised 10.6 3,067 5,493 -35.5 2.4 1.5 99.8 98.3 82.8 74.6 47.8 1.6 163 161 0.5%
4 Maximum Supply Diversity 7.5 3,038 4,406 -38.3 2.2 1.5 110.1 106.2 67.9 46.4 40.2 4.0 61 61 0.2%
5 Generation Self-Sufficiency 13.1 3,038 5,851 -39.3 2.0 2.4 106.8 104.5 53.4 25.4 32.1 2.7 86 86 0.3%
6 Minimum Energy Imports 13.1 3,037 5,850 -39.3 2.0 2.4 106.8 104.5 54.0 26.1 32.3 2.7 151 128 0.5%
7 Renewables Maximised 14.0 5,763 8,553 -38.9 2.3 2.1 110.2 104.9 45.9 43.7 21.2 5.7 250 169 0.9%
8 All Options Considered With EE12.1 3,027 5,685 -34.1 2.7 1.8 102.8 100.1 56.0 51.1 31.1 2.7 116 88 0.4%
Table 4-19: Scenario evaluation matrix
Page 125 of 222

5 Conclusions
within the context of the 1998 White Paper on Energy Policy. Here it was important to
determine the technical feasibility and economic implications of the various options. The third
was to explore the potential and role of renewable resources in these generation mixes.
It has been shown that Namibia faces a pending electricity crisis. If Namibia continues to rely
on imported electricity and does not invest in local generation capacity then unserved
energy, power outages in other words, could be as high as 9.7% of total demand. Peaking in
2012/13 cumulative unserved demand could be as high as 10.8% of total demand. It has
also been shown that, based on some specific assumptions that a 24 hour blackout for one
day a month, the equivalent of a 3.29% drop in electricity supply, would reduce Gross Value
Added by up to 3.7%. These calculations suggest that there could be some dramatic impacts
on the Namibian economy because of the shortage of electricity.
The Energy White Paper (MME 1998) identifies a number of policy options including self-
sufficiency, security of supply, inclusion of renewable energy source, sustainability and cost
effectiveness and efficiency. Each of these has been explored separately in this study within
the context of eight separate scenarios. What has been found is that there is no, one, single
policy option that satisfies all of the policy criteria. Some of the options are economically
efficient but do not maximise the use of renewable resources. Others are cheap but do not
guarantee security of supply. Others, in turn, maximise the use of renewable resources but
are expensive. The consequence is that there is trade off between the various options and
the final decision needs to taken at the political level. What this report does is to inform on
the consequences of the various policy decisions.
There is however one conclusion that is unequivocal. Demand side management measures
should be introduced as a matter of urgency. They have been shown to be cost effective.
They are desirable from an environmental perspective. Probably the most important short
term advantage is they will help to alleviate power outages and in the long term reduce the
need for generation capacity. The modelling indicates that the investment needed for
demand side management breaks even at an unserved energy value between NAD 2/kWh
and NAD 10/kWh. Both these values are substantially less than the “low” value attached to
unserved energy in South Africa which has been assumed at R20/kWh in the 2004 NIRP.
There is a second conclusion that is almost as unequivocal although more muted.

Generation from renewable resources is desirable. It is environmentally responsible, will
create jobs, makes an important contribution to the economy and ranks well from an
economic efficiency perspective. However this option is not without risks and cannot be
expected to supply all Namibia’s electricity need, at least in the short term. The ‘all options
considered’ scenario demonstrates that, excluding hydro, it is possible to have renewable
energy producing 20% of electricity needs without escalating end consumer prices more than
about 5% above the cost of importing electricity from South Africa. It also avoids the power
outage problems having unserved energy of as little as 0.4% of total demand.
From an economic perspective the least desirable option is relying on imports from South
Africa. While this option has the lowest average price it also has the highest amount of
unserved energy and therefore the worst benefit cost ratio. Probably the most important
observation that can be made is that such an option would continue to leave Namibia at the
mercy of South African politics and policy.
Page 126 of 222

One policy option is that of security of supply. What has been found in this study is that it has
a higher cost, because it means more diversified supply sources and larger spare capacities.
It also means making use of both the cheapest supply sources as well as small amounts of
alternate sources that have little or no correlated risks with the mainstream supply sources.
This option has one of the worst impacts on the current account, has the second lowest
benefit cost ratio and one of the highest electricity prices.
What this study has shown, and where the research moved to as the results became evident,
is that a balanced approach to electricity generation and demand side management is
probably the most desirable option. In this study this became known as the ‘all options
considered’ scenario. This generation mix relies on a small coal plant, biomass to electricity
plants and Baynes hydro as the backbone of base load supply. This is complemented by an
array of diverse renewable generators, each limited in size to ensure grid stability and cost
effective grid integration by avoiding grid issues that would arise for larger versions of the
same plant type. This scenario also implements demand side management. This option is
the most efficient and desirable from an economic perspective. It makes a contribution to the
economy which is little different to the top contributing scenarios. It creates a handsome
number of jobs although not as many as the ‘renewables maximised’ scenario. It has the
least power outages, low environmental emissions and average electricity price that it not
that far out of line with the cheapest options.
Namibia stands at a cross roads and the price of not taking action could be very indeed. We
trust that the work that has been done in this study will assist policy makers in making
informed decisions that will be beneficial for all Namibians.
Page 127 of 222

References
AGAMA Energy 2003, Employment Potential of Renewable Energy in South Africa.

Sustainable Energy and Climate Change Partnership.
American Council for an Energy-efficient Economy (ACEEE), Web accessed August 2007,
http://www.aceee.org.
Argonne National Laboratory 2002, Prepared for The World Bank, ‘Turkey Energy and
Environmental Review’, Task 7: Energy Sector Modelling, Executive Summary, August 2002.
Australian Uranium Association 2007, ‘The economics of Nuclear Power’, Briefing Paper 8,
Melbourne, Australia.
Awerbuch, S. and Berger, M. (2003): Applying portfolio theory to EU electricity planning and
policy-making. IEA/EET working paper. EET/2003/03.
Bicon Namibia 1999, Assessment of solar and wind resources in Namibia, Ministry of Mines
and Energy/GTZ. REEE1/98.
Black & Veatch for the United States National Renewable Energy Laboratory 2006,
NREL/SR-550-39291, April 2006.
Bradford,T and Maycock, P 2007, ‘PV market update’, Renewable Energy World, vol 10, no
4, p60.
Bringezu, Stefan, S Ramesohl, K Arnold, M Fischedick, J von Geibler, C Liedtke, and H

Schutz 2007, for Wuppertal Institute for Climate, Environment and Energy, 'What we know
and what we should know - Towards a sustainable biomass strategy', Discussion Paper, No.
163 - June 2007, <http:// http://www.wupperinst.org/uploads/tx_wibeitrag/WP163.pdf>.
Brown, Lester March 2003, 'Restructuring the Global Energy Economy', The Globalist, Web
accessed August 2007, http://www.theglobalist.com/DBWeb/StoryId.aspx?StoryId=2991.
Cameron A and Jones J 2007, ‘Solar movement’, Renewable Energy World, vol 10 no 4,
p168.
Carstens, Rinus 2007, ‘Van Eck Power Station’. Presentation at the World Energy Council
and UPDEA Workshop. NamPower, Windhoek, Namibia. 12-13 September 2007.
Climate Change 20 March 2007. Environment Canada, Quebec, Canada, <

http://www.ec.gc.ca/default.asp?lang=En&n=6EE576BE-1>.
Conningarth (2007) Manual for Cost Benefit Analysis in South Africa.
CSIR Environmentek and NamPower 2004, Environmental Assessment and Environmental

Management Plan for the Proposed Kudu Combined Cycle Gas Turbine (CCGT) Power
Plant at Oranjemund – Background Information.
Page 128 of 222


Plant at Oranjemund, Volume 1: Final Environmental Impact Report, CSIR Report No. ENV-
S-C 2004 073.

Plant at Oranjemund, Volume 2: Specialist Studies, CSIR Report No. ENV-S-C 2005 057.
DECON 1999, ‘Project Studies for Wind Parks in Walvis Bay and Lüderitz’, Final report
prepared for GTZ.
Department of Energy, South Africa 2008, National Response to South Africa’s Electricity
Shortage.
Directorate of Environmental Affairs <http://www.dea.met.gov.na>.
ECB 2006, Demand Side Management Study for Namibia – Report 2: Implementation Plans
for Six DSM Options.
ECB 2007a, Effects of cost reflective electricity tariffs in Namibia.
ECB 2007b, Coal-fired power station at Walvis Bay. License Application, Binvest.
Eilperin, Juliet 15 February 2008, 'Study Finds Humans' Effect on Oceans Comprehensive',
Washington Post, Washington, DC, USA, <http://www.washingtonpost.com/wp-
dyn/content/article/2008/02/14/AR2008021401992.html?wpisrc=newsletter>, p. 11.
Electric Power Research Institute (EPRI), Web accessed August 2007, http://www.epri.com.
Emcon Consulting Group 2005, Assessment of Feasibility for the Replacement of Electrical
Water Heaters with Solar Water Heaters, Prepared for the Ministry of Mines and Energy,
Namibia.
Energy and Environmental Analysis 2000, Environmental Benefits of Distributed Generation

(Draft).
Energy July 13, 2007, Environmental Literacy Council, Washington, DC, Web site accessed
November 2007, < http://www.enviroliteracy.org/category.php/4.html>.
Energy Watch Group 2007, Coal: Resources and Future Production, EWG Paper No 1/07,
Updated version 10 July 2007.
Enermodal Engineering Limited and Marbek Resource Consultants Ltd 1999, Cost reduction
study for solar thermal plants, Final report prepared for World Bank.
Eskom 2008, Reduction in Water Consumption, Web accessed March 2008,

http://www.eskom.co.za.
European Commission 2007, Concentrating Solar Power – From Research to

Implementation, Luxembourg: Office for Official Publications of the European Communities,
Belgium.
Page 129 of 222

European Commission December 2007, 'Nuclear Issues', European Commission Energy,

Web accessed 9 March 2008,
http://ec.europa.eu/energy/nuclear/decommissioning/index_en.htm.
European Commission Energy Research 2007, Web accessed August 2007,

http://ec.europa.eu/research/energy.
EWEA 2004, European Wind Energy Association, Facts on Wind.
Government of Victoria 2003, The Economic Impacts of Energy Efficiency, Australia.
Hager, C, Schultz, R and von Oertzen D. 2007a, Turning Namibian Invader Bush into
Electricity. 12th Congress of the Agricultural Scientific Society of Namibia.
Hager, C, Schultz, R and von Oertzen D. 2007b, Costing calculations for Invader Bush to
Electricity, Spreadsheet, Desert Research Foundation of Namibia.
Hohmeyer, Olav, Florian Wetzig, and David Mora ?2007, ‘Wind Energy - The Facts’,
University of Flensburg, Germany. Environment, vol 4.
Housing Census 2004 (National Planning Commission, Namibia).
IEA 2007, ‘World Energy Outlook 2006’, IEA, Paris, France, Web accessed 5 March 2008,
http://www.iea.org/textbase/press/pressdetail.asp?PRESS_REL_ID=187.
International Energy Agency Coal Research 2001, ‘The Clean Coal Centre, Profile. Air
Pollution Control Costs for Coal-fired Power Stations’.
International Energy Agency 2007, Web accessed August 2007, http://www.iea.org.
Jargstorf B 2004, ‘Wind regimes of Africa’, InWent.
Jindal, Rohit 2006, ‘Carbon Sequestration Projects in Africa: Potential Benefits and
Challenges to Scaling Up’, for World Resources Institute.
Jones, J 2007, ‘CSP lifts off’, Renewable Energy World, vol 10 no 3, p36.
Labour Force Survey 2004, (Namibia Ministry of Labour).
Lang, Steve 14 September 2007, IPS News, Viewed 14 September 2007, 'Southern Africa:
Plans to dam the Cunene River are shelved again', <http:// www.ipsnews.net/>.
Laumer, John 4 March 2007 'There is no free trade: cargo shippers emit twice the CO2 of
airlines, yet go completely unregulated', Treehugger, Brooklyn, New York, Web accessed 5
March 2008, < http://www.treehugger.com/files/2007/03/there_is_no_fre.php>.
Low Impact Hydropower Institute 2007. Portland, Maine (USA), Web accessed March 2008,
<http:// www.lowimpacthydro.org>.
Maritz, P 2000, ‘The Baynes Alternative’, Presentation at NamPower, Namibia.
Maurer, Crescencia 2003, ‘Financing Carbon: Export Credit Agencies and Climate Change’,
for World Resources Institute.
Mendelsohn, J. et.al. (2002): Atlas of Namibia – A Portrait of the Land and its People.
Published for the Ministry of Environment and Tourism of Namibia by David Philip.
Page 130 of 222

Ministry of Environment and Tourism <http://www.met.gov.na/index>.
Ministry of Mines and Energy 1998, Energy White Paper of Namibia.
Morthorst, PE ?2004, ‘Wind Energy – The Facts, Volume 2, Costs and Prices’, Published by
European Wind Energy Association.
NAMANG (Consortium of NorConsult, SwedPower, Burmeister & Partners, and SOAPRO)

1998. Summary of the environmental assessment report on the Baynes project. A
component of the Feasibility Study Report on the proposed Lower Cunene hydropower
scheme.
Namibian Draft Environmental Management Bill 1998, (December 1998, a sixth and ‘final’
draft of the Bill).
Namibian Ministry of Environment and Tourism (MET) 2007 The National Climate Change
Program, Windhoek, Namibia, Web accessed March 2008, http://www.met.gov.na.
Namibian Ministry of Environment and Tourism (MET) 2007, Directorate of Environmental

Affairs, web accessed August 2007 and March 2008.
Namibian Ministry of Environment and Tourism 1995, Namibia’s Environmental Assessment

Policy.
Namibian Ministry of Environment and Tourism 2007, Directorate of Environmental Affairs,

web accessed August 2007.
Namibian Ministry of Environment and Tourism, 2003, Namibia and Climate Change.
Booklet.
NamPower 2002, Generation Investment Plan Draft 04-09-02.
NamPower 2006a, Transmission Country Report 2006, presentation slides.
NamPower 2006b. Generation Master Plan Update May 2006.
NamPower 2007 and 2008, Web accessed various times. http://www.nampower.com.na.
NamPower 2007, Kudu Power Project, presentation slides by Margaret van der Merwe on 16
October 2007 at EPC Contracting, FIDIC in Windhoek.
National Association for Clean Air n.d. The air that we breathe - a source book on air
pollution for learners.
National Environmental Network, Web accessed August 2007,

http://www.net.org/energy/impacts.vtml.
NEPRU 2004, ‘The Namibian Economic Policy Research Unit’, Working Paper No. 97,
October 2004.
NER 2004, National Integrated Resource Plan 2.
NERC 2006, Generating Availability Report November 2006, Historical availability statistics
1982 – 2005, Fossil Gas Primary – all MW sizes.
NREL 2004, ‘PV FAQs: What is the Energy Payback for PV?’.
Page 131 of 222

Oak Ridge National Laboratory n.d., 'Biomass fuels, energy carbon, and global climate
change', Web accessed 9 March 2008,
http://www.ornl.gov/info/ornlreview/rev28_2/text/bio.htm.
Ocheltree, M 2006, Energy Issue Brief. Carnegie Endowment for International Peace.
OECD Nuclear Energy, International Atomic Energy Agency 2003, Uranium 2003,
Resources, Production and Demand. Quoted from Renewable Energy World, vol 10 no 1,
p58.
Orjala, M and Seppälä R 2007, ‘Power Generation: Feasibility of biofuels boiler and cofiring’.
Chemitec consulting oy, Research report VTT-R-07761-07.
PBMR 2006, Pebble Bed Modular Reactor, Brochure published by PBMR (Pty) Ltd.
Pearce, David 2001, ‘Energy Policy and Externalities: An Overview’. University College of
London. Paper prepared for OECD Nuclear Energy Agency. Keynote address to Workshop
on Energy Policy and Externalities: the Life Cycle Analysis Approach. Paris. November 15-16
2001.
Population and Housing Census 2001, National Planning Commission, Namibia.
Power Scorecard 2007, Electricity from Biomass, Pace University, White Plains, New York,
Web accessed December 2007, www.powerscorecard.org.
Renewables 2007, ‘Global Status Report’, REN21.
Rich, David 2008, ‘Designing a Greenhouse Gas Emissions Registry’, for the World
Resources Institute, 8 pp.
RWE 2005, World Energy Report 2005.
SAD-ELEC 2006. ‘Electricity Prices in Southern and East Africa’.
Scholes, Dr. B, M van der Merwe, M de Wit, M van Tienhoven with contributions from J
Hoffman, K Schade, and D Motinga 2003, 'Namibia and Climate Change.' Booklet. For the
Namibian Directorate of Environmental Affairs, Ministry of Environment and Tourism,
Windhoek, Namibia.
Science in Africa 2004, 'Alternative fuel could significantly reduce CO2 emissions', Web
accessed 9 March 2008, http://www.scienceinafrica.co.za/2004/june/biomass.htm.
Silverstein, Ken 12 November 2007 Climate Change Dexeriy', EnergyBiz Insider. Energy
Central, Auroa, Colorado, Web site viewed November 2007, <
http://www.energycentral.com/centers/energybiz/ebi_detail.cfm?id=411>.
Solar Energy International n.d., 'Energy Facts', Solar Energy International, Carbondale,
Colorado, USA, Accessed 6 March 2008, http://www.solarenergy.org/resources/
energyfacts.html.
Solel 2007, 'Ten facts about solar thermal power', Solel, Israel, Web accessed December
2007, http://www.solel.com/faq/.
South African Department of Environmental Affairs and Tourism 2000, White Paper on
Integrated Pollution and Waste Management for South Africa, no. 20978.
Page 132 of 222

SouthSouthNorth 2005, 'Kuyasa Low-cost Urban Housing Energy Upgrade Project -

Khayelitsha (Cape Town) South Africa', SouthSouthNorth, Cape Town, South Africa,
Accessed 6 March 2008, http://www.southsouthnorth.org/download.asp?name=SA%20
Kuyasa_Profile11.2005.doc&size=1485312&file=documents/SA%20Kuyasa_Profile11.2005.
doc.
SouthSouthNorth 2005, 'Kuyasa Low-cost Urban Housing Energy Upgrade Project.
Stoddard, L., Abiecunas,J. and O' Çonell R. 2006. Economic, Energy, and Environmental
Benefits of Concentrating Solar Power in California - May 2005-April 2006. Black & Veatch
for the United States National Renewable Energy Laboratory April 2006. NREL/SR-550-
39291.
Tarr, Peter and Jacquie Tarr 2003, 'Namibia.' In Southern African Institute for Environmental
Assessment: Environmental Impact Assessment in Southern Africa, Windhoek: Southern
African Institute for Environmental Assessment, 26 pp.
Tarr, Peter April 1994, 'Namibia's Environmental Assessment Policy', Compiled and edited
for the Directorate of Environmental Affairs, Ministry of Environment and Tourism, Windhoek,
Namibia.
Thomas, S 2005, ‘The economics of nuclear power: analysis of recent studies’, Public
Services International Research Unit.
Tullow/NamCor 2006, Kudu Gas Project, presentation at the Fossil Fuel Foundation
Colloquium Natural Gas – Southern Africa, Johannesburg, 20 September 2006.
United Stated Environmental Protection Agency April 2006, Clean energy- environment
guide to action - policies, best practices, and action steps for states.
United States Department of Energy (USDOE) 1999, Issues in Midterm Forecasting and
Analysis 1999, Web accessed June 2007,
<http://www.eia.doe.gov/oiaf/issues/aeoissues.html>.
United States Department of Energy (USDOE) 2007, Tracking New Coal Fired Power Plants,
Presentation slides dating 1 May 2007.
United States Department of Energy (USDOE) 2007a, Energy Information Administration,

Energy Facts, 'Energy Efficiency’, Web accessed August and November 2007,
http://www.eia.doe.gov/.
United States Department of Energy (USDOE) Energy Efficiency and Renewable Energy
January 2006, 'PV Panel Disposal and Recycling', Web accessed 6 March 2008, <
http://www1.eere.energy.gov/solar/panel_disposal.html>.
United States Department of Energy (USDOE) May 2007, Energy Information Administration,
Energy Facts, 'Wind Energy - Energy from Moving Air'’, Web accessed August 2007,
<http://www.eia.doe.gov/>.
USDOE Energy Information Administration n.d., 'Nuclear Power and the Environment',
USDOE EIA Nuclear Issues Paper, Washington, DC, Web accessed 9 March 2008,
http://www.eia.doe.gov/cneaf/nuclear/page/nuclearenvissues.html.
Page 133 of 222

Voss, A 2000 study (cited in Bertel, E and P Fraser 2002, 'Energy policy and externalities',
Nuclear Energy News, Volume 20, Issue 1, p.4,
<http://www.nea.fr/html/pub/newsletter/2002/20-1-Energy_policy.pdf>).
Wikepedia 2008a. List of Countries by Ratio of GDP to Carbon Dioxide Emissions. Web
accessed 5 March 2008, http://www.wikipedia.org.
Wikepedia 2008b. ‘Clean Development Mechanism’, Web accessed 9 March 2008,

http://en.wikipedia.org/wiki/Clean_Development_Mechanism.
Wikipedia 2008c, Pebble bed reactor, Web accessed 5 March 2008, <
http://en.wikipedia.org/wiki/Pebble_bed_reactor#South_Africa>.
WNA report (2005), The new economics of nuclear power, World Nuclear Association.
World Association of Nuclear Operators 2005, < http://www.wano.org.uk/>.
World Business Council for Sustainable Development, Web accessed August 2007,
http://www.wbcsd.org.
World Coal Institute, Web accessed August 2007, <

http://www.worldcoal.org/pages/content/index.asp?PageID=127>.
World Energy Council 2004, Comparison of Energy Systems Using Life Cycle Assessment.
World Resources Institute 2007, ‘Earth Trends: Environmental information’,

www.earthtrends.wri.org, Washington, DC: World Resources Institute.
WorldWatch Institute 2008, ‘State of the World 2008’.
Page 134 of 222

Appendix A. Terms of Reference

Terms of Reference checklist
Scope of Work as per TOR Response to TOR
1) Describe and quantify the impact on Addressed through the various scenarios
Namibia’s economy that results from in different degrees.
Namibia’s dependence on imports of electricity
as well as coal and diesel for electricity Sensitivity was conducted for the price of
generation. A sensitivity analysis of the most coal, oil and diesel, as well as the length
important parameters shall be undertaken. of period of importing electricity from
Eskom.
The annual NAD amounts leaving the

country for electricity imports, generation
oil and generation coal have been
quantified per scenario.
Status: Addressed
In section: 4.1
2) Assess, describe and quantify, per The impact on GDP of supply

economic sector, how regular electricity supply interruptions at different levels of severity
interruptions have and will influence the has been modelled through the different
competitiveness of the Namibian economy, scenarios and per sector as per Social
Namibia’s gross domestic product, foreign Accounting Matrix
direct investment, balance of trade,
employment and job creation. Employment and job creation: No – why
not Any estimate would be pure
speculation. Some supply interruptions
could be absorbed without leading to job
losses, even if there was a reduction in
output.
Competitiveness and foreign direct

investment were excluded with
agreement from the client because these
two factors are the result of a wide range
of factors, not just electricity.
Status: Addressed
In section: 4.1
3) Assess, describe and quantify, per Inherently addressed in the modelling

economic sector, how annual increases of the through the dispatch of different
Page 135 of 222

electricity price (peaking at NAD 0.58 per kWh generators and resulting tariff changes.
by 2011 as per the projection of Namibia’s This is followed through into the economic
Electricity Control Board) influences the model where the impact of increased
competitiveness of the Namibian economy, electricity prices is determined on output
Namibia’s gross domestic product, foreign (by making use of appropriate elasticities
direct investment, balance of trade, per economic sector). The results are
employment and job creation. inherently included in the output of each
scenario.
Competitiveness and foreign direct

investment were excluded with
agreement from the client because these
two factors are the result of a wide range
of factors, not just electricity.
Status: Addressed
Addressed in section: 4.1
4) Assess and quantify the capacity and Addressed in principle through a base
energy needs of the Namibian economy case growth percentage (3.7%) and a
should it grow at 1%, 2%, 3%, 4%, 5%, 6%, sensitivity analysis around this base case
7% and 10% per annum for the next ten years. (2% and 6%). 2% is seen as very low
growth and 6% as the maximum likely to
be achieved in reality.
The project leader agreed to this

approach, agreeing that a 10% growth
rate is unrealistic and that the trends can
be shown through the above approach.
Status: Addressed
Addressed in section: 4.1.2
5) Assess and quantify the electricity The project leader agreed to this being
requirements for Namibia to achieve the Vision dropped from the study
2030 goals.
6) Develop scenarios and assess and quantify Using the Business As Usual scenario,
the impact on the Namibian economy this was modelled through constraining
assuming that from 2007, the country’s ability Eskom imports and changing the time
to import electricity from the Southern African horizons for the constraints. Different
Power Pool is reduced by 10%, 20%, 40% and scenarios also imported different amounts
60%. of electricity from Eskom. For example,
the Generation Self-Sufficiency and
Minimum Energy Imports scenarios
imported far less electricity from Eskom
than the ‘base case’ scenario.
The import constraints were based on

realistic expectations and recent
developments were taken into account
Page 136 of 222

(e.g. Hwange)
Status: Addressed
7) Assess and quantify the impact on the Addressed through a range of different
Namibian economy as well as the short-, scenarios while taking cognisance of
medium- and long-term costs and benefits suitable and mature technologies.
associated with the introduction of large-scale
(i.e. tens or even hundreds of MW) centralised Characteristics of each supply technology
generation capacity, including the costs of are discussed in detail.
transmission infrastructure, by making
reference to the envisaged: Inga hydro-plants was removed from the
scope of work
Kudu combined-cycle plant (as well as in
combination with a solar thermal power Status: Addressed
station)
Addressed in section: 4.1 and Appendix C
Baynes hydro-plant
coal-fired plant at Walvis Bay (as well as in

combination with a solar thermal power
station)
Inga hydro-plant in the Democratic Republic of

the Congo
nuclear power station
large-scale wind farm at Lüderitz (as well as in

combination with pumped storage)
solar thermal power station (assessing the

most commonly commercially available
technologies) in the Namib, and
biomass conversion plants for utilising invader

bush.
8) Assess and quantify the impact on the Addressed through a range of different
Namibian economy as well as the short-, scenarios while taking cognisance of
medium- and long-term costs and benefits suitable and mature technologies.
associated with the introduction of small-scale
(i.e. a few or even tens of MW) decentralised Characteristics of each supply technology
renewable energy plants by making reference are discussed in detail.
to possible:
Mini-hydro plants were removed from the
Biomass conversion plants using invader bush scope of work.
Solar photovoltaic plants Status: Addressed
Solar thermal plants Addressed in section: 4.1 and Appendix C
Page 137 of 222

Wind farms
Mini-hydro plants, and others.
9) With reference to Namibia’s wind, solar and Wind, solar and biomass resources and
biomass resources, assess and quantify the potential contribution were addressed.
contributions that each of these renewable
energy sources has in providing a meaningful Status: Addressed
contribution to Namibia’s energy supply
situation in general, and specifically Namibia’s Addressed in section: 4.1.6
electricity supply capacity.
10) Model the impact on the Namibian This was addressed through assessing
economy if 10%, 15%, 25% and 35% of the realistic impacts of a range of DSM
current electricity consumption is saved by options, based on the Demand Side
introducing energy efficient technologies and Management Study (Emcon 2006). DSM
practices. options were activated in selected
scenarios resulting in a range of energy
savings outcomes.
Status: Addressed for more realistic

percentages.
11) Develop scenarios and quantify the costs, This has been addressed in different
benefits and opportunity costs associated with scenarios where new generation is
a rapid (e.g. within the next 5 years) versus a dispatch more rapidly (Business as
slow (e.g. within the next 25 years) Usual) and slower (Maximum
introduction and adoption of renewable energy Renewable). The costs and benefits to
technologies in Namibia. society were quantified and the
opportunity cost calculated by making use
of a social discount rate (real rate of 8%).
A benefit cost ratio (BCR) was used to
rank and compare the alternative
scenarios to one another
Status: Addressed
12) Assess, quantify and compare the total Inherently modelled through the different
cost per kWh, including generation, scenarios, specifically in the difference
transmission and distribution costs that would between “business as usual” versus “all
have to be paid by urban and rural Namibian options considered”
consumers, if large-scale centralised
conventional electricity generating options Status: Addressed
versus small-scale decentralised renewable
energy generating options were taken into Addressed in section: 4.1
account.
13) Identify, assess and quantify which energy The bid submitted by the Team proposed
Page 138 of 222

efficiency technologies and practices could to look at three DSM measures, which is
meaningfully reduce the current electricity done in this study.
consumption in the Namibian residential,
commercial, tourism, mining, fishing and The project leader agreed that the
agricultural sectors, taking prevailing socio- sectoral analysis was dropped in the final
economic realities into account. scope
Status: Addressed as per proposal
14) Identify, assess and quantify the costs and Based on the DSM study, the most
benefits of introducing energy efficiency promising measures were modelled and
technologies to reduce the country’s peak cost/benefit calculated.
electricity demand.
Status: Addressed
15) Assess and quantify the potential and Removed from final Scope of Work
impact of improved building regulations with
regard to energy efficiency (including but not
limited to compulsory installation of solar water
heaters, double glazing and double walling) on
domestic, commercial and industrial electricity
consumption in Namibia.
16) Recommend how electricity consumption Removed from final Scope of Work
is to be reduced in Namibia’s government
institutions.
17) Assess, describe and quantify what Removed from final Scope of Work
support measures (tax and other incentives)
could be introduced to promote greater uptake
of renewable energy and energy efficiency
technologies and measures and reduced
uptake of energy inefficient technologies.
18) Assess, describe and quantify the impact Removed from final Scope of Work
on Namibia’s economy that different SACU
revenues would have if Namibia could
generate an additional 10%, 25%, 50% and
100% of its existing electricity generation
capacity by means of conventional and/or
renewable energies.
19) By developing different scenarios, quantify The study reports on Namibian Balance
if and when cash outflows currently occurring of Current Account for the ‘base case’
due to the import of electricity are balanced by and ‘all options considered’ generation
savings made through the introduction of scenario for operating costs. The capital
renewable energy technologies within costs were excluded because there is no
Namibian borders. consensus about whether projects would
Page 139 of 222

be funded from Namibia, by Namibia

borrowing offshore or through foreign
investment.
20) Assess, describe and quantify the Removed from final Scope of Work
opportunities that the increased use of
renewable energy technologies has on
Namibia taking the country’s future
participation in international trade in carbon
credits into account.
21) Recommend how, and in which sectors of Based on the DSM study options. Mainly
the Namibian economy, priority residential sector technologies were
implementation of energy efficient modelled because this is where concrete
technologies and practices should be data is available to estimate costs and
introduced, and quantify the macro-economic scope. The macro economic impact has
impact that such an introduction would have. been calculated.
Status: Addressed
22) Assess and quantify Namibia’s agricultural Removed from final Scope of Work
potential to provide liquid fuels from the
country’s existing and future biomass and
biofuels industry.
23) Describe and quantify the potential and Addressed in one scenario, where
impact that a large scale decentralised biomass was given a larger part of the
conversion of Namibia’s invader bush has on electricity generation part. Biomass plants
the national electricity supply in conjunction have been identified as being one of the
with increased area to be utilised for livestock lower cost generation options. The
production. potential for grid in-feed has been
assessed in a brain storming meeting with
NamPower which has made it possible to
estimate the maximum in-feed possible
without major transmission system
investments.
However this was not balanced off

against the anticipated benefits in terms
of livestock gains from improved land
productivity.
Status: Addressed
24) Develop recommendations as to how Removed from final Scope of Work

behaviour change projects could be
implemented in Namibia, and present a cost-
Page 140 of 222

benefit analysis of various types of behaviour

change projects that are recommended.
Social and Environmental
Environmental and social perspectives are As far as Environmental Costs are

critical to provide a holistic macro-economic
concerned we have taken carbon credits
overview of the supply and demand option into account, and quantified emissions.
impacts on the Namibian electricity sector and
Any more fine-grained environmental
economy: costs would warrant a detailed
investigation on a project-by-project
These perspectives are to be developed for basis.
the following supply options:
Social costs were addressed by splitting
Kudu combined-cycle power plant job categories into skilled, unskilled and
mixed income. Furthermore, the impact
Baynes hydro power station on SMME creation was also addressed.
Thermal coal fired power station Status: Addressed
Nuclear power plant Addressed in section: 4.2, Appendix C

and Appendix D.
Wind farms
Solar thermal power station
Solar photovoltaic power plant
Biomass to electricity power plants
Consideration for environmental and social Addressed in section: 4.2, Appendix C

costs and benefits for demand side options: and Appendix D.
Energy efficient lighting
Solar Water Heater programme
It is the intention to describe the costs and

benefits for each of these options in the
Namibian context and where possible attach
actual costs to impacts so that these can also
be modelled in the cost/benefit and macro-
economic analysis.
Page 141 of 222

Appendix B. Discussion of the existing

Regulatory and Power Supply Infrastructure
B.1. REGULATORY ENVIRONMENT
The Namibian Electricity Supply Industry (ESI) is regulated by the Electricity Control Board
(ECB) subject to the powers vested in the regulator under the Electricity Act (Act 2 of 2000).
Under the Act, any person engaged in the generation, transmission, distribution, supply,
import or export of electricity must obtain a license from the ECB for such operations. There
are well-defined processes for such license applications and the general license conditions
have been developed by the ECB and are generally well understood.
It is important to note that the ECB’s independence is limited by the fact that the final
decision on licensing matters rests with the Minister of Mines and Energy. The ECB’s role is
to evaluate the applications and make recommendations to the Minister. Experience so far
has shown some cases where the Minister has overturned the ECB’s recommendations, for
example granting longer license terms to REDs than recommended by the ECB. It can
therefore be said that some tension exists between the regulator and the political echelons.
B.2. ROLES AND RESPONSIBILITIES OF ESI INSTITUTIONS

An overview of the major industry role players is given in Figure B-1.
B.2.1. Policy Level

The Ministry of Mines and Energy (MME) is the policy maker in the ESI. The MME has
developed the Energy Policy that has, since 1998, been the overall guiding document for the
ESI. The Electricity Act was developed under guidance from this policy, as have the
restructuring actions that have taken place. The Energy Policy is therefore to this day a valid
document and serves as baseline for policy in the ESI.
B.2.2. Regulatory Level

The ECB is the regulator in the ESI. The ECB has a non-executive Board and is staffed with
a Chief Executive Officer supported by various staff members who perform technical and
administrative functions. The ECB deals with the day-to-day administration of the ESI, which
principally amounts to the management of licenses, management of tariffs and interactions
with licensees. The ECB is also instrumental in developing regulations under the Act. The
ECB is currently in the process of reviewing the Act, with a number of proposed significant
changes that will increase the powers of the regulator and put in place certain key legislation
to enable the orderly implementation of REDs.
Page 142 of 222

MME
Policymaker
ECB
Regulator
Foreign
Generators
Licenses
Regulated Entities
NamPower
Future IPPs Rules
Generation
ECB levy Regulations
NamPower SORED
Trading
Central RED
NamPower
Transmission
Erongo RED
CENORED
NORED
Foreign Foreign Transmission

RED Customers
Trading Partners Customers Customers
Payment flows
Energy flows
Figure B-1: The Namibian ESI
The ECB and MME are further in the process of investigating the possibility of changing the
role of the ECB from an electricity regulator to an energy regulator. It is at this point uncertain
whether and when this may go ahead, but such a development is probably some years
away, but would have far-reaching implications for Namibia’s energy sector as a whole.
B.2.3. Licensees
The main licensees in Namibia are NamPower and the Regional Electricity Distributors
(REDs).
NamPower encompasses three main ring-fenced businesses, namely generation, trading

and transmission. There is currently also a distribution component within transmission, but it
is expected that this will be spun off to the last two REDs to be formed (Central and
Southern) within the next year or two. NamPower in its role as system operator and trader
currently has the important role of balancing supply and demand, and it is the contracting
party for imports, primarily from Eskom in the Republic of South Africa (South Africa) but also
from the Zambian Electricity Supply Company (for the Caprivi Region). As such, NamPower
seeks to optimally balance its own generation with imports and thereby obtain the best
possible cost scenario.
B.3. POWER SUPPLY INFRASTRUCTURE

This section describes the Namibian electricity generation, transmission and distribution
infrastructure.
Page 143 of 222

B.3.1. Generation
The three existing power stations in Namibia are all owned and operated by NamPower.
They are located far apart, Ruacana in the extreme north of the country, van Eck in the
centre and Paratus on the west coast. This makes the transmission system a critical
component in delivering power all over the country, especially considering that supply from
local stations is supplemented by imports from Eskom in South Africa via transmission lines
connecting Namibia to that country. Roughly half of the electricity used in Namibia has in the
past years been generated locally while most of the other half has been imported from
Eskom, which has had excess generation capacity for a long time and has offered power at
very competitive rates to neighbouring countries.
The availability of this cheap electricity from Eskom has contributed critically as an inhibiting
factor to the construction of any new power stations in Namibia (and in the region), and to
this day dominates the price expectations in Namibia, which still makes it very difficult for
new generators to establish a foothold.
The following power stations are presently utilised on the interconnected power system:
B.3.1.1. Ruacana Hydro-Electric Power Station
This power station is situated in the north of Namibia, where the Kunene River becomes the
Border between Namibia and Angola.
Rating: 240/249 MW: 3 x 83MW Francis Turbines
This station has black start up diesel generators.
One 330 kV transmission line (570 km from Ruacana to Omburu Transmission station).
Commissioning of the power station: 1978
Expected life: No end of life expected at this time.
The civil works for the addition of a fourth machine are completed, but the installation of the
turbine and generator has been postponed because of external factors. The station is mainly
operated as a run-of-the-river power plant, because all upstream storage dams have not
been completed or are damaged by the previous civil war activities in Angola. A decision has
recently been announced by NamPower to install a fourth turbine and generator to better
utilise the water available during the rainy periods. This has not been modelled since the
announcement was made after the modelling was finalised.
A small diversion weir just upstream of Ruacana is situated in Angola. This weir only allows
water regulation for approximately 1 day a week. For that reason the SAPP schedules for
Ruacana operations are on a weekly base. During the raining season (from about February
until /May each year) the station is run on full load and therefore operated as a base load
power plant, while for the remainder of the year, it is operated as a run-of-river plant.
For the purposes of this project, Ruacana has been modelled as mid-merit and peaking
plant27 according the way it is utilised for the majority of the year.
27
A mid merit power plant is typically operated 24 hours a day, but at varying output to follow the demand on the system. A
peaking plant only runs during peak times and is not utilised during off-peak times. A base load plant runs at full output all the
time. Ruacana is run as base load when there is enough water available in the Kunene river. As water flow reduces after the
Page 144 of 222

B.3.1.2. Van Eck Coal-Fired Power Station
This station is situated at Windhoek.
Rating: 120 MW: 4 x 30MW generators
This station needs an external power supply to commence with generation.
The plant has been designed as a dry cooled power station because of water supply
constraints in the Windhoek area.
Commissioning of the power station: Units 1 – 3: 1972/1973, Unit 4: 1979
High-grade coal is presently supplied by coal mines in South Africa and transported by ship
from Richards Bay to Walvis Bay, and by railway to Windhoek. There is a very high cost of
transport: 24 – 27 MJ/kg. The plant is normally operated as a standby and peaking power
station respectively, although during the more recent regional constraints it has been run
mid-merit to base load.
Van Eck is characterised by high fuel cost (largely due to transport costs) and very low
thermal efficiency (due partly to the intermittent operation, but also due to the old age of the
plant and its deteriorating boiler condition). It is therefore not a plant that is economically
viable to run except in emergency situations.
For the purposes of this project it has been assumed that van Eck should be retired as soon
as sufficient replacement capacity has been procured / installed. In most scenarios analysed
in this study this does not happen earlier than 2012, meaning that van Eck will have to do
service for a number of years to come, even at its very high cost (short run marginal cost
given as being around NAD 60c/kWh (NamPower, pers. comm., September 2007).
Van Eck power station has very little cleaning technology installed, and therefore emits high
levels of pollutants. This adds to the rationale for retiring the plant as soon as feasible.
B.3.1.3. Paratus Diesel Power Station
This station is situated in Walvis Bay. The specifications are as follows:
Rating: 24 MW: 4 x 6MW diesel generators
This station has a black start up diesel generator.
Commissioning of the power station: 1976
Light fuel oil (normal diesel) is used for start up and stopping and heavy fuel oil (bunker oil) is
used for normal operation.
This plant is mainly used as a standby and peaking power station respectively, but it is also
contractually bound as an emergency standby plant for the city of Walvis Bay (mainly for the
fishing industry).
rainy season the plant can no longer run at full output for 24 hours a day. It then follows the load during the day while water is
dammed up in the diversion weir during low demand times (typically at night). As water flow drops even further the water is
dammed in the diversion weir for use only during peak demand time in the evening and early morning.
Page 145 of 222

Paratus is in good workable condition, although it runs at very high short run marginal cost of
just under NAD 80c/kWh (NamPower, pers. comm. October 2007). For the purposes of this
project it is assumed to continue operating as peaking plant for the foreseeable future.
B.3.2. Transmission and Distribution Network

B.3.2.1. Transmission and Trading
NamPower Transmission is divided into two businesses, the wires business and the supply
business. The wires business manages and operates the transmission network, while the
supply business looks after the transmission customers. These transmission customers
include some large mines, the REDs, plus some smaller supply points that are connected
directly to the transmission grid largely for historic reasons.
NamPower Single Buyer / Transmission Trading is at this point the only trader in the country.
Imports from and exports at trading level to neighbouring countries (South Africa and
Zambia) are contracted and managed there. NamPower Generation’s output is also
effectively purchased by this unit. The output of new generators will also have to be
purchased by the Trading unit in NamPower, and this is the entity with which power purchase
agreements must be negotiated.
NamPower also holds licenses for import and export of electricity. These can be divided into
those at trading level (transactions with Eskom) and at transmission supply level (supplies to
Angola and Botswana, supply from Zambia).
There are also some cross border supplies where NamPower Transmission supplies power
at distribution or sub-transmission level to Angola, Botswana and South Africa.
There is one exception to the above system, namely power supply to the Skorpion zinc
project in southern Namibia of close to 90MW. In order to ensure firm supply and a
competitive price an arrangement was negotiated whereby Eskom supplies the power which
is wheeled by NamPower to the mine.
B.3.2.2. Distribution
The REDs are responsible for the distribution and supply of electricity to end consumers
within their respective areas. Three REDs have already been established and are operational
NORED (see Figure B-2) was established in 2002 while CENORED and Erongo RED started
operating in 2005. Within the proposed Southern RED and Central RED areas the local and
regional authorities as well as NamPower are still the licensed distributors, but the process to
establish these REDs is under way.
The REDs distribute and sell power on to end users in their respective areas of supply. They
serve approximately 150 000 end consumers, the vast majority of which are domestic users.
Page 146 of 222

Figure B-2: The Namibian Electricity Network Map
Page 147 of 222

Appendix C. Detailed discussion of

Generation Options
This section describes a range of electricity supply options that are being considered for
Namibia. These are:
1. Hydro-electric power generation at Baynes on the Kunene River
2. Coal-fired power generation
3. Power generation from natural gas sourced from the Kudu gas field off Namibia’s
southern coast
4. Nuclear power generation
5. Power generation from biomass
6. Concentrating solar power
7. Solar photovoltaic power plants
8. Wind farms
9. Integrated solar combined cycle plants
The supply options are investigated with respect to their technical details, the resources
available in Namibia, the electricity generation potential within the Namibian context, their
environmental and social implications, as well as their costs. The investigation focuses on
issues such as:
o Technology maturity, track record and commercial availability
o Modularity
o Centralised versus decentralised/distributed plants
o Embedded generation
o Lead times
o Finite energy resources
o Generation potential
o Availability: dispatchability, intermittency, predictability
o Generation type
o Environmental impacts
o Social benefits and impacts
o Cost of supply option
Some of these aspects are defined in more detail below:
Page 148 of 222

C.1. TECHNICAL ISSUES

Modularity: This aspect looks at the modularity of a supply option. A supply option is
considered to be modular if the capacity can be expanded in MW to tens of MW steps such
as for example wind power. Modular supply options have the advantage in Namibia’s
relatively small power sector, to increase generation capacity with growth, without getting into
extreme oversupply situations.
Centralised versus decentralised: Centralised plants, such as hydro-electric and coal-fired

power stations, are single, large-scale power generation facilities that are connected to the
grid through transmission lines. Generated electricity is thus distributed throughout the
country through a transmission network which is designed to transmit large-scale generation
capacity.
Decentralised plants such as biomass to electricity and solar photovoltaic plants can be
installed closer to the point of consumption, typically on the distribution grid, with resultant
reduction of transmission losses.
Some supply options are also suited for embedded generation, i.e. integration of small-scale
generators (such as solar PV) at low voltage level. This means that the consumer also
becomes a generator – a model that is becoming more and more popular in other countries.
However the success of such a model depends significantly on a consistent feed-in tariff
which allows the prospective generator a return on the investment. Feed-in tariffs for different
embedded generation options have as yet not been established for Namibia.
Construction, maintenance and monitoring requirements of the different modular generating

options are essential criteria in determining the suitability to decentralise plants. Whereas
wind is a modular generating technology, it does not lend itself to decentralisation due to loss
of economies of scale as well as the need for regular maintenance, compared to for example
solar PV generation.
Lead times: In this study, lead times are divided into two parts – the pre-construction lead
time (project development, design, environmental compact assessment (EIA), permissions,
legal framework) and the construction lead time (actual implementation). It is essential that
lead times are taken into account, particularly while there is a capacity shortage in the SAPP.
Availability: Each supply option is defined through a generation profile, which defines its
availability. Typically a nuclear power plant has availability in excess of 90% while a wind
park may have an availability (or capacity factor) of 30%.
Availability is further defined by the degree of dispatchability, i.e. if the supply option is 100%
dispatchable, then the supply option is based on a fossil/nuclear fuel generator or a hydro
generator with a large storage dam and there is no uncertainty in the supply. This however
assumes undisrupted supply of fuel for the fossil/nuclear generation plants.
An intermittent generator has a variable generation profile. Some intermittent generators

have a high level of predictability (e.g. solar) while others have a lower predictability level
(e.g. wind).
Generation type: This is classified into three categories: base-load, mid-merit and peak-
load. Base-load generators are designed for continuous operation (high availability factor).
Mid-merit plants are designed for generation during periods of higher demand (i.e. during the
day and including peaks). Peaking power plants are suitable for generation during the peak
load period mainly due to high cost of fuels or limited pumped storage.
Page 149 of 222

Pumped storage: This is a facility which is based on a hydro power station with a
conventional dam at an elevated point but with an additional dam at a lower level (post
turbine). A pumped storage scheme is able to reverse the water flow and pump water from
the lower dam into the higher dam during off-peak times. Traditionally, a pumped storage is
an attractive means to cope with peak demand in the daily demand load profile. A pumped
storage scheme is also attractive in combination with wind generation in order to even out
the intermittency of wind. The basic requirements for a pumped storage scheme are:
• Sufficient head between the higher and lower dam levels
• Geographical realities which allow the construction of dams of sufficient size to

provide sufficient hours of storage at full capacity output
• Sufficient availability of water
• Normally a pumped storage scheme would imply a hydro power station designed and
built from the outset with the ability to reverse the turbines and use them to pump the
water back up. Separate installed pumps are technically possible but detract from the
economic viability.
NamPower has previously investigated the option of pumped storage and has come to the
conclusion that there are no suitable sites in Namibia that can be developed at competitive
costs under the current electricity cost regime28. Consequently, a pumped storage in
combination with wind generation has not been developed into a supply option in this study.
It is expected that Namibia can in the near future consider other forms of storage that do not
rely on geographical requirements and water. An example of such storage facilities is
compressed air storage, which has the ability to store energy for days to weeks.
C.2. ENVIRONMENTAL ASPECTS

All aspects of energy, i.e. the way it is produced, distributed, and consumed, can affect local,
regional, and global environments. This can occur through change of land use, land
degradation, air pollution, the acidification of water and soils, and through global climate
change via greenhouse gas emissions (Energy 2007).
Due to time and resource constraints of this study as well as the broad scope, this study
does not assess each and every social and environmental impact that might arise for every
supply option proposed. Therefore, no attempt was made to discuss in great detail the
individual impacts for each and every supply option put forward. Instead, various
assumptions were made for each of the supply options and are highlighted in the relevant
sections following this introduction.
The key environmental29 areas considered for this study were:
28
Information provided by Udo Kleyenstüber, NamPower; established during a meeting on 22 October 2007.
29
The “Environment'' is defined as: the biosphere in which people and other organisms live. It consists of:
• renewable and non-renewable natural resources such as air, water (fresh and marine), land and all forms of life,
• natural ecosystems and habitats, and
• ecosystems, habitats and spatial surroundings modified or constructed by people, including urbanised areas,
agricultural and rural landscapes, places of cultural significance and the qualities that contribute to their value.
Integrated pollution and waste management is a holistic and integrated system and process of management, aimed at pollution
prevention and minimisation at source, managing the impact of pollution and waste on the receiving environment and
remediating damaged environments.
Page 150 of 222

• Air pollution, e.g. emissions of sulphur and nitrogen. However, it should be noted that
no costs were modelled for the abatement of NOx and SOx emissions in this study;
• Greenhouse gases (e.g. Carbon dioxide, CO2), their effects on Climate Change, and
the costs of carbon emissions trading;30 In general, the 'carbon footprint' of fossil
fuels are higher than the other supply options considered and thus have a higher
overall contribution towards climate change.
• Water as a valuable natural resource;
• Sustainable development, i.e. questioning how sustainable an energy supply option
is for future generations giving consideration to the support and sourcing of local
resources, goods and services.
A more detailed account of the environmental impacts to be assessed will have to be

finalized through a formal environmental impact assessment (better known as an 'EIA').
Policies, programmes and projects will need to follow a formal environmental assessment
(EA) procedure as set out by the Namibian Environmental Assessment Policy (Tarr 1994).31
In relation to this study, the Environmental Assessment Policy states that following activities
would apply:
• Rezoning applications,
• Land acquisition for national parks, nature reserves, marine reserves, protected
natural environments or wilderness areas,
Pollution'' for the sake of this study will be defined as: the introduction into the environment of any substance property including
radiation, heat, noise and light that has or results in direct harmful effects to humanity or the environment or that makes the
environment less fit. (South African Department of Environmental Affairs and Tourism, 17 March 2000).
30
Climate change is a shift in long-term average weather patterns, which can include changes in temperature and in
precipitation amounts. The international scientific community agrees that there has been a significant change in global climate
in recent years, due largely to the burning of fossil fuels for transportation and industrial processes. These activities emit
greenhouse gases that trap heat in the atmosphere (Climate Change 20 March 2007, Environmental Canada). Two greenhouse
gases that are released through the burning of fossil fuels are carbon dioxide (CO2) and Nitrous Oxide (N20). Methane (CH4) is
released during fossil fuel extraction and decaying garbage.
The Namibian government could apply various mechanisms to promote the reduction of carbon emissions produced by
NamPower, either through Carbon Taxes or a Cap-and-Trade scheme. The idea of a Carbon Tax is to place the government in a
position of taxing the power utility (in this case NamPower) according to their carbon footprint which can be more readily
monitored and measured. Carbon taxing is considered more efficient and more equitable of the two mechanisms. Additionally,
the tax revenues received could help to fund research and development in alterative fuels, new technologies.
According to the Brookings Institution and the World Resources Institute, a $15 tax on each ton of carbon dioxide would cause
the price of coal to nearly double. Further, they claim that this would cause coal consumption to drop by one-third (Silverstein,
Ken 12 November 2007).
Conversely, a more market-based approach is Cap-and-Trade which sets limits to greenhouse gases from various sectors like
the power utility sector to a determined level within certain time frames, e.g. cap the emissions at 2002 levels by 2020. Those
companies that are unable to meet their levels could then buy 'credits' from those who exceed their objectives. While more
flexible in its approach, some argue that the 'credits' need to be auctioned from the outset while others support credits being
initially given away to allow the market time to work. Additionally, it is argued that the cap-and-trade favours those with money
to buy credits.
31
To pre-empt this Policy and its procedure would be irresponsible and unacceptable. The authors of this study cannot
professionally proclaim with absolute certainty what the related impacts of each supply option are and their associated costs are
projected to be without going through a project and site specific process which would include information and data collection,
data analysis and review, proposing and assessing alternative site locations for the proposed project, and a public awareness
and outreach programme (to name but a few).
Page 151 of 222

• Any government policy, programme or project on the use of natural resources,

• Human population growth / management programmes,
• Nuclear installations,
• Transportation of hazardous substances and radioactive waste,
• Mining, mineral extraction and mineral beneficiation, power generation with an output
greater than 1 MW,
• Electrical substations and transmission lines having equipment with an operating
voltage in excess of 30,000 volts phase-to-phase,
• Storage facilities for chemical products,
• Industrial installation for bulk storage of fuels,
• Bulk distribution facilities,
• Major roads,
• Railways,
• Major pipelines,
• Major canals, aqueducts, river diversions and water transfers,
• Major dams, reservoirs, levees and weirs,
• Reclamation of land from the sea,
• Human resettlement,
• Effluent plants,
• Waste disposal sites,
• Alternative energy programmes, and
• The use of pesticides, herbicides and defoliants
(Tarr 1994).
Potential environmental and social impacts related to the activities of electricity generation
project are listed in Table A1 for electricity generation projects. While the full list of impacts
may not apply to every supply option considered in this study, the list provides an idea of the
breadth and scope of environmental impacts that might need to be considered by the
Namibian electricity sector in its final decision.32
Other legislation pertaining to the environment will need to be taken into consideration. No
attempt was made to review and analyze all the environmental and social legislation that
might apply in an energy generation project.
In economics, an 'externality' is a cost or benefit resulting from an economic transaction that

is borne or received by parties not directly involved in the transaction. Externalities can be
32
It should be noted that there will be site specific issues like demographics data and baseline data for air quality, water quality,
surrounding land-use patterns and sensitive and endangered flora and fauna species (to name but a few) that will need to be
collected when a supply option is finally proposed. Again, to determine the more detailed review for a given generation option is
part of the required environmental assessment process and cannot be predetermined, pre-empted, qualified or quantified ahead
of time. Further, the concept of 'Life Cycle Assessment' may also be applied to determine the impacts and costs associated from
inception to decommissioning of a power station (also known as 'Cradle to Grave.')
Page 152 of 222

either positive, when an external benefit is generated, or negative, when an external cost is
imposed upon others. Central to environmental economics is the concept of an 'externality'
which refers to some effects of an activity that are not accounted for in its price.
The types of environmental and social impacts that might be addressed in future projects
include:
• Significant positive and negative impacts;
• Direct and indirect impacts;
• Secondary and/or cumulative impacts;
• Immediate and long-term impacts;
• Impacts which are unavoidable or irreversible impacts; and
• Construction, operation, and decommissioning
In general terms, the potential environmental and social impacts of electricity supply options
might include further investigations and quantification of the following impact areas:
Air Climate Change

• Dust (particulate matter) • Carbon emissions (tonnes)
• Sulphur and nitrogen emissions, e.g. SOx • Carbon intensity (tonnes per MWh)
and NOx
• Change of weather patterns
• Greenhouse gases, CO2 and CH4
• Extreme weather events
• Mercury
• Loss of ecosystems
• Waste heat
• Loss of biodiversity
• Indoor air pollution
• High insurance premiums
• Acoustic disturbances (noise pollution)
• Water scarcity
• New and unforeseen energy and carbon
laws and regulations
Water pollution and effluent management Land-use management

• Run-off • Soil and land disturbances and associated
water pollution (surface and/or
• Coastal zone management (if applicable) groundwater)
• Oil and grease • Conflicts with existing land uses and local
• Ash populations
• Effluent from the work areas and wash- • Conflict with other potential developers
bays soaking directly into the ground • Impacts on vegetation and agriculture
• Effluent from the on-site ablution blocks • Off-road driving during construction
• Run-off from the work and storage areas • Designated and proclaimed protected
• Oil spills and other spillage’s (leaches from areas, and/or flora and fauna
solid wastes dumping with direct toxicity or • Coastal zone management (if applicable)
effect on soil pH can prevent vegetation
from growing) • Temporary, permanent and access roads
Page 153 of 222

(during construction and operation)

• ‘No-go’ areas or zones
Waste management (during construction and Storage of materials

operation);
• Hazardous and benign
• Ash
• Recyclables (paper, metals)
• Storage and disposal of (hazardous) waste
• Paints, thinners,
• PCB’s (transformers)
• Oil and grease
• Asbestos
• Batteries
• Batteries
• Redundant equipment, scrap and waste
• Uncontrolled storage and dumping of
product and waste
Wildlife interaction
• Marine fisheries (if applicable);
• Birds
o Seasonal impacts
o Species impacts
o Migration routes
o Breeding places – effect on
breeding
o Disturbance of forage locations
• Mammals
Social Impacts Social Impacts
Positive socio-economic impacts (due to access) Negative socio-economic impacts

• Employment possibilities • Accidental risks, emergency situations and
response;
• Poverty alleviation
• Construction camps and power line
• Light construction activities;
• Cooking • Damage to cultural and natural assets
• Refrigeration • Human health impacts, e.g.
• Hot water electromagnetic fields from power lines
• Health and safety • Resettlement (if applicable) and the

associated costs of compensation.
• HIV /AIDS
• Effects on Aviation
• Hospitals and clinics, e.g. refrigeration of
medication • Effects on tourism industry
• Schools • Visual impacts and Aesthetics

• Shadow disturbances
Table C-1: Potential Environmental and Social Impacts of Energy supply Options
Page 154 of 222

C.3. COST ISSUES

The cost of the considered supply options takes into account the capital costs, the operating
costs and the fuel costs.
Capital cost: This consists of the overnight turn-key implementation costs which include
materials, labour, engineering and project management, the IDC, based on the construction
lead time calculated at 6% real weighted average cost of capital (WACC,) and the
decommissioning costs as percentage of the capital cost. The interest during construction is
divided into equal amounts over the construction period, while no IDC is charged during the
project development phase (pre-construction) as it was considered negligible compared to
the actual IDC during construction.
Operating and fuel cost: These are presented as a cost per kWh.
VAT: All pricing is exclusive of VAT.
Exchange rate: All costs stated in foreign currency are adjusted for inflation of that particular
currency up to 2007 and then converted to Namibian dollars at real exchange rates of 7.00
for USD and 10.00 for Euro.
C.4. HYDRO-ELECTRIC POWER GENERATION

The proposed Baynes hydro dam is located on the Kunene River, which forms the border
between Namibia and Angola, downstream from the Ruacana hydro power station. The
catchment area of the river is pre-dominantly on the Angolan side.
The Baynes site has been well studied and its technicalities and environmental impacts are
well understood (NAMANG 1998). Baynes has been extensively compared to the Epupa site
further upstream and is seen as having a lower environmental and social impact than Epupa.
C.4.1. Technical Issues

The Baynes development has the following technical details specifications (NAMANG 1998:
2-1):
Dam capacity: 2.6 billion cubic metres
Active storage: 1.8 billion cubic metres
Dam wall height: 200m
Dam wall length: 1,200m
Generation output: 360MW (with options for 540MW and 600MW)
The service life of the hydro-electric power station is expected to be 40 years.
Page 155 of 222

The implementation lead time for the development is estimated at 2 years pre-development
time, three years construction time and an additional 5 years33 for the dam to fill up.
C.4.2. Generation Potential

The potential generation output for the Baynes dam is estimated at approximately 1,147GWh
per annum. However, should the Gove dam in Angola be taken into operation again (with the
necessary repairs and coordination from Angola) then the annual generation potential of the
Baynes Power Station can reach about up to 1,700 GWh.
The Baynes hydro power plant is modelled as a base load station with a 360MW output in
this study. The capacity factor is assumed at approximately 90%.
Considerations have been expressed by NamPower to operate Baynes more as a mid-

merit/peaking plant as opposed to a base load plant. In that case the installed capacity would
be increased from 360MW to 540/600MW.
C.4.3. Environmental and social aspects

Possible environmental and social impacts of hydro-power generation are listed below. While
at first glance this list appears to be just environmental impacts it is important to note that any
changes or degradation to the natural environment will impact individuals and communities at
large. Water is a scarce resource in Namibia. This valuable resource is vital to the welfare
and livelihood of people. Informal settlements may be affected by a hydro power station
altering smaller communities' food production and availability of water for drinking, washing
and cooking. Any decisions made to use this resource for the sole purpose of generating
electricity will have to take into the account how this water is used and what impact it will
present to the natural and human-made environment up-stream and down-stream of the
generating unit(s). Altering the flow of water will change the surrounding ecosystems and the
very environments that people and communities rely upon.
The Summary Environmental Assessment Report on the Baynes Project (NAMANG 1998)
highlighted a number of issues, namely:
• The biophysical environment (i.e. regional and local setting, impacts related to the
different project phases, impacts on the aquatic habitats and species, impact
downstream of the reservoir, mitigation and management, monitoring), and
• The human environment (i.e. analysis of the current situation, analysis of project
impacts, analysis of mitigation, community participation and public awareness in the
EA formal process, and costs for social mitigation).
Impact-Mitigation-Valuation matrices were developed for the biophysical and human

environments. Six impact zones were defined to structure the matrices in the Report
(NAMANG 1998). These include:
• Upstream watershed and river reaches;
• Reservoir area and water body;
33
Information provided by Udo Kleyenstüber, NamPower during a meeting on 22 October 2007.
Page 156 of 222

• General project area, access and transmission corridors;
• Dam, spillway and power station;
• River reach dam to tailrace;
• River reach downstream to tailrace including delta complex.
The Report calls for the development of an Environmental Management Plan (EMP). The
EMP consists of a set of mitigation, monitoring and institutional measures to be taken during
the implementation and operation of the project (NAMANG 1998). It is the vehicle for
minimising and offsetting the adverse impacts to the biophysical and human environments. It
addresses the activities of both the Social Mitigation Programme, the Client’s organisational
set-up for project implementation, the Contractors and the project Consultant(s). Pending the
Angolan and Namibian governments’ decision regarding project implementation and
establishment of a Social Mitigation Programme, the EMP must remain of a preliminary
nature drawing from and systematising the mitigation and monitoring requirements presented
in the EA report. For this reason what is presented in the EA report is termed an Initial
Environmental Management Plan.
The Initial Environmental Management Plan encompasses the following aspects of

environmental management:
• Institutional arrangements and responsibilities
• Integrated catchment management
• Socio-economic safeguards
• Water quality safeguards
• Terrestrial ecology safeguards
• Aquatic ecology safeguards
• Pollution control
The summary statements made in the NAMANG Report (1998) of the environmental impacts
of the Baynes project are:
Terrestrial Environment
Although a considerable area is flooded the impacts are regarded as of limited
significance and largely capable of mitigation given the characteristics of the
inundation area and surrounds.
Aquatic Environment
The impacts are significant on some critically endangered fish species and on the
downstream with regard to river regulation, but they are not totally unacceptable
providing compensation measures are implemented.
Social Environment
The impacts are significant; the Project will have serious consequences on the
sociocultural environment of local people unless comprehensive and effective social
Page 157 of 222

compensation programmes are developed through consultation with the directly

affected communities.
Environmental Mitigation and Monitoring Costs

Financial costs associated with mitigating above summarised impacts are moderate
(2.0% of construction costs) and economically insignificant in the overall cost picture.
External Costs
Quantifiable external costs associated with the Baynes Project are small (0.02% of
construction costs) in relation to financial economic costs.
In addition to the quantifiable external costs, there will be non-quantifiable environmental

losses (wilderness, biodiversity, imposed social change etc.) associated with implementing
the Baynes Project. These losses could be avoided by opting for supply scenarios with small
or negligible environmental consequences for the country, such as the import of power from
Republic of South Africa (RSA) or generation by a domestic combined cycle plant run on
Kudu gas. In terms of present value of measurable costs over the life of the schemes, import
of power is USD 8 million more costly and domestic thermal generation USD 60-65 million
more costly, both alternatives compared to the Baynes project with Gove in operation.
The question that the decision makers need to address when it comes to unquantifiable
environmental impacts, is the following: Is the preservation of the environment worth the
additional cost of USD 8 million, in the case of increased reliance on imports, or USD 60-65
million, in the case of domestic thermal generation? If the answer is in the affirmative, the
development of the project should not go ahead, and vice versa (NAMANG 1998).
The Low Impact Hydropower Institute as cited by Power Scorecard (2007) has created a Low
Impact Hydropower Certification Programme. It has identified eight environmental criteria to
minimize the impacts of hydropower dams. These include:
• River flows
• Water quality
• Fish passage and protection
• Watershed protection
• Threatened and endangered species protection
• Cultural resource protection
• Recreation
• Facilities recommended for removal
Further, the Power Scorecard (2007) highlights various impacts and by implication the
associated costs of hydropower. These include:
• Relocation of communities
• Relocation of cultural and heritage sites, e.g. burial sites
• Wildlife mitigation programmes
• The construction of bypass and diversion systems for certain fish species
• Converting a dam to a run-of-river operation to ensure the natural flow of the river
remains
Page 158 of 222

• Dredging, flushing and piping techniques for moving non-contaminated sediments

downstream
• Mitigating technologies to improve dissolved oxygen levels
• Minimum flow turbines
• Re-regulating weirs
• Pulsed operation at peak efficiency and minimum discharges to improve habitat
during seasonal or annual drought / flood conditions
The positive and negative socio-economic impacts are listed in the table below:
Positive socio-economic impacts (due to access)
Employment possibilities Limited possibilities for the local population since employment will
firstly be small in numbers and secondly mostly skilled.
Poverty alleviation Little positive effect can be expected from the power project.
Negative effects are possible due to disturbance on the traditional
way of life and access to the river.
Light, Cooking, Refrigeration, Hot Water Little likely positive effect due to the dispersed households. Rural
electrification in the area is not automatically guaranteed in sync with
the dam construction.
Health and safety The power station staff village is likely to have at least basic medical
facilities which should also be accessible for local population.
HIV /AIDS Likely negative effect due to sex work opportunities arising during
construction and thereafter due to non-permanent staff.
Hospitals and clinics, e.g. refrigeration of Little likely positive effect.

medication
Schools The power station staff village may get a school.
Negative socio-economic impacts
Accidental risks, emergency situations No significant risk

and response;
Construction camps and power line Temporary major impact during construction
construction activities;
Damage to cultural and natural assets Permanent major impact on local population’s cultural assets
Human health impacts, e.g. Negligible impact due to very low population density
electromagnetic fields from power lines
Resettlement (if applicable) and the Significant impact

Effects on Aviation None
Effects on tourism industry Permanent impact. Scale of impact depends on which dam option is
executed (e.g. whether Epupa falls disappear or not)
Page 159 of 222

Visual impacts and Aesthetics The dam will alter the landscape, but the visual impact of man-made
structures will be limited to the dam wall, the power line(s) and the
settlement that will be constructed for the power station staff.
Shadow disturbances None.
Table C-2: Social Impact Checklist – Baynes Hydro
C.4.4. Financial Overview

The cost of Baynes in 1998 was estimated at USD 555 million (NAMANG 1998: 2-2). The
Team estimated that the cost of the project in today’s USD is estimated at USD 3.9 million
per MW. This is motivated by the high world wide demand for generation equipment and the
specific risks associated with building the Baynes dam wall. The operating cost is estimated
at NAD 5 c/kWh and is based on the Ruacana figures.
The full capital costs including interest during construction and decommissioning used for
modelling is NAD 43.5 million per MW installed which translates to NAD 37.5c/kWh over an
operating life of 40 years and a yield of 90% of installed capacity. Operating cost is modelled
at NAD 5c/kWh.
C.4.5. Risks
The development of a large scale hydro dam brings with it a number of significant risks. In
the case of the Baynes development there are also specific risks which have previously been
identified. The risks include:
• The construction of the dam wall of the proposed Baynes hydro scheme is risky as
this type of construction has not been done before (Maritz 2000). It is likely that
construction methods have developed to such an extent that this risk can be
mitigated to a better degree today.
• A further risk is potential upstream developments on the Kunene River within Angola.
Major irrigation projects can have an impact on the flow of the river and therefore its
power generation capacity.
• Border River to Angola and shared water rights and the potential political risks
associated with that. This is further exacerbated through the risk of political instability
which has been the case for Angola as well as Namibia.
• Low river flow due to drought or related to global warming would have severe
implications. A second hydro power scheme on the same river raises the risks for the
Namibian power sector significantly as the diversification of generation is very low
(“all eggs in one basket”).
• Construction delays due to high demand for power stations and related equipment
globally.
• Resistance to the development of the dam.
The risks with their likelihood and severity are shown in the risk chart in Figure C-3.
Page 160 of 222

Challenging dam wall
Severe
construction
Water supply future
climate change
Angolan up-stream
water use developments;
irrigation
I
Moderate
M
P Border river:
Polical instability
A Environmental issues:
C resistance to project
development
T
Construction delays due
to high global demand
Minor
Low Medium High

LIKELIHOOD
Figure C-3: Risk chart for Baynes hydro power
Page 161 of 222

C.5. COAL-FIRED POWER GENERATION

Coal fired power stations have been around since the very first days of electricity generation.
The figure below shows that in the last thirty years there has been little change in the status
of coal as primary energy source worldwide.
Figure C-4: World Primary Energy Sources34
Power generation from coal is an old technology, and is very well established and proven. It
is one of the traditional default power generation choices, both in terms of proven technology
as well as overall costs. Today, coal is becoming more controversial due to its high level of
greenhouse gas and other emissions (if unmitigated) when compared to other primary
energy options. Some “cleaning” of the emissions is possible, but this is not entirely devoid of
controversy since emissions are not eliminated but only re-packaged into other forms of
waste, which still represent disposal issues. Sequestration of CO2 emissions is being widely
discussed and pursued, but is yet to become main stream technology
(http://www.co2storage.org.uk/, accessed March 2008).

A coal-fired power station comprises a coal fired boiler to convert the energy contained in the
fuel (coal) into heat and pressure (steam) which is used to drive a steam turbine which is
coupled to an electricity generator.
Various improvements to the basic technology have been developed over the years, some of
which have entered the main stream and are by now proven technology, while others are still
considered experimental, specifically the concept of clean coal. The cost of clean coal
technology is not yet clear, but carbon capture and storage may increase the capital cost of a
coal power station by 35% (http://www.parliament.uk/documents/upload/postpn253.pdf,
accessed March 2008)
Namibia has an existing coal power plant, the van Eck power station in Windhoek. This plant
is a dry cooled plant with a nominal capacity of 120MW. It has been used as standby plant
for most of its life due to the lower cost of power imports from South Africa and the high cost
of coal transport.
34
Source: RWE World Energy Report 2005
Page 162 of 222

Coal power plants have a typical service life of 30 to 50 years. Van Eck power station is 30
years old and still in operable condition, although its life is being negatively affected by the
intermittent emergency only operation which is putting abnormal wear on its boilers.
The lowest cost fuel supply option is to build the coal power plant at a coal mine, thus
eliminating coal transport costs as well as logistical issues. This is not an option for Namibia
currently since there are no proven coal reserves of a magnitude and quality that would
support a reasonable size power plant. However there are some unexploited coal reserves in
Southern Namibia which may in future be extracted. Coal plant in Namibia would therefore
currently rely on coal imports.
Existing technology coal power plants of smaller capacity such as would be considered for
Namibia (e.g. up to 600MW), currently have a lead time of at least four years. However, high
international demand for new power plants as well as construction resource constraints may
increase this lead time, but smaller power plants of say 150MW to 300MW may not be
subject to the same international demand as large power plants, and may be more readily
available. Coal plants considered for Namibia would be standard technology and hence
essentially “off-the-shelf” items. Unless constructed at the coast and cooled using sea water
such plants would have to be dry cooled due to the general scarcity of water in Namibia.
C.5.2. Resource
There is currently no proven coal resource in Namibia that could fuel a sizeable coal power
plant, although there are some reserves which may in future be exploited. For the purpose of
this report it is thus assumed that coal would have to be imported. This would typically be
done through the port of Walvis Bay, with rail transport to the site of the plant. The most likely
location of such a future plant would be at Walvis Bay because this would eliminate both the
cost and logistical risk of land transport of the coal. It would also enable the possibility of
using sea water for plant cooling (although this poses environmental and corrosion
challenges).
Coal supply would be secured through a long term coal supply contract with one or more
coal mines. Regional coal mines should prove to be cheapest sources due to shorter
transport routes and lower transport cost, but this may be offset by anticipated increasing
transport costs (escalation on price of diesel), supply risks and increased handling costs due
to political and economic instability on the African continent.
World coal resources are not considered to be significantly constrained (RWE 2005: 26) over
the normal lifetime of a coal power plant (although there are dissenting views on this topic
(Energy Watch Group 2007)).
If a long-term supply agreement can be established at acceptable price from a source in a

stable environment then the fuel supply risk should be limited largely to the exchange rate
risk since such an import contract may be indexed in foreign currency. It must be noted that
there are other opinions (Energy Watch Group 2007) which state that coal reserves are
being vastly overestimated. This may lead to higher prices or more difficulty in sourcing coal.
In the modelling it was assumed that coal prices would remain stable at 2007 prices
experienced by NamPower (NamPower, pers. comm., October 2007) on short term supply
contracts. A sensitivity analysis was run on coal prices doubling from 2007 level over five
years.
Page 163 of 222

C.5.3. Generation potential

Generation potential for a new coal power station would be determined mostly by its price
and policy interpretation since it is not constrained by local resource availability. For large
plant above 700MW transport constraints may become a factor (such as harbour and rail
capacity), but very large plants are unlikely to be built in Namibia because they would not be
able to compete with the low fuel costs of large coal plants in South Africa which are situated
at coal mines.
Coal plants are normally base-load plants, with high availability. Coal plants could supply
Namibia’s entire base and mid-merit load from a technical perspective. Plant sizes up to
700MW have been contemplated in the industry and for the purposes of this study plant size
steps of 175MW have been used between 175MW and 700MW.
The transmission grid to the central costal area has recently been strengthened. This would
allow for a plant of 175MW located at the coast to feed its output into the national grid without
much new investment in network capacity. Existing load at the central is around 80MW and
new mining and desalination activities are expected to add at least 200MW in the next 2-3
years. This would make a 350MW coal plant at Walvis Bay feasible without new transmission
capacity needing to be added (although strategically NamPower may wish to strengthen the
transmission capacity to be able to fully export the coastal generation capacity)
The modelling for this study has been done on the basis that a coal plant in Namibia should
be as small as possible to provide Namibia with sufficient base load capacity for the next 5 to
10 years while import constraints persist. Thereafter it is expected that due to high transport
cost a local coal plant will not be able to compete with regional coal plants built at mine
mouth, and it would only be run in emergencies. This reflects the past operating experience
with van Eck coal plant which was also not able to compete with imports.
C.5.4. Environmental and Social Aspects

Coal-fired power stations have numerous environmental and social impacts to manage
before, during and after the life-time of the station. Over the life-time of a coal-fired power
station, which can span over 40+ years, environmental and social issues will arise that will
need to be accounted for.
Further to this, the construction, operation and decommissioning phases of the power station
each have environmental and social implications that will require adequate budgets and
sound management decisions and planning. For instance, during the construction phase,
contractors need to be made aware of the sensitivities of the surrounding environment and
take care to avoid unnecessary harm and damage. During the operation phase, the natural
environment within the direct boundaries of the site as well as the equipment used to help
prevent air and water pollution will have to be planned and budgeted for, maintained and
managed.
As already highlighted in section 4.2.6.1.1, the proposed Walvis Bay coal-fired power station
would receive coal supplies from Richards Bay in South Africa by ship vessels travelling
around the Cape peninsula. It is estimated that the annual emissions from shipping range
between 600 and 800 million tonnes of carbon dioxide, or up to 5% of the global total. This is
nearly double Britain's total emissions and more than all African countries combined (Laumer
4 March 2007). Carbon dioxide from ships do not come under the Kyoto Protocol. What this
value would equate to for the power station in Walvis Bay would depend upon the size of
vessel and large the cargo would be and how many trips were taken between Richards Bay
and Walvis Bay.
Page 164 of 222

In addition to air emissions, the vessels could cause pollution to the ocean along its journey
by various means including oil, grease and petrol leaks and spills, general waste disposal,
and even possibly sewage. These activities will impact the marine environment. Humans are
reshaping the seas through overfishing, air and water pollution, commercial shipping and
other activities (Eilperin 15 February 2008). Although not the scope of this document it is
assumed that any vessels destined for Namibia will abide by the laws of the seas with
owners and operators of such vessels taking every precaution to avoid undue harm to the
ocean and its environs.
Upon arrival at Walvis Bay the vessels filled with coal may need to dock at sea if the harbour
and off-loading bays are occupied. Again, there is the potential for oil, grease and petrol
leaks and spills, general waste disposal and even sewage into the surrounding waters unless
closely monitored and regulated. The sensitivity to air and water pollution of the Walvis Bay
lagoon poses a threat to marine flora and fauna. As other industrial activities will be taking
place in the Bay activities related to the off-loading of the coal will have to be closely
monitored and regulated. Emergency response plans and procedures will need to be drawn
up to establish accountabilities and responsibilities in case of accidents. Such a system may
be part of the Walvis Bay terminal or port authority's ISO 14001 Environmental Management
System.
There are numerous environmental and social issues surrounding the operation and
maintenance of coal-fired power stations. The proposed location of such a station in Walvis
Bay brings a number of unique challenges that will have to be thoroughly addressed in an
environmental impact assessment and public health risk assessment to meet the pre-
condition of the Namibian Port Authority (ECB License Application June 2007). Some of the
issues that will likely to be addressed are:
• Emissions from the transfer of coal from the transfer station to the coal stock-yards of
the power station;
• Land-use management including coal storage, any ash dumps or disposal sites, and
mixed wastes;
• Waste management;
• Noise pollution;
• Health and safety of employees and the surrounding community;
• Monitoring and reporting to authorities; and
• Training and awareness of employees as well as local outreach of the community
Table C-3 (World Coal Institute August 2007) outlines in greater detail some of the key
environmental areas that may need to be addressed in Walvis Bay if a coal-fired power
station is pursued. Important to note is that when coal is burned as fuel, it gives off carbon
dioxide, the main greenhouse gas that is linked with global warming.35 Burning coal also
produces emissions, such as sulphur, nitrogen oxide (NOx), and mercury, which can pollute
the air and water. Sulphur mixes with oxygen to form sulphur dioxide (SO2), a chemical that
can affect trees and water when it combines with moisture to produce acid rain. Emissions of
nitrogen oxide help create smog, and also contribute to acid rain. Mercury that is released
into the air eventually settles in water. The mercury in the water can build up in fish and
shellfish, and can be harmful to animals and people who eat them (USDOE EIA November
2007).
35
Emissions trading is a market-based policy instrument that allows those who produce greenhouse gases to reduce their
emissions at minimum cost. This type of policy instrument is a move away from governments’ traditional regulatory approach.
Page 165 of 222

Environmental Challenges Technological response Status
Particulate Emissions
Such as ash from coal Electrostatic precipitators and Technology developed

combustion. Particulates can fabric filters control particulate and widely applied in
affect people's respiratory emissions from coal-fired power both developed and
systems, impact local visibility and stations. Both have removal developing countries
cause dust problems. efficiencies of over 99.5%
Trace Elements
Emissions from coal-fired power Particulate control devices, Technologies

plant include mercury, selenium fluidised bed combustion, developed,
and arsenic. They can be harmful activated carbon injection and commercialised and
to human health & the desulphurisation equipment can widely applied in
environment all significantly reduce trace developed countries.
element emissions.
NOx
Oxides of Nitrogen (NOx) are NOx emissions can be cut by Widely applied in
formed during the combustion the use of specialised burners, developed countries,
process. Can contribute to smog, advanced combustion methods, and although increasing
ground level ozone, acid rain and catalysts and 'selective non- in developing countries,
greenhouse gas emissions. catalytic reduction'. Over 90% could be more widely
of NOx emissions can be used.
removed using existing
techniques.
SOx
Mainly sulphur dioxide, produced Technologies are available to Widely applied in

from the combustion of elemental minimise SOx emissions by developed countries,
sulphur present in many coals. removing the gas from the and although increasing
Emissions can lead to acid rain waste stream or by using in developing countries,
and acidic aerosols. advanced power generation could be more widely
methods. Emissions can be used.
reduced by over 90%.
Waste from Coal Combustion

Consists primarily of
incombustible mineral matter Waste can be minimised before Technologies are
(by coal cleaning) and during proven and continually
(using high efficiency systems) improving. Awareness
coal combustion. Residual of the recycling
waste can be reprocessed into opportunities (e.g. the
construction materials. use of ash in cement
making) is steadily
increasing.
Page 166 of 222

Environmental Challenges Technological response Status
Carbon Dioxide
(reduction)
In the short to medium term, Coal-fired power station
Carbon dioxide is formed when substantial reductions in efficiencies greatly
fuels containing carbon are burnt, greenhouse gas emissions can improved during the last
and is a significant greenhouse be made by increasing the part of the 20th century
gas. Progressively reducing CO2 efficiency of coal-fired and with the further
is an essential element of a global generation. development of current
response to the risks of global methods will improve
warming and climate change. yet further.
Carbon Dioxide
(near zero emissions)
Near zero emissions CO2 separation, capture
Achieving near zero emissions of technologies enable the and storage is under
CO2 from fossil fuel based power separation and capture of CO2 development around the
offers the prospect of balancing from coal-fired power world with projects
growing energy demand with generation for permanent and announced in the USA,
stabilising greenhouse gases in safe storage underground. Australia and Europe.
the atmosphere.
Table C-3: Environmental challenges at coal fired power stations
Employment possibilities Limited possibilities for the local population.
Poverty alleviation Little positive effect can be expected from the power project.
Light, Cooking, Refrigeration, Hot Water No likely effect due to location in urban area.
Health and safety No likely positive effect due to location in urban area. Likely negative
effect due to coal storage and emissions.
Hospitals and clinics, e.g. refrigeration of No likely effect due to location in urban area.
medication
Schools No likely effect due to location in urban area.
Accidental risks, emergency situations Significant accident risks due to location in populated area will need
and response; to be mitigated.
Page 167 of 222

Construction camps and power line Insignificant likely effect due to location in urban area.
Damage to cultural and natural assets Insignificant likely effect due to location in urban area.
Human health impacts, e.g. Possible significant negative impacts due to coal handling and
electromagnetic fields from power lines emissions.
Resettlement (if applicable) and the None.

Effects on tourism industry Insignificant likely effect due to location in industrialised area.
However if the plant is located close to the lagoon it may impact on
tourism.
Visual impacts and Aesthetics Insignificant likely effect due to location in industrialised area.
However if the plant is located close to the lagoon it may impact on
tourism.
Shadow disturbances Insignificant likely effect due to location in urban area.
Table C-4: Social Impact Checklist – Coal Station at Walvis Bay
C.5.4.1. Associated Costs
The costs for air-pollution control systems are highly site- and performance-specific,
depending on technical and financial factors such as the degree of retrofit difficulty, unit size,
capacity factor, removal efficiency required, labour cost and interest rate, which vary
considerably from country to country and from plant to plant.
Various processes exist to mitigate the sulphur emissions of coal-fired power stations. These
include wet limestone, ammonia process, spray-dry, sorbent-injection, circulating fluidized
bed, and redeemable processes. However, this study will not attempt to describe each of
these processes. As a general guide, Flue Gas Desulphurization (or better known as FGD)
can increase the cost of electricity by about 10%. (International Energy Agency Coal
Research 2001, International Energy Agency 2007)
As highlighted in section 4.2.6.1.1 of possible environmental concern is the suggestion that

the desulphurization plant be dropped from the proposal not only to save costs, but because
'no international environmental standard requires desulphurization for this site' (ECB License
Application 2007, Attachment 1, p 11). It is debatable whether it is wise to even debate the
installation of the seawater flue gas desulphurization given the nature of the location of the
proposed power station site.
Selective Catalytic Reduction is a very effective but expensive means to control NOx
emission. Capital costs for installing Selective Catalytic Reduction on new industrial boilers
range from 6,000 to 10,000 USD/Moto/hr (International Energy Agency Coal Research 2001,
International Energy Agency 2007). In retrofit application, capital costs are about 30 to 50%
higher (International Energy Agency Coal Research 2001, International Energy Agency
2007). The increase in cost is primarily due to modifications to existing ductwork, the cost of
structural steel and reactor construction, auxiliary equipment costs (e.g., additional fans,
ammonia vaporizer, air compressor), and engineering costs. In addition, significant
demolition and relocation of equipment may be required to provide space for the
reactor.(International Energy Agency Coal Research 2001, International Energy Agency
2007).
Page 168 of 222

Figure 3.3 (International Energy Agency Coal Research 2001) offers some general relative
operating costs for some of the major systems at a coal-fired power station.
Figure C-5: Relative operating costs of systems at coal fired power stations
C.5.4.2. Future Trends
There are a number of technology trends being researched, demonstrated and applied
throughout the world to address air pollution and more specifically, carbon emissions. These
include, but are not limited to the utilization of low-sulphur coal, the use of clean coal (usually
washed to reduce the ash content), improvements in the filters of bag-filter technology, the
conversion of coal to gas or liquid fuel, carbon injection into underground caves, and carbon
sequestration into limestone.
Market-based trends include carbon or energy taxes with revenues being used to research
and promote alternative energy sources and carbon trading. This study will not attempt to
give a full description of these mechanisms to control the overall carbon contribution to
climate change.
Page 169 of 222

Coal Plant Investment vs Capacity
1 800
Millions
1 600
y = 1 782 466x
1 400
1 200
Investment USD
1 000
800
600
400
200
-
0 100 200 300 400 500 600 700 800 900
MW Capacity
Figure C-6: Coal plant investment versus capacity
The figure above shows the overnight capital cost trend in relation to plant capacity (derived
from USDOE 2007). This shows a clear linear trend of just under USD 2 million per MW
capacity. Externalities are not included in these costs, and the level of emissions control is
not specified.
Operating costs used in the model were similarly derived from USDOE (1999: 2-4). Coal
costs were modelled with reference to costs stated by NamPower for van Eck, and adjusted
for shorter transport (assuming future plant would be built at the central coast) and for high
operating efficiency. This results in a modelled coal cost of 27.9 c/kWh (NAD, 2007) and
modelled operating cost of NAD 5.4 c/kWh (2007).
The full capital costs including interest during construction and decommissioning used in the
modelling is NAD 19.2 million per MW installed which translates to NAD 19.9c/kWh for a
plant life of 30 years and a yield of 90% of installed capacity. The total cost per kWh
modelled is thus 53.2 c/kWh.
Page 170 of 222

C.5.6. Risks
Severe
Coal availability
Coal transport price
I
Moderate
Coal Price
M
P
A
C Environmental Pollution Grid strengthening
and Carbon required
T
Minor
Low Medium High

LIKELIHOOD
Figure C-7: Risk chart for clean coal power plant
The risks identified in relation to coal plants are illustrated in Figure C-7:
• The highest expected impact risk is the future availability of coal. Coal is a fossil fuel
and thus subject to resource limitations in the long run, but for the expected plant life
it is assessed as a low likelihood risk and has been ignored for the modelling.
• While the likelihood of availability constraints is assessed as fairly low the likelihood
of rising prices is assessed as medium to high, especially due to the high transport
cost component which is sensitive to oil prices which are very likely to remain high.
The risk is that coal is an international commodity and prices will be set on
international markets that Namibia does not influence, making it a combined price
and currency risk. This risk has been reflected in the sensitivity analysis on high coal
price.
• For a coal plant exceeding 350MW at the central coast the transmission grid
(between coast and Windhoek) would need to be strengthened which will have a
small impact on the transmission price path. In the modelling a coal plant exceeding
350MW will not be built, hence this risk has been ignored for the modelling.
• The other significant issue with coal plant is emissions. It is expected that any new
plant in Namibia will be specified to be a “clean” plant with much lower emissions
than van Eck. The policy maker has not made any known determinations in this
regard. Hence there is a risk that developers will seek to minimise cost and invest in
as little cleaning technology as they will be allowed to do. This risk has not been
modelled.
Page 171 of 222

C.6. NATURAL GAS POWER GENERATION

Power generation from natural gas is a well established, mature technology used all over the
world. A combined cycle gas turbine (CCGT) plant is a high efficiency machine available as
off-the-shelf technology (most new gas plants in the USA are CCGT technology
(http://en.wikipedia.org/wiki/CCGT)) and as such has relatively short lead times, however not
taking into account the high current worldwide demand for large plants which is very likely to
raise prices and extend lead times.
Gas in the Namibian context is to be supplied by the Kudu gas field. This gas field has been
subject to attempts at exploitation for more than 20 years, so far without success. The last
seven or so years have seen concerted efforts by NamPower to develop a power station
based on the Kudu gas field. This has, however, so far not resulted in an investment
decision, and most recent indications are that the project will not go ahead in the short term
(NamPower, personal comm. 2007).
The challenges with developing the Kudu gas field include the relatively high costs of
developing the upstream gas side with the field being located 170 km out at sea (CSIR &
NamPower 2004) and at a depth of 170m and disagreement about the price and currency in
which the gas would have to be paid.
years (CSIR & NamPower 2004). It could thus sustain a 400MW plant (which would be much
more suitable for the Namibian system) for 40 years if the gas infrastructure and repayment
period could be arranged accordingly.
More recent variations of the Kudu theme have included the possibility of building a 400MW
plant in Namibia and piping gas down to the RSA, or even piping all the gas down to RSA
(NamPower 2007, slide 5). The end result so far has however been a failure to get
agreement on gas price and currency risk sharing as mentioned above.

The combined cycle gas turbine plant works with dual generators, one driven by a gas
turbine, the second being a steam turbine driven by steam produced by using the exhaust
gases from the gas turbine. This is illustrated in the diagram below
(http://en.wikipedia.org/wiki/CCGT, accessed March 2008):
Figure C-8: CCGT Schematic Layout
Page 172 of 222

A CCGT plant typically achieves efficiencies of up to 59% for electricity only plant. If the heat
is also utilised then the output efficiency can rise to 85% (http://en.wikipedia.org/wiki/CCGT,
accessed March 2008).
Due to the single gas resource in Namibia a CCGT plant would be a centralised facility
located in the extreme south west of the country. Gas turbines come in various sizes and
could thus be sized in a modular fashion. The Kudu developers have considered plants
between 400MW and 800MW.
C.6.2. Resource & Generation Potential

years (Tullow/NamCor 2006, slide 5). The resource could thus theoretically sustain a 400MW
plant (which would be much more suitable for the Namibian system) for 40 years if the gas
infrastructure and repayment period could be arranged accordingly. It is not clear what
obstacles might be encountered in pursuing this avenue or what the cost implication thereof
would be.
A CCGT plant would be expected to have an availability of 85% (NERC 2006) or better, as
these plants are robust and well established technology. Dispatchability for the Kudu plant
would depend very much on the gas flow issues being resolved. In the configuration initially
contemplated Kudu would have been a base-load plant. More recently mid-merit operation
has apparently been considered. However few details have been made public.
The modelling for the purposes of this report has therefore remained with the “older”
assumption of a base-load plant.

Due to confidentiality of the Kudu project the figures used in the modelling have been derived
exclusively from public international sources (USDOE 1999: 5). The final numbers come
fairly close to pessimistic estimates that have previously been mentioned for Kudu (ECB
2007a: 27).
The figures that have been used stem from research and statistics published by the USA
Department of Energy (USDOE 1999: 5).
Figure C-9: Gas Plant Fuel Cost Trend
Page 173 of 222

Figure C-10: Gas Plant Operating Cost Trend
The above two charts show fuel and operating cost trends for gas and coal plants in the
USA. From these a typical fuel cost of USD 4 cents/kWh and a typical operating cost of USD
0.5 cents/kWh have been derived and used in the model. When converted to NAD values
this amounts to NAD 4.5c/kWh operating cost and NAD 36.1c/kWh fuel cost.
The full capital costs including interest during construction and decommissioning used in the
model is NAD 6 million per MW installed capacity. This translates to a fixed cost of NAD
7c/kWh assuming a plant life of 20 years and a yield of 90% of installed capacity. The total
cost modelled is NAD 47.6 c/kWh.

The possible future development of a gas field off the coast of Namibia presents a number of
challenges. Environmental impact assessments were developed for the proposed Kudu
combined cycle gas turbine (CCGT) in 1998 and 2004 (CSIR Environmentek 2004). These
two EIAs and stakeholder consultation identified matters affecting Oranjemund and its
environs and site specific issues. Only the highlights will be listed below.
Issues affecting Oranjemund and its environs included:

• A large, 'foreign' workforce;
• Impacts to the roads and traffic;
• Impacts on security;
• Impact on noise particularly at night;
• Impact on construction noise;
• Impact on power station operational noise;
• Air pollution with emissions of nitrogen being considered the most important on its
impact to vegetation;
• Visual impact of the power station with particular focus on the Orange River Mouth
which was designated a Wetland of International Importance (also known as the
Ramsar Convention on Wetlands of 1971) on 23 August 1995. Any changes as a
Page 174 of 222

result of technological developments, pollution or other human interference must be

communicated by Namibia to the International Union for the Conservation of Nature
and Natural Resources (IUCN) (CSIR Environmentek 2004);
• Visual impact of power transmission lines across the main entrance of the town of
Oranjemund;
Site specific issues highlighted included:

• Impacts of low hummock vegetation during construction and operational phases;
• Managing the impacts of construction and operational phases. It was noted that due
to mining activities in the area that relatively little pristine areas remained on the
proposed site. Construction activities should not be allowed to approach the
undisturbed hummock vegetation;
• General wastes should be disposed of at the waste facilities for the town while any
hazardous waste is to be delivered by Namdeb to the Windhoek hazardous waste
disposal facility in Kupferberg;
• Purge water discharge beyond the surf zone from the power plant will comply with the
World Bank guidelines for effluent disposal in terms of temperatures and biocides.
However, elevated salinities and temperatures in the surf zone may have a significant
barrier effect on larvae of fishes and invertebrates that are transported by the littoral
drift (CSIR Environmentek 2004).
As the authors did not have access to the environmental studies of the proposed gas field, it
can only be assumed that the exploration, drilling, piping and related facilities could impact
the local marine environment (flora and fauna). Water quality monitoring will need to be
installed to ensure fuel leaks and spills do not occur and in the unlikely event of a spill, that
appropriate action can be taken to contain and clean ocean and coastal environments.
Emergency response measures will need to be developed in case of an event, e.g. fire.
In addition to piping of the fuel under water and onto land, any on-land storage presents its
own challenges. These include the risk of fuel leaks and spills, contamination of surface- and
ground-water, the release of air emissions into the atmosphere from the storage and transfer
for the fuels, and in case of emergencies, fire. Appropriate measuring, monitoring and
reporting mechanisms will need to be developed and implemented to address these factors.
Page 175 of 222

Employment possibilities Limited possibilities for the local population due to low staff numbers.
Poverty alleviation Little effect can be expected from the power project.
Light, Cooking, Refrigeration, Hot Water No likely effect due to location near urban area.
Health and safety No likely positive effect due to location near urban area.
medication
Accidental risks, emergency situations Unlikely to have significant accident risks due to likely location
and response; outside populated area.
Construction camps and power line Insignificant likely effect due to location near urban area.
Damage to cultural and natural assets Insignificant likely effect due to location near urban area and location
in a mining area where nature has already been extensively
disturbed.
Human health impacts, e.g. Not likely to have significant local effects.
Resettlement (if applicable) and the None.

Effects on tourism industry Insignificant likely effect due to location in mining area which has very
little tourism potential.
Visual impacts and Aesthetics Limited likely effect due to location in mining area and industrial zone.
Shadow disturbances None
Table C-5: Social Impact Checklist – Gas Power Station in Oranjemund Area
Page 176 of 222

C.6.5. Risks
Severe
Gas reserve /
gas extraction issues
Gas Price
Large station - forex risks
surplus export price
and sales risk
I
Moderate
M
P
Lead time
A extended due to
C high demand for
power stations
T
Minor
Low Medium High

LIKELIHOOD
Figure C-11: Risk chart for Gas CCGT power plant
Figure C-11 illustrates the identified risks for a gas power station in Namibia:
• The most critical risk lies in the significant forex risks inherent in the gas pricing, as
well as the risk of not achieving an acceptable export price for excess power that
would be produced by a plant of this type. These issues and the interplay between
them are seen as having contributed significantly to the Kudu project not reaching an
investment decision.
• There is some risk inherent in a gas field, without proven actual extraction, that
extraction of the gas may prove more difficult than expected. This risk is however
perceived to be of low probability, but might be catastrophic if it does materialise.
There is also risk that technical issues in extracting the gas and piping it to shore may
prove more difficult than expected, leading to possible increases in the gas cost.
• Another significant risk is that due to huge increased international demand for gas
turbines as well as gas installation experts and equipment the lead lead-time for
getting the Kudu gas project into production may turn out to be far longer than
expected. This might critically impact the electricity supply of the country should Kudu
be relied on to serve demand with no alternatives available.
Page 177 of 222

C.7. NUCLEAR POWER GENERATION

Nuclear power plant (NPP) technology has been in use for over 50 years and currently about
30 countries rely to some extent on NPP for their power generation needs. In 2005 nuclear
energy contributed 16% to world electricity generation (Ocheltree 2006).
During the 80’s and 90’s the interest in NPP waned considerably, however recent increases
in coal and gas prices as well as climate change concerns have led to a resurgence of
interest in nuclear power.
C.7.1. Technical issues

The most distinctive part of a nuclear power plant is its nuclear reactor. The fuel in the
reactor undergoes nuclear fission chain reaction, releasing energy in the process. The
energy is used to generate steam that drives a steam turbine generator. A substance called
moderator is necessary to sustain the chain reaction, which in conventional plants is
controlled by insertion of control rods in the reactor.
Conventional NPP reactor life is quoted as 30 years but many NPP operators are currently
considering to ‘life extend’ existing reactors to operate for 40 years. Newer generation units
(3rd generation) have a life of about 60 years (Thomas 2005: 17).
For a Pebble Bed Modular Reactor (PBMR) this study assumes a 30 year service life simply
due to its current developmental stage and lack of track record.
Nuclear power plants are considered large-scale centralised plants.
The lead time for the implementation of a NPP is assumed to be 6 years for the project
development phase (design, planning, EIA, permissions etc) and 4 years for the
implementation of the plant (Thomas 2005: 14).
C.7.2. Resource
Although Namibia has the fourth largest mining output of uranium in the world, the uranium
still has to go through a number of processes before it is converted into reactor fuel. While
Namibia is endowed with the resource for reactor fuel, only mining and production of
“yellowcake” currently take place in the country. The processing of uranium to reactor grade
fuel is a high tech and costly process. The cost of uranium in the reactor grade fuel only
make 25% of the overall fuel cost (Australian Uranium Association 2007). The more
significant cost portion of the fuel costs come from the enrichment process.
There is little consensus in literature as to the actual worldwide reserves of uranium.

Assuming continued operation of all existing NPPs one source estimates that the reserves
will last for 70 years (OECD Nuclear Energy, International Atomic Energy Agency 2003).
Page 178 of 222


The capacity of conventional NPPs varies within a wide range – from about 400MW to more
than 2,000MW. The average conventional NPP size is 1,000MW. Economies of scale are
achieved in large plants.
The generation potential is limited by the total finite world supply of uranium, world wide
demand and ability to utilise the generated electricity. It is not clear whether Namibia would
be able to secure the purchase of enriched nuclear fuel as there may be particular
requirements which Namibia cannot meet or only at very high cost. This would have to be
explored in more detail but exceeds the scope of the study. For the purposes of this study it
is assumed that Namibia will be able to secure enriched nuclear fuel.
Considering that Namibia’s demand reaches 430MW, a single conventional NPP of

approximately 900MW (currently the smallest commercially available plant size) would
generate in excess of Namibia’s demand and can only be considered if the necessary long
term power purchase agreements can be secured. In addition the associated energy
production portfolio risks are substantial if such a power plant is pursued.
The more attractive option is the Pebble Bed Modular Reactor (PBMR) which is currently
being developed in South Africa and has an anticipated size of 165MW. This size of plant
can be integrated into the Namibian generation portfolio in terms of its size and energy mix.
Since this technology is not yet commercially available it has clear risks attached, in
particular if such a plant is planned for implementation in the near future. After discussions
with the client it was however decided that the PBMR is the more reasonably sized NPP as
opposed to the conventional and that this study rather models a PBMR. Cognisance is taken
of the uncertainties in the successful commercial development of this technology by
assuming that PBMR technology will only become available by 2017 since the first plant is
scheduled to go online in South Africa in 2013.
Nuclear power plants are base-load generators that operate at more or less full capacity.

The possible social and environmental impacts of conventional nuclear power plant
operations are (World Business Council for Sustainable Development 2007 and USDOE
Energy Information Administration n.d.):
• Main environmental impacts tend to occur at the beginning and end of the fuel cycle,
i.e. when fuel is being processed and waste disposed. During normal power plant
operations environmental impacts relate to the use of cooling water.
• Radioactive releases from normal operation versus those from unexpected
occurrences.
• Release of radioactive particles would increase the level of radioactivity in the
receiving environment including air, water, soils, animals, plants, buildings, etc.
• No SO2, NOx or CO2 emissions (refer below).
Page 179 of 222

The Environmental Protection Agency (EPA) identifies the following average emission levels
in the production of 1 MWh of electricity
Pounds of Emissions per MWh
Coal Oil Natural Gas Nuclear
Carbon Dioxide 2249 1672 1135 0
Sulphur Dioxide 13 12 0.1 0
Nitrogen Oxides 6 4 1.7 0
Source: www.epa.gov/clean energy/impacts (as cited in USDOE EIA n.d.)
Table C-6: Average Emission Levels

• Radiation exposure concerns.
• Public health and safety.
• Uncertainty and associated risks of nuclear power (perceived and real).
• Nuclear power industry image (real and perceived).
• Waste from a nuclear plant is primarily solid waste, spent fuel, and some process
chemicals, steam and heated cooling water. The greatest environmental waste
concern at an operating nuclear power plant is spent fuel disposal (USDOE EIA n.d.).
• Disposal of radioactive waste (low-level, intermediate level and high level). Nuclear
power produces around 2,000 metric tonne/per annum of spent fuel. This amounts to
0.006 lbs/MWh. If a typical nuclear power plant is 1000 MW in capacity and operates
91% of the time, waste production would be 45,758 lbs./annum or slightly less than
23 tons. The solid waste from a nuclear power plant is thus not the volume of the
waste, which is very small, but the special handling required for satisfactory disposal.
A similar amount of electricity from coal would yield over 300,000 tons of ash,
assuming 10% ash content in the coal. Processes (specifically scrubbing) for
removing ash from coal plant emissions are generally highly successful but result in
greater volumes of limestone solid wastes (plus water) than the volume of ash
removed (USDOE EIA n.d.).
• There are very few, if any, long-term waste disposal sites internationally. The
associated costs of storing and keeping these wastes safe for hundreds of years are
as of yet, unknown. As the wastes remain radioactive for many years to come, future
generations will have to pay for their long-term management.
• Decommissioning costs might represent up to 50% of the discounted investments
made for the nuclear part of the power plant (European Commission 2007). Sound
financial provisions for decommissioning should reduce the potential burden on future
generations. The study assumes a 30 % decommissioning cost.
The PBMR uses pyrolytic graphite (instead of water) as the neutron moderator, and an inert
or semi-inert gas such as helium, nitrogen or carbon dioxide as the coolant (at very high
temperature), to drive a turbine directly (Wikipedia 2008c). The gases do not dissolve
contaminants or absorb neutrons as water does, so the core has less in the way of
radioactive fluids and is more economical than a light water reactor (Wikipedia 2008c).
The PBMR primary coolant is helium. The helium directly turns low-pressure turbo-
machinery, without intervening losses from heat-exchangers. Helium is well-favoured
Page 180 of 222

because it is chemically inert, and neutrons do not transmute it to a radioactive element

(Wikipedia 2008c). This means that the turbo-machinery should not become radioactive,
even though it operates on primary coolant. While the fuel design is robust, a fuel defect
could still contaminate the power production equipment. One disadvantage is that the turbine
must be somewhat larger, and therefore more expensive.
Various environmental and social issues concerning pebble bed reactors are that (Wikipedia
2008c):
• Claims that its reactor was designed to desalinate seawater. This could be seen as a
positive spin-off of such a project in a region short of fresh water;
• Most reactor systems are enclosed in a containment building designed to resist
aircraft crashes and earthquakes;
• Encasing the fuel in potentially combustible graphite poses a hazard;
• There is less experience with the production of PBMRs than light water reactors.
Claims made by both proponents and opponents are theory-based rather than
practical-based;
• Since the fuel is contained in graphite pebbles, the volume of radioactive waste is
much greater, but contains about the same radioactivity when measured in
Becquerels per kilowatt-hour. The waste tends to be less hazardous and simpler to
handle;
• The wastes will need to be safely contained adding to the already apparent waste
disposal problems of nuclear waste. The radioactive waste must be stored for many
human generations, reprocessed, transmuted in a different type of reactor, or
disposed of by a method yet to be devised;
• Defects in the production of pebbles may cause problems. The graphite pebbles are
more difficult to reprocess due to their construction. However, this could be seen as a
positive outcome as it is difficult to re-use PBMR waste for nuclear weapons.
In supporting a nuclear power industry, either conventional or pebble bed, Namibia would
have to develop and implement a number of programmes. These might include:
• Skills development and training of nuclear engineers;
• Monitoring of radioactivity at and around nuclear power station;
• Legal obligations and associated costs;
• Handling, transport, disposal and management of radioactive waste (low-level,
intermediate level and high-level);
• Decommissioning of nuclear power plant sites;
• Safety management systems; and
• Conditions of license agreements which may require further monitoring and reporting
to authorities.
The trends of the nuclear industry include renewed support for nuclear power due to sharp
raise in conventional fossil-fuel prices, the renewed support for nuclear power due to greater
awareness and evidence of climate change, the quantity of waste per unit of electricity
produced has fallen in most countries, the collective radiation exposure to employees has
fallen, and the sharp rise in processing uranium and all associated materials due to high
demand (World Association of Nuclear Operators 2005). However, this form of energy may
not support one of the overall objectives to provide sustainable energy to the country of
Page 181 of 222

Namibia without doing harm to the environment and taking into consideration future
generations.
Employment possibilities Limited possibilities for the local population due to low staff numbers.
Poverty alleviation Little effect can be expected from the power project.
Health and safety No likely positive effect due to location near urban area.
medication
Accidental risks, emergency situations Major accident risks due to location near populated area.
and response;
Damage to cultural and natural assets If located outside town significant effects on natural assets are
possible and will need to be taken into account during site selection.
Human health impacts, e.g. Not likely to have significant local effects.
Resettlement (if applicable) and the Not likely.

Effects on Aviation Not likely.
Effects on tourism industry Possible significant effects since this is a major tourism area.
Visual impacts and Aesthetics Possible significant effects depending on site selection.
Shadow disturbances Not likely.
Table C-7: Social Impact Checklist – Nuclear Power Station at Walvis Bay

The review of literature (Thomas 2005, WNA report 2005, Ocheltree 2006) shows that the
cost of nuclear power plants varies significantly, i.e. there is little worldwide consensus on
cost. One reason for this lack of consensus is that many utilities may hide the actual cost of
nuclear energy in their books in order to make nuclear power seem more attractive due to its
otherwise controversial status. It is accepted though that nuclear power plants are capital-
Page 182 of 222

intensive implementations. Despite the high capital costs nuclear plants can however prove
more cost effective in scenarios of high gas prices and carbon penalties.
The costs of nuclear power generation are categorised as:
• pre-development cost,
• capital cost,
• operating cost,
• waste disposal cost,
• decommissioning cost.
The costs for a PBMR are currently based on estimates received from PBMR (Pty) Ltd
(Personal comm. with Claasen, G from PBMR Company). The capital cost of a PBMR was
quoted as USD 3.3 million per MW. The study assumed a cost of USD 4 million per MW due
to additional project development costs, grid extension costs and likely escalation cost in
bringing this product to the market. In addition to this capital cost a decommissioning cost of
30% was assumed. This covers the decommissioning of the site, but does not cover the
long-term costs of nuclear waste storage facilities which would add significantly to the price
of the plant, in particular the long-term obligations which arise with the safety issues of such
waste.
The operating and fuel costs are estimated at USD 1 c/kWh each. There are sufficient
precedents for operating costs of a nuclear plant. However, this is not the case for the fuel
costs related to the operation for of a PBMR plant. The fuel cost has therefore been based
on the cost of fuel for conventional reactors with an additional margin for the specialised
manufacture of the fuel pebbles.
Although the price of uranium has increased twenty fold over the last few years, this will only
impact on about a quarter of the overall enriched nuclear fuel price since the conversion and
enrichment are a more costly process than the cost of uranium itself (Australian Uranium
Association 2007).
operating life of 30 years and a yield of 95% of installed capacity. Operating cost is modelled
at NAD 7c/kWh and fuel cost at NAD 7c/kWh.
C.7.6. Risks
The risks associated with nuclear power plants are extensive. An overview is provided in the
risk chart in Figure C-12. The main risks are listed below.
• Although the likelihood of major accidents is very low they cannot be excluded. And if
a catastrophic accident does take place, then the consequences are often detrimental
far beyond the economic benefit of such a nuclear power station.
• Long-term nuclear waste storage remains a major issue. The full cost implications
have not been established and the storage costs are in many countries carried by
government. This is the case in the USA, where government will build the final
storage facility in the desert, but has not yet done so (after more than 20 years). The
Page 183 of 222

cost of such storage is therefore in most cases not included in the financial figures
even though the economy bears the burden.
• Furthermore, the safety and security of the waste storage facility has to be ensured
while the waste remains radio-active (thousands of years) – this cost too is ignored in
current financial models for NPPs.
• Nuclear power plants cannot be insured against the consequential damages in the
case of an accident. An accident with major consequences would bankrupt any
insurance company. Therefore society actually carries the risk.
• There are resource concerns for uranium due to the current renewed interest in
NPPs. Again, there is not much consensus in terms of the remaining resources but
should there be a resource constraint then this would have major implications,
besides leading to risks of resource wars such as are seen for oil. It is unlikely that
Namibia will not be able to procure reactor grade fuel since Namibia is an exporter of
yellow cake, which provides some leverage for entering into long term supply
contracts.
• Namibia does not have a technological nor manufacturing base that can meaningfully
support such an industry, which implies that most technical components would have
to be imported. This exposes the operator/owner to unforeseeable forex risks and
price escalations.
Inability to
Risk of
procure fuel or
Severe
catastrophic
enter into contracts for
accidents
fuel
High level specialised

skills required for Inability to get
operation- problem international permissions
Security risks and
to attract & retain for plant construction
I associated costs, ability
to secure the installation
Moderate
M against attack PMBR technology Spent fuel disposal,

not yet proven commercially, Fuel transport,
P may extend lead time long term waste
and/or increase cost storage
A
C
T
Volatility of
Minor
fuel prices
Uninsurable
Low Medium High

LIKELIHOOD
Figure C-12: Risk chart for nuclear PBMR power plant
Page 184 of 222

C.8. ELECTRICITY GENERATION FROM BIOMASS

Biomass is a collective term for organic matter that could be used in energy production.
Diverse fuels derived from wood, agricultural wastes, food processing wastes, animal
manure, solid municipal waste and others embody the term biomass.
Biomass can be burned directly in boilers but the pollution produced and the low energy gain
when this method is utilised have been the driving forces in the development of new methods
of energy production from biomass.
The Desert Research Foundation of Namibia (DRFN) is currently implementing a proof-of-

concept project referred to as CBEND (Combating Bush Encroachment for Namibia’s
Development), which is to demonstrate the gasification technology and focus on the
operational and grid integration issues to guide a future larger scale role-out of biomass to
electricity plants within Namibia. The costing information of this supply option is largely based
on the DRFN concept paper (Hager, Schultz & von Oertzen 2007a).
This study focuses on the CBEND approach and models small wood gasification plants
(0.5MW to 1MW) for a number of reasons which include:
• Limitations of connecting large scale plants to the existing networks (smaller than
2MW).
• Mobility of plants requires smaller scale plants in order to harvest the bush in close
vicinity and not to incur higher transportation costs, which will be dependent on
imported fuels.
• Provide a macro-economic analysis of using up to 60MW of wood gasification plants,
based on the CBEND approach
Although the biomass to electricity power plant options are extensive (Hager, Schultz & von
Oertzen 2007a, Orjala & Seppälä 2007) it is beyond the scope of this study to investigate the
optimal solutions but rather to base this particular supply option on the approach taken by
DRFN. This is further motivated in the text below.

Energy generation through biomass gasification goes trough the following stages36:
o harvesting or biomass collection
o processing of the biomass into fuel-which involves chopping, briquetting, drying
o gasification of the organic fuel
o gas cleaning
o electricity and heat production
During the gasification process, pyrolysis (thermal decomposition of biomass in the absence
of oxygen) causes the release of gases that are later used in controlled combustion. It is
preferable to use biomass with low moisture content as the moisture (water vapour) subdues
the pyrolysis process. In addition to moisture, content feedstock particle size, form, bulk
36
Source: http://members.tripod.com/~cturare/ene.htm
Page 185 of 222

density, energy content, ash content and composition of the fuel are all factors that influence
gasification (Hager, Schultz & von Oertzen 2007a).
The solid remaining product of pyrolysis is known as char or charcoal and can be used as a
crop fertiliser while the inevitable heat produced in the process can be used for drying or
other heat requiring technologies.
The service life of the technology is assumed to be 20 years.
The technology is highly modular in terms of plant size and can be moved to different
locations, provided that the grid infrastructure is present.
Implementation lead times for biomass to electricity plants are of the order of 2 years.
C.8.2. Resource
As pointed out by Hager, Schultz & von Oertzen (2007a), 26 million hectares of invader
bush, mostly in Northern Namibia, are to be found, representing a growing problem for
farmers and have caused a loss in reducing the number of cattle bred of estimated NAD 700
million in 2005. An estimate indicates that a maximum of 18 metric tons invader bush of 35%
moisture content can be harvested per hectare. Averages are however much lower.
Expectations are that the “harvested plot will show significant re-growth when the next
harvesting cycle commences five to eight years after the first cycle. In this meantime, the
land can be used as grazing area for cattle.” depending on if and how an aftercare
programme was applied.
Figure C-13 (Hager, Schultz & von Oertzen 2007a) provides an overview of the geographic
spread of the invader bush resource in Namibia.
Figure C-13: Overview of biomass resources in Namibia
Page 186 of 222


According to Hager, Schultz & von Oertzen (2007a) the sustainable energy yield per hectare
is about 2MWh in bush-infested areas in Namibia. If for example 500GWh per annum should
be generated through biomass then it will take approximately 250,000 hectares to supply this
demand. It therefore becomes apparent that the resources at 26 million hectares are
substantial and that this technology can contribute significantly to the energy mix in Namibia.
Since there is little international experience in operating many gasification plants and feeding
their output into the electricity grid, this study considers a maximum of 60MW of biomass to
electricity plants. This is significantly below the generation potential of Namibia’s resources.
The energy output from such a 60MW plant would be 378GWh per annum.
Compared to solar and wind renewable energy generators, biomass to electricity generation
has the advantage of being dispatchable power due to the stored energy in the biomass.
This allows this power generation to be utilised as base-load to mid-merit plant.

Various social and environmental impacts of biomass projects are cited by the US
Department of Energy and the Power Scorecard, however, they may not apply in total for the
Namibian situation:
• Air pollution from the burning of carbon-based wood or plant material
o Sulphur dioxide emissions are lower than in coal-fired power stations.
o Reduction in nitrogen oxide (NOx) emissions if biomass is co-fired with coal.
Also, NOx emissions will vary depending upon design and control of the
combustion facilities.
o Carbon monoxide (CO) levels that are higher than those of coal-fired power
plant.
o Particulate emissions may impact public health if not managed through
appropriate filter technology.
• Biomass is assumed to yield no net emissions of CO2 because of the sequestration of
biomass during the planting cycle. However, the entire process (harvesting,
transportation, and the conversion to electricity) can be considered to be a small,
positive net emitter of CO2.37
37
There are two common, but mutually exclusive, impressions about biomass fuels and carbon dioxide (Oak Ridge National
Laboratory n.d.). One first impression is that biomass fuels and fossil fuels are not different because, when burned, both yield
carbon dioxide. This is true if land from which biomass is harvested for fuel is not replanted and instead is converted to other
uses. However, if the biomass is produced sustainably, the growing trees and other plants remove carbon dioxide from the
atmosphere during photosynthesis and store the carbon in plant structures. When the biomass is burned, the carbon released
back to the atmosphere will be recycled into the next generation of growing plants. When biomass is used for fuel in place of
fossil fuels, the carbon in the displaced fossil fuel remains in the ground rather than being discharged to the atmosphere as
carbon dioxide. The productivity, or rate of growth of the trees, becomes an important consideration. While slow-growing trees
can take a very long time before the released carbon is recaptured in the next generation of trees, fast-growing trees can
recycle carbon rapidly and will displace fossil-fuel use with every cycle.
A second impression is that biomass energy systems, because they recycle carbon, produce no net emissions of carbon
dioxide. This is not strictly true either. It takes some energy, much of it now provided by fossil fuels, to grow and harvest
biomass fuel crops and to haul the fuel to a power plant. The use of biomass fuels does result in some discharge of carbon
dioxide. The extent to which biomass fuels can displace net emissions of carbon dioxide will depend on the efficiency with which
they can be produced and used.
Page 187 of 222

• Possible reduction in the production and release of methane gas due to the diversion
of wood/crop waste from landfill / rotting on ground.
• Possible soil erosion of land disturbed for harvesting of biomass both by the removal
of vegetation, but also in the roads and infrastructure to access the biomass.
• Possible loss of soil quality from the loss of nutrients that the forest /crop would have
provided. This could result in the loss of long-term productivity of the land.
• Variation in yields due to differences in weather, geology, and soil conditions, but also
the loss of yields due to forest /crop fires.
• Fertilizers, pesticides and herbicides and the potential for groundwater and surface
water pollution.
(United States Department of Energy, Energy Information Administration August and

November 2007 and the Power Scorecard. Electricity from Biomass December 2007).
Changes in land use and land coverage important drivers of water, soil and air pollution
(Bringezu, Stefan, et al. 2007). Vegetation removal leaves soils vulnerable to massive
increases in soil erosion by wind and water, especially on steep terrain, and when
accompanies by fire, also releases pollutants into the atmosphere. This can degrade soil
fertility over time, reduce the suitability of the land for future agricultural use, but also release
vast quantities of phosphorus, nitrogen, and sediments to streams and other aquatic
ecosystems (Bringezu, Stefan, et al. 2007).
Further to this discussion, Bringezu, Stefan, et al. (2007) suggests that impacts on global
land use may result in significant changes in land cover. They state that the impacts are
directly related to the survival or extinction of plants and animals (e.g. forest clearance) and
whole species (e.g. if the cleared land area exceeds the population area) questioning
whether the unlimited expansion of arable land at the expense of major destruction of natural
ecosystems should be regarded as a sustainable solution (Bringezu, Stefan, et al. 2007).
Bringezu, Stefan, et al. (2007) state that within a consistent sustainability strategy the level of
regional consumption of any raw material should not deteriorate critical environmental
resources on a global scale. They list the following for sustainable use of biomass:
• Energy and material crops can contribute only a certain share to a country's (and
world's) material and energy supply,
• Environmental and social impacts of increased biomass use should be considered

and include global challenges such as poverty reduction, access to water and energy,
implications of climate change, etc.,
• Due to substitution and competition effects, any biomass strategy need to consider
the interrelationship of material, energy and land use and should be embedded into a
cross-sector strategy for sustainable use and management of resources,
• A significant increase in resource efficiency considering renewable and non-

renewable resources is necessary to fulfil a rising demand of the world economy for
material and energy services.
Forests that are not harvested do not continue to accumulate carbon indefinitely. They eventually approach maturity and
achieve, over time, a balance between the carbon taken up in photosynthesis and the carbon released back to the atmosphere
from respiration, oxidation of dead organic matter, and fires and pests. If fossil fuels continue to be used to meet society's
energy needs, reforestation or forestation of ever larger areas would be needed to prevent increasing concentrations of
atmospheric carbon dioxide.
Page 188 of 222

A positive impact of developing a biomass industry in Namibia would be the creation of jobs
in areas where the biomass is grown and managed and the possible alleviation of poverty in
these areas. Based on Hager, Schultz & von Oertzen (2007b) it is assumed that 4 skilled and
52 unskilled jobs are created per operational MW, while half a skilled job and one unskilled
job is created per MW during construction of the power plant. In a World Wide Fund for
Nature and European Biomass Association (AEBIOM) study reported by Science in Africa
(2004), the production of biomass will create up to 400,000 jobs by 2020, particularly in rural
areas. In the Northeast, Southeast and West Coast regions of the US alone, the biomass
industry has already provided about 70,000 jobs.
A very strong motivator for this option would also be the inclusion of communities in the
industry to ensure that they have a direct role in the management of the land in their area,
taking ownership for the care of the vegetation and thus supporting sustainable development
principles by promoting a cleaner, renewable fuel within a local context. It must be assured
through agreements that the affected community benefits from the biomass-to-energy supply
option, either through joint land-use, monetary rewards and/or other forms of compensation.
Employment possibilities Major possibilities for the local population.
Poverty alleviation Major positive effect can be expected from the power project due to
the large numbers of potential employees and the low level of skill
required.
Light, Cooking, Refrigeration, Hot Water Limited possible effect since most settlements in the target areas are
already electrified.
Health and safety No likely positive effect. Likely negative effect due to emissions and
physical nature of bush harvesting activities.
HIV /AIDS Unclear what the effect may be.
Hospitals and clinics, e.g. refrigeration of No likely effect due to settlements and facilities already having been
medication electrified.
Schools No likely effect due to settlements and facilities already having been
electrified.
Accidental risks, emergency situations Significant accident risks due to physical nature of harvesting
and response; operations, also fire risk during dry seasons.
Construction camps and power line Insignificant likely effect due to decentralised small scale nature of
construction activities; the power project and likely use of existing power lines.
Damage to cultural and natural assets Insignificant likely effect due to decentralised small scale nature of
the power project and likely use of existing power lines.
Human health impacts, e.g. Insignificant likely effect due to decentralised small scale nature of
electromagnetic fields from power lines the power project and likely use of existing power lines.
Resettlement (if applicable) and the None likely.

Effects on Aviation None likely
Page 189 of 222

Effects on tourism industry Insignificant likely negative effect due to decentralised small scale
nature of the power project and likely use of existing power lines. May
have positive effect due to the removal of invader bush.
Visual impacts and Aesthetics Insignificant likely negative effect due to decentralised small scale
nature of the power project and likely use of existing power lines. May
have positive effect due to the removal of invader bush.
Shadow disturbances None likely.
Table C-8: Social Impact Checklist – Biomass to Electricity

The costs for a biomass to electricity operation are contained in the hardware, the fuel costs
for the biomass collection efforts and in the labour. The costing assumes that the actual
biomass fuel is free-of-charge as the land owner will be rid of the invader bush to free up
land for better productive uses.
The investment costs are calculated as NAD 16.7 million per MW for the plant and the grid
integration hardware (including protection requirement) as well as the project management
and engineering costs. These costs are based on (Hager, Schultz & von Oertzen 2007b) and
allow for additional grid integration cost.
The operating costs are estimated at NAD 4.7c/kWh and the fuel costs (transport, shredding)
at NAD 5.4c/kWh. These cost estimates are based on the work of CBEND (Hager, Schultz &
von Oertzen 2007b).
modelling is NAD18.6 million per MW installed which translates to NAD 25.9c/kWh over an
operating life of 20 years and a yield of 74.4% of installed capacity. Operating cost is
modelled at NAD 4.9c/kWh and fuel cost at NAD 5.6c/kWh.
C.8.6. Risks
The potential risks associated with this supply option are shown in the risk chart in Figure
C-14. The risks as listed as follows:
• The technology is not sufficiently mature to allow off-the-shelf ordering and operation,
which is significant if up to 60MW is to be introduced into the Namibian power sector.
• As in any power plant, the reliable operation is essential in order to guarantee the
supply. In that sense the biomass to electricity generation is exposed to labour
related volatilities since the working conditions out in the field will always be
challenging. The work force needs to be moved around in remote areas and will not
reside at home. The power plant operation will therefore be exposed to labour issues
that can impact significantly on the ability to deliver contractually binding power to
NamPower, the REDs or any other party that signs the power purchase agreement.
• A number of uncertainties will impact on the cost of this supply option. This includes
logistical inefficiencies that may not have been taken into consideration, possible
charges for the invader bush biomass resource by the land owner and additional
costs for grid integration of the generators. For that reason, the proof-of-concept is
essential in establishing some real cost figures.
Page 190 of 222

• It is not clear whether the environmental impact of this supply option may not have a
number of negative impacts as listed in section C.8.4.
Severe
Technology may not

work as intended or not
be as reliable as expected
Grid integration more
costly and complex than
Labour stability issues
anticipated
in fuel harvesting affecting
labour cost and hence
I Logistics and transport financial performance
Moderate
fuel costs for biomass

M collection exceed
P estimates Environmental impacts:
* land degradation
A Landowners may require
* plant and animal life impacts
* SOx and NOx
C payment for supply of
fuel wood
T Labour stability issues
in fuel harvesting affecting
production reliability
Minor
Low Medium High

LIKELIHOOD
Figure C-14: Risk chart for biomass to electricity power plant
Page 191 of 222

C.9. CONCENTRATING SOLAR THERMAL POWER PLANTS

Concentrating solar power (CSP) plants have been operating in the Mojave Desert in the
USA for more than 20 years (Jones 2007: 36). There has been no new multi-MW CSP plant
construction in the last 15 years but this is about to change as approximately 2,600 MW of
CSP plants is in the pipeline, some of it under construction (Cameron & Jones 2007).
C.9.1. Technology Status

The principle of operation of concentrating solar power plants is much the same as that of
conventional power plants. The difference is that the energy input in CSP plants is obtained
by concentrating solar radiation and converting it to high temperature steam or gas that
drives a turbine or a motor engine. CSP plants consist of two major components: the first
being the solar radiation reflector and the second being the conversion of the solar thermal
energy to electric power.
CSP plants are expandable to some extent, provided that this has been incorporated at the
initial planning stage. CSP plants do not lend themselves to distributed generation as the
maintenance and operating requirements are more suited to centralised stations.
Implementation lead times are in the range of about 3 years and include 1 year of project
development and two years for construction. Jones (2007: 40) indicated that a 64MW facility
was implemented in just over one year’s time.
The service life of CSP plants is estimated at approximately 30 years, based on the Mojave
Desert performance. This is an assumption.
The CSP plants are classified in terms of their solar collector type as parabolic trough,
central receiver, and parabolic dish plants. The availability of CSP can be extended beyond
daylight hours by the use of thermal storage.
C.9.1.1. Parabolic Trough Plants
Parabolic trough CSP constitute the largest installed capacity and have been operating
reliably since the 1980s (Cameron & Jones 2007: 168). Parabolic trough CSPs are modular.
Each module consists of long parallel rows of reflectors (glass mirrors) in the shape of a
trough, which tracks the sun during the course of the day. A fluid, referred to as heat transfer
fluid, circulates in pipes/tubes, placed along the focal axis of the trough. The heat transfer
fluid is heated up to about 400°C by the concentrat ed solar rays. The steam or gas is
converted to electric energy in a conventional steam turbine generator. The relatively low
temperatures achieved in parabolic trough CSP are the cause for lower efficiency of this
technology. The solar-to-electric efficiency is about 14%. Parabolic trough CSP can be
implemented with thermal storage.
C.9.1.2. Central Receiver CSP
The first central receiver CSPs went on line towards the end of 2006 (Cameron & Jones
2007: 170). Previous plants were implemented mostly as demonstration and research
projects. A circular array of large individually tracking mirrors (heliostats) concentrates
sunlight on to a central receiver, mounted on top of a tower. A heat transfer medium in the
receiver absorbs the concentrated radiation, heating up to temperatures in the range of
500°C to 1,000°C. The higher temperatures in a cent ral receiver CSP are responsible for
Page 192 of 222

increased efficiency of this type of technology. The conversion from solar radiation to
electricity (electrical conversion efficiency) is in the range of 20% for central receiver
systems. Thermal storage in the form of molten salts, liquid sodium or thermal oils is utilised
in central America receiver CSP, allowing electricity production for hours after sunset.
C.9.1.3. Parabolic Dish
This is the least commercially mature solar technology with prototypes used only in research
and demonstration. A parabolic dish-shaped reflector (very similar to a satellite dish)
concentrates sunlight on to a receiver, situated at the focal point of the dish. A fluid or gas in
the receiver absorbs the concentrated radiation, heating up to 750°C or even to more than
1,000°C, and is then used to generate electricity i n a small piston, sterling engine or micro-
turbine, attached to the receiver. The dish makes use of dual-axis tracking control. Electrical
conversion efficiency of parabolic dish is in the range of 30% to 40%. Parabolic dish systems
are modular with a single dish producing 25kW of electricity.
C.9.1.4. Thermal Storage
Thermal storage offers a number of benefits to CSP, even though it has not been used
extensively in CSPs to date. Thermal storage:
o Can shift the output of CSP to several hours after sunset. Continued generation for
three to eight hours will in most cases coincide with the evening peak.
o Increases the dispatchability of the power plant to cover peak demands.
o Increases the solar capacity factor of the plant, which is the percentage of time the
plant is operating (at full load) on solar.
The annual solar capacity factor in systems without thermal storage is about 25%. The solar
capacity factor can be increased to over 50% if the solar field (i.e. the reflective/mirrored
area) is increased and thermal storage is added. Recent developments offer thermal storage
of 6 to 7.5 hours, eliminating the need for fossil-fuel back up (European Commission 2007:
14 to 16) while systems with longer storage are being planned (Cameron & Jones 2007:
170).
C.9.1.5. CSP Technology for this Study
The study models the parabolic trough CSP with thermal storage. The technicalities and the
costing will be closely based on the currently implemented Andasol 50MW plant with 7.5
hours of full output after sun-down (European Commission 2007: 14 to 16).
C.9.2. Resource
Regions where the annual irradiance level is 1,700kWh per square metre or higher are
considered suitable for CSP. The best suited sites, however, are those in excess of 2,700
kWh per square metre of annual solar radiation which translates to an average daily
irradiance of 7.4kWh/m2/day. Based on the average solar radiation data in Namibia as shown
in Figure C-15, all regions in Namibia are suitable for CSP operation, except the narrow strip
along the Swakopmund to Walvis Bay coastline.
Page 193 of 222

Figure C-15: Map of average daily irradiation levels for Namibia

A concentrating solar power plant requires approximately an area of 4 hectares per MW,
based on the current Andasol design (European Commission 2007: 14 to 16). The
generation potential for CSP in Namibia is therefore many times more than the overall
demand. Even the generation potential along existing transmission lines exceeds Namibia’s
energy demand manifold.
The CSP option is therefore not limited by the resource but by access to the grid (which is
expandable) and by technical grid integration and energy mix limitations.
A CSP plant would make contributions to the day-time energy needs of Namibia with an
approximate coincidence factor of about 92% (330 clear days per annum). Furthermore, the
associated thermal storage of 7 hours would be dispatchable during the evening peak.
Source: National Renewable Energy Laboratory.
Page 194 of 222

No significant air emissions occur from power generation using CSP technology (Stoddard,
Abiecunas, O'Connell April 2006). Using the natural gas combined cycle plant – the cleanest,
most efficient fossil technology – as a proxy, data for criteria air emissions reductions were
developed. By way of example, for a 4,000 MW deployment scenario, at least 300 tons per
year of NOx and 7.6 million tons per year of CO2 would be avoided. If the fossil displacement
is simple cycle gas turbines or coal fired plants, these values would be larger (Stoddard,
Abiecunas, O'Connell April 2006).
While CSP plants may have environmental benefits due to emissions reductions, they do
require significant land area. A 100 MW CSP plant is estimated to cover approximately 800
acres (comprised mostly of the solar field) while a 500 MW combined cycle plant would
occupy about 20 acres (Stoddard, Abiecunas, O'Connell April 2006).
Black & Veatch (2006) calculated the total land area in California with sufficient resource
to support power generation on comparably flat land outside of environmentally sensitive
areas by using solar insolation data provided by NREL. Table C-9 shows available land
with high solar resource and land slope not greater than 1 percent, a preference for
trough and power tower plants. The land area for each technology type, along with
potential generation capacity in MW and GWh, is presented in Table C-9. Capacity and
generation were based on CSP systems without thermal storage…The total generation
capacity as of 2004 for the state was roughly 58,000 MW (Stoddard, Abiecunas, O'Connell
April 2006).
Solar Resource Land Capacity Potential, MW Generation Potential,

Area, mi² GWh
Parabolic Trough, no 5,900 661,000 1,614,000

storage < 1 % slope
Parabolic Trough, six 5,900 471,000 1,640,000

hours storage < 1 % slope
Power Tower, six hours 5,900 342,000 1,233,000

storage < 1 % slope
Parabolic Dish, < 3 % 11,600 1,480,000 3,371,000

slope
Parabolic Dish, < 5 % 14,400 1,837,000 4,196,000

slope
Concentrating PV, < 3 % 11,600 1,235,000 2,859,000

slope
Concentrating PV, < 5 14,400 1,534,000 3,558,000

% slope
Table C-9: Concentrating Solar Power Technical Potential
Solel (2007) offers the following reasons to support CSP:
Page 195 of 222

• Unlike conventional power stations the CSP stations receive their energy through
concentrated solar radiation, rather than fossil fuels and as such emit no air pollutants
in producing the electricity;
• In sunny, arid locations, one square kilometre of land can generate as much as 100
gigawatts hours (GWh) of electricity per year, enough to power 50,000 households;
• CSP plants reduce air pollution, i.e. for each square meter of CSP concentrator
surface is enough to reduce annual consumption of 200 to 300 kilograms (kg) of
CO2;
• The 'energy payback' is five months which compares well with their useful life of 30 to
40 years;
• Reliable and available when needed most, i.e. during peak demand;
• CSP can be built in their entirety within a few years and can follow the demand more
closely than most conventional power projects;
• While all power plants require land and have an environmental impact, the best
locations for solar power plants are on land, such as deserts, for which there might be
few other uses;
• CSP can be designed for solar-only or for hybrid operation to provide all day power;
• CSP are labour intensive and generally create more jobs per investment than
conventional power stations;
• CSP are primarily indigenous resources with no need to import energy sources;
• Electricity produced from CSP plants is a fixed-cost generation resource generally
sold through long term power purchase agreements in which the cost to the customer
is known in advance;
• CSP can be cheaper than power from fossil fuels when all cost externalities are
considered.
Page 196 of 222

Employment possibilities Minor possibilities for the local population due to low employment
numbers for this technology.
Poverty alleviation Little likely effect due to likely remote location in unpopulated area.
Light, Cooking, Refrigeration, Hot Water Little likely effect due to likely remote location in unpopulated area.
Health and safety Little likely effect due to likely remote location in unpopulated area.
HIV /AIDS Little likely effect due to likely remote location in unpopulated area.
Hospitals and clinics, e.g. refrigeration of Little likely effect due to likely remote location in unpopulated area.
medication
Schools Little likely effect due to likely remote location in unpopulated area.
Accidental risks, emergency situations Insignificant accident risks.

and response;
Construction camps and power line Little likely effect due to likely remote location in unpopulated area.
Damage to cultural and natural assets Little likely effect due to likely remote location in unpopulated area.
Human health impacts, e.g. Little likely effect due to likely remote location in unpopulated area
electromagnetic fields from power lines and no emissions.

Effects on tourism industry Little likely effect due to likely remote location in unpopulated area.
Visual impacts and Aesthetics Major effect on site, but little likely impact on humans due to likely
remote location in unpopulated area.
Shadow disturbances Major effect on site, but little likely impact on humans due to likely
Table C-10: Social Impact Checklist – Solar Concentrating Plant in Namib Desert

The cost estimates for CSP are based on the actual cost of the Andasol parabolic trough
CSP currently being implemented in Spain (European Commission 2007: 16). The costs are
Euro 5.2 million per MW with the operating costs being estimated at approximately USD 1.2
c/kWh (World Bank 1999).
Page 197 of 222

calculated at NAD 10.4c/kWh.
C.9.6. Risks
The risks are listed below and presented in the Figure C-16:
• The risks associated with the technology is predominantly in the area of performance
as there has been little development in the last twenty years other than the
performance monitoring of the existing plants (Cameron & Jones 2007: 168 and
Jones 2007). This risk however only applies to the solar thermal part of the plant as
the gas turbine part is well known technology.
• A further risk may be the lead times as there is currently substantial demand in this
sector.
• Furthermore, extreme weather conditions could have a damaging impact on this
plant.
Severe
Technology may not

I work as intended or not
be as reliable as expected
Moderate
M
P
A
C Implementation lead
time may be extended due
T to international
demand
Minor
Low Medium High

LIKELIHOOD
Figure C-16: Risk chart for concentrating solar power plant
Page 198 of 222

C.10. SOLAR PHOTOVOLTAIC POWER PLANTS

Photovoltaic (PV) technology was specifically developed for powering satellites and space
stations. The first PV modules developed for this application date back to the 1950’s. Since
then the technology has matured significantly through improved efficiency, life expectancy
and lowering of costs.
Today PV is used pre-dominantly in grid-connected systems in parts of the world where

incentives are provided for power generation through renewable energy. Off-grid applications
in regions with good solar resources are often served through PV since grid extensions or
diesel generator systems are too costly.
To date the total world-wide installed PV capacity is approximately 7.7GWpeak (Renewables

2007). The annual production for PV in 2006 was 2.5 GW peak (Bradford Maycock 2007: 62).

PV modules convert sunlight into DC electricity through what is referred to as the
“photovoltaic” effect.
The focus of this supply option is on grid-connected PV plants which that consist of PV
arrays and grid interactive DC to AC inverters that synchronise to the grid and thus feed
power into the grid. PV systems in conjunction with battery storage are generally used for
stand-alone applications and are not considered here due to the high cost of the storage and
the high maintenance requirements.
PV plants operate autonomously and have very low maintenance requirements. Service
intervals range between bi-annual and annual visits.
PV technology is highly modular as modules can be strung into array sizes from a few Watts
to a few MW. Some of the larger PV plants have reached sizes of 5MW peak. The modularity of
PV allows a high degree of distribution and embedding into low and medium voltage
networks. Due to the autonomous operation and low maintenance requirements the added
costs for operating distributed systems are insignificant.
PV plants are mostly installed on the ground or on roof tops (assuming correct orientation).
Installation is simple, fairly rapid (delivery time being the major time component) and requires
no specialised equipment or plant in terms of civil works and machinery. The grid
interconnection can be at low (230/400V) or medium voltage (11, 22 or 33kV). Namibia has
the local capacity to install, operate and maintain large grid connected PV plants.
C.10.2. Resource
Namibia is endowed with an abundant solar resource and has amongst the highest
irradiation levels in the world. The average irradiation levels in the centre of Namibia are in
the region of 5.8 to 6.2 kWh/m2/day as shown in Figure C-15. For the purpose of this study it
will be assumed that the annual average irradiation level for a PV plant is 6 kWh/m2/day
(Bicon Namibia 1999).
The seasonal variation of the solar resource on a flat surface is shown in Figure C-17. The
amplitude of the “cosine” shape of the curve is reduced through the angle of tilt. Since a PV
plant would essentially contribute to day time base load one would choose the angle of tilt for
Page 199 of 222

an optimal summer and winter contribution. As a matter of interest: It would be conceivable

that a PV plant targets particular loads within the overall Namibian demand profile. This could
for example be air-conditioning loads which would be naturally “synchronised” to the solar
irradiation levels of the day. This would result in a smaller angle of tilt to harness more solar
energy during summer.
10
9
Irradiation [kWh/m2day]
5
4
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month
Keetmanshoop (WMO) Ondangw a Windhoek Average
Figure C-17: Seasonal variation of the solar resource
Namibia has approximately 330 days of full sunshine per annum providing one of the most
consistently and predictable renewable resources in Namibia (Mendelsohn 2002).

The area needed to generate 1MW of power at noon can be calculated with the following
assumptions:
1. The overall PV plant efficiency is 9% (12.5% PV array efficiency, 80% output of
nominal standard test conditions, 95% grid inverter efficiency, 5% transformer and
line losses) and
2. The extra space required to avoid shading is 50%38 between rows of PV modules at
an angle of tilt of 30°.
3. The incident solar irradiance is 1kW/m2 on the tilted surface.
The resulting area required to generate 1MW output from a PV plant is 16,000m2 or 1.6
hectares39 while the overall PV array size is 1.4MW peak. A PV plant with a 50MW output
therefore requires an area of about 80hectares and a PV array size of 69MW peak. This is
insignificant in terms of Namibia’s size and can be located in the vicinity of the existing grid or
near a substation such as Kokerboom.
38
In order to avoid shading and assuming that the lowest position of the sun at noon is 50° from the ho rizontal, then the
additional space needed between rows of PV modules which run on east/west axis is tan (30°)/tan (50°) = 48%.
39
Calculated as 1MW / (1,000 x 9% / (1 + (48% x cos (30°)))) = 15,772 ~ 16,000m 2.
Page 200 of 222

The generation potential of solar PV in Namibia is therefore not limited by the resource or the
space but rather by the amount of PV power that the Namibian grid can accommodate in
terms of network stability and how it fits into the overall generation mix.
For the purpose of this study it will be assumed that the maximum output from potential PV
plants can reach a maximum of 40MW (10% of peak demand or 12% of daytime peak). RE
contributions in European networks have exceeded these values and are therefore a safe
assumption.
The actual PV capacity required to generate 40MW output during the midday period is
55MW peak. The average daily power generation curve is presented in Figure C-18
50
PV plant output [MW]
40
30
20
10
0
0 2 4 6 8 10 12 14 16 18 20 22 24
Time [hours]
Figure C-18: Average daily power generation profile of a 55MWpeak plant
Based on an average annual irradiance level of 6kWh/m2/day a 55MW peak plant will generate:
o 240MWh per day
o or 87GWh per annum.
The availability of the plant is directly coupled to the generation profile as shown in Figure
C-18.
The probability factor that the plant is able to deliver is based on the ratio of sunny days
versus the total number of days, i.e. 330 over 365, which results in a probability factor of
92%. Should the plant(s) be located in the southern part of Namibia then this factor will be
higher due to the low rain fall and consequently fewer days of inclement weather.

The environmental impact of a photovoltaic system is minimal. It is a renewable energy
based generation technology with zero emissions and no water requirements, generating no
by-products. Like all other generating options, energy is needed to manufacture the
generation plant, in this case photovoltaic cells, glass front and frame. A PV module
generates a multiple of the energy used for its manufacture over its lifetime. It has been
shown that the energy payback of a crystalline module takes place within a 2 to 3 years
period whereas a thin-film module requires about 2 years (NREL 2004).
PV modules are either installed as roof-mounted or roof integrated arrays or as ground

mounted arrays. In the latter case the area requirements are approximately 1.5 hectare (or
Page 201 of 222

15,000m2) per MW40. Considering the large areas of desert land in Namibia, the land-usage
is not a critical issue.
Because solar modules (panels of PV cells) have useful lives of up to 30 years, the amount
of waste generated by retired modules is currently very small. By about 2020, however, this
growing industry will produce a growing PV waste stream. PV products are generally safe for
landfills, because PV materials are usually encased in glass or plastic, and many are
insoluble. Some modules, however, could be classified as hazardous waste, a situation that
is prompting the PV industry to develop recycling processes for modules (USDOE Energy
Efficiency and Renewable Energy 2006).
Because PV systems are widely dispersed, and because each system has relatively small
amounts of semiconductor material per cell, recycling PV will be a challenging task.
However, the industry is looking ahead and preparing for this challenge. The PV industry
hopes to learn from new processes for recycling batteries and electronic devices such as
computers.
40
This is based on an module efficiency of 10%, an irradiance level of 1,000W/m2 and 50% extra area to allow for an
appropriate angle of tilt for Namibian conditions.
Page 202 of 222

Poverty alleviation Little likely effect due to likely remote location in unpopulated area.
Light, Cooking, Refrigeration, Hot Water Little likely effect due to likely remote location in unpopulated area.
Health and safety Little likely effect due to likely remote location in unpopulated area.
HIV /AIDS Little likely effect due to likely remote location in unpopulated area.
Hospitals and clinics, e.g. refrigeration of Little likely effect due to likely remote location in unpopulated area.
medication
Schools Little likely effect due to likely remote location in unpopulated area.
Accidental risks, emergency situations Insignificant accident risks.

and response;
Construction camps and power line Little likely effect due to likely remote location in unpopulated area.
Damage to cultural and natural assets Little likely effect due to likely remote location in unpopulated area.
Human health impacts, e.g. Little likely effect due to likely remote location in unpopulated area
electromagnetic fields from power lines and no emissions.

Effects on tourism industry Little likely effect due to likely remote location in unpopulated area.
Visual impacts and Aesthetics Major effect on site, but little likely impact on humans due to likely
Table C-11: Social Impact Checklist – Solar PV Plant in Namib Desert

The dominant costs of a solar PV power plant are the upfront capital costs, while the
operating costs are minor in comparison. The technology has very low maintenance
requirements, much of which is a monitoring exercise.
The major cost component of solar PV power plants resides in the photovoltaic modules.
Large PV plants (100’s of kW or MW size) have a fixed and a variable cost component:
Page 203 of 222

The variable cost components include the PV modules and the grid inverters. The fixed cost
components include the control building, transformer station, grid extension, communication
and security.
Embedded PV power plants (kW to tens of kW size) only have the variable cost component
and are often integrated into existing buildings that have existing grid connections.
The current investment costs are calculated as NAD 33 million per MW for plants in excess
of 5MW peak. This is based on the current price for PV at USD 3.75 per Wpeak (Bradford and
Maycock 2007: 65).
modelled at NAD 2c/kWh.
C.10.6. Risks
The risks associated with solar PV are minor:
• Quality is a concern and the use of reputable manufacturer’s are essential in order to
ensure that long-term performance is guaranteed.
• Extreme weather conditions.
Severe
Extreme weather
I
Moderate
M
P
A
C
Long term
T performance
Minor
Low Medium High

LIKELIHOOD
Figure C-19: Risk chart for solar PV power plant
Page 204 of 222

C.11. WIND ENERGY POWER GENERATION

The first electrical wind turbines were built before the end of the 19th century. During the early
20th century large-scale wind turbines (multi-kW) were manufactured. However, the rural
electrification programmes in USA and Europe killed off the market for wind turbines. During
the 70’s developments started again and to date this technology has been refined
significantly and has reached a mature technology stage.
To date global installed capacity is about 95GW. The recent growth in this sector has
therefore been significant (Renewables 2007).

The most common utility scale wind energy converters (WEC) are horizontal axis, up-wind
generators, which are based on a three blade design. The WEC, which have been
manufactured in the last ten years are pre-dominantly based on pitch-control which allows for
more refined wind generator control and allows for higher wind power exploitation. The
described WECs achieve typical efficiencies of up to 45%.
The grid integration of modern WECs is through power electronic converters which
synchronise to the grid and feed-in the generated power. Similar to grid connected solar PV,
the grid voltage needs to be present to allow for power generation. If the grid fails then the
wind generator goes off-line.
Wind generators have an operational life time of about 20 years (Morthorst ?2004). During
this time the units require professional regular maintenance once to twice per annum.
Integration of wind power does require additional control equipment (regulating power,
spinning reserves) due to the variability of wind and since wind power virtually does not
contribute to primary frequency control. The cost of such control equipment has been
estimated at 15% of the capital costs and forms part of the assessment.
Wind energy generation is a modular technology and wind farms constitute a number of wind
turbines, which are connected together to the grid feeder line via a substation. Although wind
generation is modular it is essential to implement wind power in clusters of wind generators,
i.e. this is not a technology which is suitable for distributed generation such as solar PV and
biomass to electricity plants. This is mainly due to the EIA requirements, civil works
(foundations), grid interconnection, maintenance and supervision of operating plant. On the
other hand locating wind farms of tens of MW at different wind sites has the advantage of
evening out fluctuations in generation.
One of the key advantages of WECs is the short lead time. Wind farms can be implemented
within a two year period, assuming that the resources have been studied and the
permissions have been granted. This would allow for a one year project development period
and a one year construction period.
C.11.2. Resource
The wind resources along the Namibian coast are substantial in particular near Lüderitz and
along parts of the Skeleton Coast (Bicon Namibia 1999). The wind speeds in these areas are
strongest in summer months. The most promising sites where more detailed measurements
Page 205 of 222

have been conducted are Lüderitz and Walvis Bay. Focus has also been on these sites due
to existing civil and grid infrastructure.
Annual average wind speeds at Lüderitz are in the region of 7.5m/s (DECON 1999) to
8.36m/s (Jargstorf 2004) and in the region of 6.5m/s (Jargstorf 2004) to 6.8m/s (DECON
1999) at Walvis Bay, transposed to a 50m hub height.

The wind generation potential exceeds Namibia’s demand and is limited only in terms of the
grid capacity and the intermittent nature of the power which therefore has practical/technical
grid integration limits. The capacity factors for Lüderitz and Walvis Bay at 50m hub-height are
quoted as 35% and 21% (Jargstorf 2004) respectively which is based on one hour averages.
One hour averages usually lead to an underestimation. This study assumes an average
capacity factor for the overall wind energy contribution in Namibia at 31%.
With the current grid infrastructure to Lüderitz the maximum size of a wind park located in
Lüderitz is 30MW. Since the wind resources are substantially better in Lüderitz than in Walvis
Bay it is assumed that any wind park development would commence in Lüderitz. However,
beyond the 30MW capacity further wind parks would be developed in Walvis Bay and then in
Oranjemund.
The daily generation profile is largely based on the wind regimes in Lüderitz with wind
generation peaking at 14:00 and with the lowest output at 06:00. The stronger wind season is
taken to be from October to March while the lower wind season is between April and
September, i.e. the seasonal generation profile is inversely related to the main demand
periods in winter41.
For the purposes of this study it is assumed that maximum installed capacity from wind is
90MW, although in most scenarios it does not exceed 60MW. This is primarily motivated by
Namibia’s lack of storage capacity through which the intermittency of wind could have been
mitigated and grid reliability ensured as well as keeping wind to less than 20% of the overall
capacity mix due to limited international experience with higher wind proportions.
Furthermore, Namibia currently imports electricity from Eskom and has to provide an hourly
schedule of the required imports 24hours in advance. This fairly long “gate closure” time
makes integration of wind challenging since accurate forecasting of wind generation over a
24hour period is problematic whereas three hour forecasting with the appropriate weather
stations and software tools has reached high levels of accuracy.

Wind is a clean fuel; wind farms produce no air or water pollution because no fuel is burned
(USDOE Energy Information Administration May 2007). Wind power does not emit air
emissions such as SOx, NOx, and CO2. A major advantage of wind power is that it is
considered a clean energy form when compared to more conventional forms of electricity
generation namely coal and natural gas (Hohmeyer, Wetzig, and Mora 2007?). Major
negative impacts include:
• the aerodynamic noise and visual aesthetics along the landscape from the turbines,
41
Data based on Elisabeth Bay, 7m AGL
Page 206 of 222

• civil works during construction may have negative environmental impacts if not
implemented with an Environmental Management Plan.
• wildlife interactions, namely birds, and
• electromagnetic interference on radio waves and microwaves used for
communication if in closer vicinity of communication posts (Hohmeyer, Wetzig, and
Mora 2007?).
As with any new development general awareness and introduction to this form of energy will
be necessary by the proponents of this supply option to the surrounding communities.
If possible, to avoid potential public uneasiness, the wind farm could be sited in more remote
and isolated areas away from human settlements. However, this will drive up the overall
price due to the costs associated with connecting this power to the grid or some form of
distribution system.
The most likely place for wind farms in Namibia is Lüderitz and Walvis Bay (DECON 1999),
both of which are coastal towns. In Lüderitz, Grosse Bucht is the favourite site for
implementing a wind farm. This is located about 10km outside of Lüderitz within the diamond
area and will therefore not have a major impact in terms of aesthetics. Similarly, a wind farm
could be constructed at Paaltjies near Walvis Bay with similarly low impact on the human
environment.
Page 207 of 222

Poverty alleviation No likely effect due to location near urban area.
Health and safety No likely effect due to location near urban area.
Hospitals and clinics, e.g. refrigeration of No likely effect due to location near urban area.
medication
Schools No likely effect due to location near urban area.
Accidental risks, emergency situations Insignificant accident risks due to likely location away from immediate
and response; populated centres.
Damage to cultural and natural assets Possible significant effect on natural assets due to environmental
sensitivity of possible sites. Little likely effect on cultural assets.
Human health impacts, e.g. No effect due to location outside of urban area and due to no
electromagnetic fields from power lines emissions.
Resettlement (if applicable) and the Limited likely effect due to likely location in unpopulated area.
Effects on Aviation No likely effect but depends on micro location.
Effects on tourism industry No likely effect due to likely location in unpopulated area but depends
on micro location.
Visual impacts and Aesthetics Likely effect depending on location.
Table C-12: Social Impact Checklist – Wind energy plants near Lüderitz and Walvis Bay.

The capital cost of wind energy is assumed to be NAD 10.5 million per MW of installed
capacity (Morthorst ?2004: 99). This is inclusive of hardware, civil works, labour, design,
engineering and project management. The interest during construction is calculated over a
one- year period and the decommissioning cost is estimated at 3% of the capital cost.
Page 208 of 222

Operation and maintenance costs are estimated at an average of NAD 6 c/kWh (Morthorst
2004: 100). This is relatively high for the first few years of operation while the WECs are still
relatively new. However, the figure represents a reasonable average over the twenty year life
expectancy of the units.
modelled at NAD 6c/kWh.
C.11.6. Risks
The risks associated with wind energy are low. The risks are depicted in Figure C-20:
• concerns about the resilience of WECs under harsh coastal conditions,
• negative impact on the financial performance of the wind park due to overestimation
of resource,
• climate change and lowering of wind regimes,
• grid integration and stability more challenging then anticipated.
Severe
I
Moderate
M Wind plant life shortened

in harsh Namib climate
P
A
C Wind resource
T overestimated
Grid stability issues
worse than expected
Minor
Grid integration cost

higher than budgeted
Low Medium High

LIKELIHOOD
Figure C-20: Risk chart for wind generation power plant
Page 209 of 222

C.12. INTEGRATED SOLAR COMBINED CYCLE PLANT

Integrated Solar Combined Cycle (ISCC) gas turbine power plants are essentially based on a
conventional combined cycle gas plant with an added “solar boost”. The combined cycle
natural gas systems are popular in areas where natural gas is available. A combined cycle
plant uses gas combustion turbines as the first stage in electricity generation. The hot flue
gases from the combustion pass trough a heat exchanger to generate steam, which drives a
steam turbine at the second stage in the electricity production process. Concentrating solar
power can be integrated in the second stage of the combined cycle plant. The solar heat
increases the capacity of the steam turbine by either generating additional steam in a heat
recovery steam generator or by generating low pressure steam to be injected directly into the
steam turbine. The conventional combined cycle systems have heat-to-electricity efficiencies
of about 55% (Enermodal Engineering Limited and Marbek Resource Consultants Ltd 1999).
The add-on peak output of the solar system accounts for 20% to 30% of the combined cycle
output. This means that the solar system can boost the output of a 100MW combined cycle
plant to 130MW.
The ISCC plants’ main advantage is that the peak capacity can be increased at a lower
capital cost and the solar power boost is available during daytime activity.
Currently at least 1 GW of total ISCC plants are being considered in a range of projects in
different countries (Cameron & Jones 2007: 170).
ISCC plants integrate a small concentrating solar component and therefore low incremental
cost and risk. Since the main power plant is based on well know CCGT technology this is
considered a mature technology.
The resources, generation potential and environmental aspects have been described in
sections 3.1.2.1.3 and 3.1.2.1.6. The cost of ISCC plants are described below.

Please refer to the discussion on environmental and social aspects in section C.6.4 and
C.9.4.

The capital costs of an ISCC power plant are estimated at USD 0.9 million per MW (1999).
This is based on estimates from the World Bank (1999).
The operating costs are estimated at USD 0.7 c/kWh (1997) which is based on a
combination of the costs between a CCGT and a CSP plant as per section C.6.3 and C.9.3.
The fuel costs of USD 3.6 c/kWh (1997) are similarly based on a combination of a CCGT
plant and the contribution from the CSP component. It is assumed that the solar plant
operates for a third of the day and contributes 25% of the energy during that time.
C.12.3. Risks
The risks are shown in the risk chart in Figure C-21:
Page 210 of 222

• The main risk for the ISCC is the financial exposure to foreign exchange risks,
assuming that the contract will be signed in foreign currency. The risk is however
already slightly mitigated through the CSP component, which uses local renewable
resources.
• A severe risk is the possibility of the gas reserves being lower than anticipated.
• Lead times are extended due to international demand in power stations.
Severe
Gas reserve /
gas extraction issues
Gas Price
forex risks
Large station -
surplus export price
I and sales risk
Moderate
M
P
Lead time
A extended due to
C high demand for
power stations
T
Minor
Low Medium High

LIKELIHOOD
Figure C-21: Risk chart for integrated solar combined cycle power plants
Page 211 of 222

Appendix D. Detailed discussion of Demand

Side Management Options
This section describes a range of demand side options with their technical details, their
characteristics and their financial implications.
Options considered
The options considered here rely on the ECB DSM study (ECB 2006) and enumerates and
ranks possible options while developing those options considered most appropriate in more
detail.
Since this present study concerns itself with major grid electricity supply issues, only those
DSM options which offer a major impact on grid electricity consumption and which can also
be quantified with a high degree of confidence have been included in this study. This limits
the options included to three, being:
• Large-scale compact fluorescent light roll out to residential consumers,
• Wide-spread implementation of ripple control for electric and electric assisted solar water
heaters,
• Large-scale replacement of electric water heaters with solar water heaters with electric
back up.
The first and last of these are energy efficiency measures with a significant impact on the
national load profile because they are aimed at residential consumers with a rather specific
pattern of using both lights and hot water. The ripple control project is a pure DSM measure
with no energy savings, which only shifts demand from one period to another. Taken
together and implemented fully these three measures have the potential to reduce peak
demand by as much as 20%.
There are many other possible DSM measures that can be undertaken in Namibia in
particular related to behavioural changes:
• Every individual has a role to play in conserving electricity and energy. Social
patterns and the behaviour of individuals, families, companies and the government
itself will impact the success of the any energy efficiency and conservation
programme. It will take a concerted effort to increase the general awareness of
people to change their behaviour on how they use electricity. Electricity can be saved
by changing one's actions. Examples of this might include making people aware of
shutting off any unnecessary lights, shutting off computers when not in use like over
evenings and weekends, purchasing more energy-efficient appliances, and shutting
off all appliances as opposed to putting them into 'stand-by' as this continues to draw
power from the units.
• To support the individual in better understanding their role in saving electricity and
how to conserve energy the Namibian government as well as NamPower could go a
long way in a training and awareness campaign for the public. This could include
understanding the basics of electricity, how to use electricity more wisely, why it is
important to save electricity in the long-run, and for those receiving electricity for the
first time, the safety aspects of using electricity. This 'training' could take the form of
Page 212 of 222

advertisements on radio, television and newspapers and even open workshops

throughout the country. Lastly, general awareness on energy conservation and
energy efficiency must be conducted on an on-going basis. That is, it should not be
assumed that once people are introduced to ways and means of saving energy that
they will continue to do so. A concerted effort will be required not only by government
and NamPower, but by individuals and businesses alike.
• It should not be assumed that one energy efficiency or demand side management
option 'fits all.' That is, there will be differences in individual and family incomes,
regional consumption patterns and demographics that will have to be taken into
account. Namibia will need to prioritize where and when it can it can anticipate the
greatest savings by the consumer. However, individuals can begin at home,
businesses can improve their overall image by taking energy (and thus sustainability)
seriously, and government can lead by example by taking steps to lower their overall
consumption of power.
• The public can feel like they are doing something for the environment by supporting
and participating energy efficiency investments. By reducing one's energy use, the
public can help reduce air emissions, air pollution, water consumed, and the amount
of natural resources extracted.
The fact that these types of interventions are not included in the modelling of this study does
by no means imply that they are not worth pursuing. They have been excluded from this
study primarily because they are difficult to quantify and even the DSM study only produced
rough estimates of what they may cost and what they may achieve. Such vague numbers
were considered inappropriate for inclusion in the modelling for this study, and developing
any of them to a more concrete level is outside the scope of this study.
Energy efficiency benefits
Efficiency and conservation is key to energy sustainability which focuses on long-term

energy strategies and policies that ensure adequate energy to meet today's needs as well as
tomorrow's needs.
One effective way of reducing environmental impacts from the current production of energy is
to increase energy conservation. By using energy more efficiently, both costs and emissions
are lowered; therefore, a significant incentive exists to improve efficiency in manufacturing
processes, the construction of new buildings, the production of new appliances, and for many
other energy uses. New housing construction standards, 'green building' codes as well as
city and municipality by-laws can assist in moving the planned housing market in the 'right'
direction. Properly designed homes, office blocks and building which includes the appropriate
orientation of the structures can play a significant role in saving energy. Existing structures
will present different challenges as due consideration to saving energy may not have been
addressed.
There are economic, social and environmental benefits to be gained from improving energy
efficiency throughout the economy. Benefits from improving energy efficiency suggest a role
for government to develop instruments and mechanisms that create the right incentives for
firms and individuals to adopt energy saving improvements (Government of Victoria 2003).
Complex measures can raise transaction costs and deadweight losses and therefore have
an impact on efficiency. Industry and individuals should be motivated to take actions to
improve energy efficiency that are in their own best interests — that is, ‘no regrets’
measures.
Page 213 of 222

This policy objective implies that voluntary measures are likely to be optimal — regulation or
the introduction of market mechanisms are likely to be costly and unnecessary when the aim
of policy is to induce firms and individuals to act in their own best interests. However,
regulation can prove effective and welfare enhancing if it is well designed and sensibly
introduced.
The remainder of this section briefly examines the key features of the three DSM options
included in the modelling. The bulk of this information is taken from the ECB DSM project
reports (2006) which are available on the ECB website for additional reading.
D.1. COMPACT FLUORESCENT LAMP ROLL OUT

The distribution of compact fluorescent lamps (CFL) is a widely used measure of energy
efficiency with a high impact on networks with an evening peak. It can be rolled out in a short
timeframe, almost to the extent that it can be used as an emergency measure for unforeseen
power supply shortages (CFL have been implemented in areas in the RSA on this basis).
The programme proposed in the DSM study (and since implemented differently by
NamPower) recommends the purchase of around one million CFLs and distribution of these
to residential consumers in exchange for used incandescent bulbs.
D.1.1. Anticipated Impact

The implementation of this programme is expected to result in a maximum demand reduction
of approximately 20MW during peak demand times and an annual energy saving of 22 GWh
(ECB 2006).
It is assumed that this programme can be implemented over a two-year period.
D.1.2. Costs and Investment

The calculations in the DSM study are based on a total number of 106,770 grid connected
households. Assumptions have been made regarding the level of penetration for converting
from incandescent to compact fluorescent lamps at household level. This was guided by the
perceived accessibility within the various centres.
The total investment cost of this energy efficiency drive is estimated at NAD 13 million. This
translates to a cost of NAD 625,000 per MW (ECB 2006).
D.1.3. Environmental and Social Impacts

There are public health concerns in the handling and disposal of CFLs at the end of their
useful life. Breakages of the bulb will release mercury into the atmosphere and increase the
exposure of humans and the natural environment to this toxic element. If the national
government and/or NamPower intend to disseminate CFLs to the public than there will be a
responsibility by these parties to better inform people on the handling and disposal of CFLs.
One possible way to ensure the safe disposal of the bulbs is to propose a return-policy by
both the manufacturers of the bulbs and/or NamPower.
The replacement and installation of CFLs will reduce the overall energy demand, However,
again, a solid public outreach and awareness programme should follow the roll-out of CFLs
to ensure that people don't have the false sense of 'saving energy' while keeping the lights
Page 214 of 222

on continuously. Dissemination of CFLs must be done hand-in-hand with a public awareness

programme on electricity use.
D.1.4. Risks
The main risks of this programme are:
• Low quality CFLs exist and may enter the market after the initial dissemination drive.
This would lead to premature failure and discreditation of the programme. It is
therefore essential to ensure that standards are being met.
• The dissemination programme may find it difficult to reach a high percentage of
households, resulting either in lower penetration or higher dissemination costs. This
would lead to either a lowering of the expected savings or to cost increases.
• Users may revert to incandescent bulbs after some time meaning that the positive
impact of the conversion would slowly erode.
Figure D-22 shows a risk analysis around the CFL energy efficiency drive.
Severe
Implementation
Free dissemination
Sustainability
stirs political
Consumers revert back to
controversy
incandscent bulbs
I over time
Moderate
M Sustainability
P Environmental concerns
regarding disposal of
A Implementation CFLs
Reaching a high % Sustainability
C of households may be Low quality CFLs in market
problematic influence consumer
T perceptions
Economic
Minor
Trade in CFLs may

start and undermine
benefits
Low Medium High

LIKELIHOOD
Figure D-22: Risk chart for CFL energy efficiency measure
Page 215 of 222

D.1.5. Summary
The key information of this demand option is summarised in the table below and is used as
input to the technical and macro-economic modelling.
Parameter Value
DSM measure Compact Fluorescent Lights
Impact Peak reduction and energy savings
Anticipated savings 22 GWh per annum
20 MW peak reduction
Service life 6 years
Roll out timeframe 2 years
Investment per MW NAD 625,000 per MW (2007)
Cost of capital: 6%
Capitalisation period 6 years
Direct jobs during Number of skilled staff: 0 per MW

construction:
Number of unskilled staff: 0 per MW
Table D-13: Summary information on CFL measure
Page 216 of 222

D.2. SOLAR WATER HEATER ROLL OUT

Solar Water Heaters (SWH) are one of the Renewable Energy Technologies which are
competitive with electrical water heaters (EWH) in terms of the financial performance. In
2005 the breakeven between domestic SWH and domestic EWH was between 3 to 9 years
(Emcon Consulting Group 2005), depending on the tariff and on the metering. In low-income
households, approximately 40% of the electricity bill goes towards electrical water heating.
Besides the financial benefit to the users when converting from EWH to SWH, SWHs can
free up substantial network capacity if deployed widely. EWH have electrical element sizes
between 2 to 4kW. With an estimated 97,000 electrical water heaters in Namibia, the
electrical load is significant, especially during the evening peak. According to the Emcon
study an estimated 106MW are contributed to the evening peak through EWH (Emcon
Consulting Group 2005).
The SWH technology is mature and well proven.
In the modelling it is assumed that the SWH programme is rolled out in addition to the ripple
control programme. This is done because a) the ripple control can be rolled out much faster
than SWH and brings relief for the critical evening peak and b) SWH will most likely still have
electric back up elements, that one would want to ripple control to avoid excessive peaks in
adverse weather conditions.

Solar Water Heaters will have an impact on demand over the whole day but also particularly
during the evening peak as this is pre-dominantly a domestic appliance. The estimated
impact on the evening peak for the purposes of this study is approximately 80MW. Assuming
an average daily consumption of 8.5kWh per EWH an annual savings of 250GWh could be
achieved (ECB 2006).
It is assumed that this programme can be rolled-out over a ten-year period.

The investment cost is based on the average cost of one SWH costing NAD 15,000.
Assuming a 50% diversity factor a 2kW geyser will displace 1kW of maximum demand.
Therefore 1,000 geysers displace 1MW. The investment cost is therefore NAD 15 million per
MW (ECB 2006).
D.2.3. Environmental and Social Impacts

Solar water heaters (SWH) could play a role in reducing the overall CO2 emissions of
Namibia. By avoiding the use of electricity to heat water in conventional water heating units
(geezers), SWH could play an important role in reducing the country's CO2 contribution. In a
CDM study conducted by SouthSouthNorth in Kuyasa, Khayelitsha in Cape Town, South
Africa (November 2005), it was calculated that 2.85 CO2 tonnes could be avoided by the use
of SWH, CFLs and an insulated roof per household – see Table D-14 (SouthSouthNorth
November 2005). It should be noted that the solar collectors for this study were 1.6 m² with a
100 litre water storage tank. Greater savings could therefore be made with larger systems.
Page 217 of 222

Solar water heaters do not pollute. By investing in one, you will be avoiding carbon dioxide,
nitrogen oxides, sulphur dioxide, and the other air pollution and wastes created when your
utility generates power or you burn fuel to heat your household water. When a solar water
heater replaces an electric water heater, the electricity displaced over 20 years represents
more than 50 tons of avoided carbon dioxide emissions alone (Solar Energy International
2008).
Given the high levels of sunlight in a country like Namibia the use of solar water heaters
supports the sustainable development principle of using local natural resources to meet ones
needs rather than importing resources like coal (either directly or indirectly through electricity
bought from South Africa produced by coal). Secondly, once the SWH unit is paid for the
owner / user of the unit basically receives 'free' hot water as opposed to buying electricity
with ever-rising tariffs.
Emissions Baseline CO2 Emissions per annum:

Reduced and Electric geysers : 1.288 CO2 tonnes/hh/year
Avoided Incandescent light Bulbs: 0.294 CO2 tonnes/hh/year
Energy needed to reach a level of thermal comfort in an un-insulated house:
6.86 CO2 tonnes/hh/year
Project CO2 Emissions per annum:

Solar Water Heaters: 0 CO2 tonnes/hh/year
Compact Fluorescent Light Bulbs: 0.066 CO2 tonnes/hh/year
Energy needed to reach a level of thermal comfort in an insulated house:
5.53 CO2 tonnes/hh/year.
2.85 CO2 tonnes/hh/year avoided as a result of the project
Table D-14: CO2 Emissions Reduced and Avoided Kuyasa CDM Energy Upgrade Project
D.2.4. Risks
The major risks of this energy efficiency drive are:
• Poor quality SWH equipment and installation practise leading to a poor reputation
and low uptake of the technology.
• Availability of second-hand EWH on the local market as a result of retrofit activity.
Low-income households might be lured into adopting EWH for their homes as a result
of the low initial cost.
• Serious exchange rate fluctuations may drive up the costs of a SWH if they are being
imported and not produced in Namibia.
• There are challenges in terms of existing installation capacity in Namibia as well as to
build further capacity to support the implementation of such an extensive roll-out.
Figure D-23 shows the contemplated risks in a risk chart.
Page 218 of 222

Severe Implementation
Poor quality SWH equipment
and installations
I
Moderate
M Sustainability
P Economic
Technical skills
shortage
Interest rate
A increases for funded
C SWH Economic
Exchange rate
T fluctuations
Economic
RSA EWH
Minor
manufacturers dumping
Economic
EWH in Namibia
Economic Ulitity risk of
Market for second reduced kWh turnover
hand EWH
Low Medium High

LIKELIHOOD
Figure D-23: Risk chart for conversion to SWH
D.2.5. Summary
Parameter Value
DSM measure Solar Water Heater Conversion
Impact Energy savings and peak reduction
Anticipated savings 250 GWh per annum
80 MW peak reduction
Service life 20 years from installation
Roll-out timeframe 10 years
Investment per MW NAD 15 million per MW (2007)
Cost of capital: 6%
Direct jobs during Number of skilled staff: 3.33 per MW

construction:
Number of unskilled staff: 3.33 per MW
Table D-15: Summary information on SWH measure
Page 219 of 222

D.3. RIPPLE CONTROL

Ripple control of electric water heaters is a well-established means for utilities to defer
electricity consumption from peak to off-peak times. Most of Namibia’s larger towns have the
potential to beneficially implement ripple control on EWH (and SWH with electric back-up) to
control billing demand peaks, as well as national demand peaks and the timing of peak
consumption.
This technology is in use by the City of Windhoek and by the Erongo RED in the town of
Walvis Bay and is therefore well understood.

Ripple control is a peak shifting facility and does not save energy. Therefore its impact is only
on maximum demand. The time of impact is related to the use of the appliance. Since EWH
are most active during night (significant contribution to evening peak) as well as during the
morning hours, those are the periods that ripple control will impact on.
It is estimated that ripple control can provide at least an additional demand shifting capacity
of 24 MW over and above the existing installed ripple control capacity. This is based on the
practical experience of the City of Windhoek where an average drop of 1kW is achieved per
EWH switched off. The expected impact is therefore based on 1kW per EWH switched (ECB
2006).
It is assumed that this programme can be implemented over a four-year period in the major
centres of Namibia.

The estimated cost for ripple control is NAD 2.5 million per MW. This is based on an
estimated cost of NAD 1,000 per receiver (including installation) and costs of around NAD 7
million for a 66kV injection station and NAD 2.25 million for an 11kV injection station (ECB
2006).
For around 27,000 receivers, one 66kV injector and eleven 11kV injectors the total project
budget is expected to be in the region of NAD 60 million. Lower budget variants have been
investigated as part of the 2006 DSM study which leave out some of the higher cost areas
and implement only in areas where the capital cost per MW is significantly lower than the
NAD 2.5 million used here.
Page 220 of 222

D.3.3. Risks
The risks that affect this DSM feature most significantly are:
• NamPower and the REDs cannot reach agreement on implementation and operation
of this DSM option.
• Significant consumer resistance may prevent the installation of receivers in more
homes than anticipated, thereby reducing the benefits of the system.
• A parallel SWH implementation roll-out may lead to over-investment in ripple control
equipment since the back-up element of SWH can also be controlled through timers.
• Utilities are gaming with the ripple system, e.g. to improve their load profile vis-à-vis
NamPower without passing the benefits thereof on to consumers.
• Technology costs are escalating, thus jeopardising the continued viability of the ripple
system.
The risk chart in Figure D-24 provides an overview of the anticipated risks as a function of
their severity and likelihood.
Implementation
Utilities cannot agree
Severe
on implementation
Implementation
Utilities cannot agree
on operation
Implementation
Regulatory
Consumer resistance
I to installaton
Utilities use
Moderate
ripple for gaming

M
Economic
P TOU tariff interaction
Sustainability with ripple reduces
A SWH uptake faster benefits
C than expected reduces
benefits Economic
T Technology costs
escalate and erode
Implementation
benefits
Single Buyer implemen-
Minor
tation influences rollout

or operation
Low Medium High

LIKELIHOOD
Figure D-24: Risk chart for ripple control DSM measure
Page 221 of 222

D.3.4. Summary
Parameter Value
DSM measure Ripple control
Impact Peak demand shifting
Anticipated demand impact 24MW
Zero MWh
Service life 20 years
Roll-out timeframe 4 years
Investment per MW NAD 2.5 million per MW (2007)
Cost of capital: 6%
Direct jobs during Number of skilled staff: 1.06 per MW

construction:
Number of unskilled staff: 1.06 per MW
Table D-16: Summary information on ripple control measure
Page 222 of 222

Namibia Electricity Supply Demand Options March 2008 FINAL

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Namibia Electricity Supply Demand Options March 2008 FINAL

Uploaded by

Copyright:

Available Formats

Renewable

Electricity Supply and Demand

This study set out with three objectives.

• Continued reliance on imported electricity could result in blackouts equal to 10% of

• Demand side management is economically efficient. It has a role to play in avoiding

• Relying on imported electricity is the worst option. It is economically inefficient, results

• The most economically efficient mix is a balanced combination of a wide variety of

• The use of some renewable resources is economically efficient.

• Within a balanced generation mix:

o Renewables, other than hydro power, could supply up to 20% of demand at a

o When hydro power is included, renewables could supply up to 90% of

o Renewables create jobs and contribute to gross value added (GVA).

INTRODUCTION TO THE EXECUTIVE SUMMARY

Page iii of xii

• Baynes Hydro-Electric Power Generation

On the demand management side the options are:

• Compact Fluorescent Lamp roll out

Scenario 4: Maximum Supply Diversity. This scenario centred on the analysis of

Scenario 5: Generation Self Sufficiency. This scenario centres on generating as much

Scenario 7: Maximum Renewable Energy Options. This scenario centred on using as

1. Cost of the Generation Options

2. Cost benefit analysis:

All amounts are given in N$ million, 2007 prices

Table 0-1: NPVs, incremental costs, incremental benefits and BCRs

The following conclusions are drawn:

3. The key conclusions from an electricity price perspective are:

4. As far as contribution to GDP is concerned:

• The ‘renewables maximised’ scenario, at NAD 14.0bn, makes the greatest

Page vii of xii

5. From a social perspective:

For jobs the following conclusions are drawn:

The following conclusions are drawn on the degree of SMME promotion:

• Scenarios ‘generation self-sufficiency’ and ‘minimum energy imports’ scenario score

6. From a renewable energy perspective:

Page viii of xii

7. As far as emissions are concerned:

• However, there are additional environmental aspects associated with the

• The ‘all options considered’ scenario has a cumulative contribution to GDP of

THE POLICY IMPLICATIONS

There is a second conclusion that is almost as unequivocal although more muted.

Page xii of xii

Ralf Tobich, for peer reviewing the draft report.

Penda Hangala, for reviewing the Vision 2030 document.

This report was prepared by:

Emcon Consulting Group: technical analysis

Uli von Seydlitz

Economics Information Services: economic analysis

Geo cc, environmental analysis

Introduction to the Executive Summary .............................................................................. ii

The Challenges: .................................................................................................................iii

Analytical Results .............................................................................................................. vi

The Policy Implications....................................................................................................... x

Table of Contents .................................................................................................................. 4

List of Tables ......................................................................................................................... 9

List of Figures ...................................................................................................................... 11

2 Current Situation and Potential Outlook ........................................................................ 20

2.1 Electricity Usage.................................................................................................... 20

2.2 Electricity Supply ................................................................................................... 23

2.3 Demand and Supply .............................................................................................. 26

2.4 Environmental Issues ............................................................................................ 28

2.5 Potential Outlook ................................................................................................... 29

2.5.3 Environmental considerations......................................................................... 32

3 Description of Methodology .......................................................................................... 36