EnDev II

Ending Energy Poverty for All

Introduction: what is the problem really?
Bringing access to electricity for the majority of people living in Africa turns out difficult, slow, and
expensive; even though urban areas experience rapid progress, rural areas continue to lag far behind. This
paper suggests one way to deal with this so that the rural population will also start to enjoy at least some
of the benefits of having access to electricity.
Economic development in urban areas progresses at a much higher pace than in most rural areas, in part
most likely because access to electricity has been available for a long time. Important development efforts
providing urban public services to improve urban living conditions are beginning to pay off in many
countries. In turn, easy access to electricity, water, and broadband Internet increases economic
productivity and triggers further rural-urban migration. Until rural population densities, urbanization rates,
and/or incomes increase considerably, new access to electricity through grid extension is not likely going to
be cost-effective; unless substantial public support (i.e., subsidies) is provided, further grid expansion is not
likely to take place.
The fact is that the majority of Africans live in rural areas and are indispensable for supporting life in urban
areas: they are responsible for supplying urban areas with food and low-cost energy (wood-based). In
general they lack access to public services and modern infrastructure and live in energy poverty. Access
rates in urban areas are indeed 6 times higher than in rural areas, as illustrated in Table 1.
Table 1: Electricity access, 2008
Total Non electrified Urban Rural
population population electrification electrification
(million) rate rate
(million)
North Africa 164 2 (1%) 99.6% 98.2%
Sub-Saharan Africa 838 587 (72%) 57.5% 11.9%
Africa 1002 589 (60%) 66.8% 22.7%
Source: World Energy Outlook, 2009, IEA
There are several logical reasons that explain this inequity: low rural population density, low rural demand
for electricity, and low disposable rural incomes, each of which makes the supply of electricity in similar
ways as realized in urban areas more difficult and increases costs. As a result, there are now almost 600
million people deprived from access to modern energy, who basically live in the dark, in ways that are not
so different from their far ancestors. With the ongoing Decentralization and expanding rural electorates,
one would have expected modern energy access rates and public services even in rural areas to increase
fairly rapidly, but this has not been the case until now.
The situation is even worse since simply maintaining the status quo of 58% urban electrification access rate
in SSA in a rapidly growing urban environment will be a gigantic task in itself: to maintain a constant access
rate an additional 180 million urban persons would need to be covered between now and 2020. If over all
access rates in SSA were to increase from 30% to 50% by 2020, 350 million people will need to gain access
Draft

and 460 million if access were to increase to 60% by 2020, but many of these newly connected would come
from urban expansion. It therefore requires a major effort but does not contribute much to improving rural
conditions.
Unfortunately, access to modern and sustainable energy services and in particular grid electricity will by
and large remain limited to the more well-to-do part of the urban population. For those living in rural and
peri-urban areas, low disposable incomes and insufficient energy infrastructure will cause electricity supply
1

to remain absent, continuing to delay sustainable development there. Current efforts to improve living
Page

conditions for rural and peri-urban households are minimal, despite the fact that they make up far more

EnDev ending energy poverty ([email protected])

than half of the population of the continent. In short, it looks like the situation is not getting better any
time soon.
There is simply no justification for this inequity. In what follows possible ways are discussed to improve
modern energy access for the rural population. Could a different approach take a majority of Africans out
of energy poverty in a relatively short period of time?

There are limits to further grid expansion

Benefits of access to electricity are well understood and it is no miracle that many efforts are underway to
accelerate large-scale access. However, realizing this is much easier said than done: (i) urban electrification
is fairly straightforward as urban clients are relatively rich and live in a closely confined area, making it
simple and cheap to build a distribution network centered around a large and efficient generation plant; it
is quite understandable that commercial utilities focus on urban areas; (ii) rural customers generally live
farther apart and are financially often less well situated than their urban compatriots, making it more
difficult and expensive to connect to a distribution network; moreover, rural clients will show low unit
consumption rates, making expansion into rural areas even more expensive. Therefore utilities will rather
opt for further developing urban instead of rural areas (or smaller towns for that matter), simply for good
business reasons; (iii) peri-urban households generally live in low-standard houses that may not be suitable
at all for grid-electricity connection for safety reasons.
In summary, there are no natural incentives for utilities to quickly expand the electricity network beyond
profitable urban areas. Put differently, incentives are required for utilities to start covering areas outside of
their normal territory: subsidies.

There are no comparable decentralized solutions

Decentralized solutions are pursued by governments and donors outside of grid areas, either on an
individual basis (i.e., for one beneficiary at the same time) or for a distinct group of beneficiaries (such as
an isolated village). PV solar home systems and/or PV lanterns are among the frequently promoted
individual solutions. However, target households find these difficult to accept as a final solution. To them, a
PV system is often perceived as a poor substitute for real grid electricity because power output remains
limited, e.g. making it impossible to use a rice cooker or a kettle. This is disappointing, as investments in a
PV system are comparable to the costs of a grid-connection. Strictly speaking, this comparison is wrong as
grid electrification provide many additional benefits to society as a whole that PV systems do not. Since
prices are expected to fall further in the future, PV systems will then commensurate more with a pre-grid
situation. Rural electrification impact surveys confirmed that PV systems can be adequate and acceptable
temporary solutions, but eventually beneficiaries would like to have better solutions when their incomes
and living standards increase.
Mini-grids, connecting a limited number of households in a village could become viable when households
desire more services than PV systems or PV lanterns can provide. Mini-grids using diesel generators would
provide high quality, high cost electricity fairly quickly. However, these high costs often result in reduced
service delivery, such as supply during a few hours per day only. Mini-grids coupled to micro hydro plants
could provide lower-cost electricity, but so far their spread has been limited for a number of reasons: not
all geographical areas are prone to micro hydro development, some specific technical skills are required,
investment costs are relatively high, and institutional problems with operation and maintenance are not
uncommon since operational aspects are frequently transferred to the local population without ownership.
The use of electricity is almost addictive, demand appears to be always increasing and never decreasing;
the more electricity clients use, the easier it will be for utilities to supply. Once households obtained access
to electricity, they have effectively started a path that for a long time to come leads to the desire of higher
Draft

consumption. If and when the demand carried by mini-grids grows, eventually interconnection between
them is likely to become viable, and so is connection to the national electricity grid.

Towards long-term solutions to end Energy Poverty

Rather than promote final solutions as the starting point – as has been and still is common practice –, a
path towards such final solutions would be more useful. At any point in time, rural and peri-urban
2

households, institutions, and private firms have then access to solutions that fit their needs and their
Page

wallets of that moment; these temporary solutions may then shift over time until the conditions satisfied

EnDev ending energy poverty ([email protected])

for grid-based supply to become attractive in the distant future. This is not a quest for rapidly extending
electricity grid supply, but an attempt to focus more attention on solutions for those who will under no
circumstances obtain grid electricity in the near to medium term future. It is evident that technological
developments are progressing very fast and network effects of (large integrated) grids can be ruled out for
the foreseeable future by technology, specific needs, smart ICT and cooking solutions and the specific
population density. Similar to such development paths for electricity, there are also paths towards cooking
solutions, whereby households move up over time from firewood, charcoal, briquettes, to kerosene, LPG
and possibly electricity. See Figure 1 below for a diagram of several paths towards better electricity
services, and Annex 1 for more details.
Until now, the bulk of most countries’ electrification efforts have been to increase urban supplies and at
best trying to replicate urban supply mechanisms in rural areas, leaving aside a few pilot projects to
explore alternatives for use in rural areas. The time has come to state that this is no longer an adequate
approach if the goal is modern energy access for the majority of the population within a reasonable time
period.
A different approach is therefore needed, conceptually, chronologically and physically. If electricity were
not considered the end solution in itself but a means to improve the quality of life, then it would be
possible to improve rural living conditions fairly quickly. Although this cannot be called rural electrification
pure sang, however, it is effectively the beginning of an era of modern services delivery that is made
possible by having at least access to some electricity; in other words, the beginning of the end of energy
poverty. For many households, modern energy services will initially limited to the use of a television and/or
modern lights, both of which will increase living conditions and may contribute to improved education,
productivity gains, and/or accelerated economic development, but this will not make them attractive
clients for a grid-based electricity utility. It is slowly trickling down in the minds of planners that rural
electrification by grid extension may not be the most economical, technologically and developmentally
appropriate solution for many geographical areas in the short term.
By the time the population has outgrown a particular service delivery level, incomes are likely to have risen
and population densities increased, resulting in entirely different economic conditions under which other
temporary solutions might now become feasible (such as added PV modules to the battery system at
home, or mini-grids connecting most village households). A grid-connected city person may find it difficult
to imagine, but even the simplest first step up from kerosene lighting, a battery-based modern CFL or LED
lamp, would immediately improve the quality of life for rural households. There are numerous options to
improve living conditions without having grid electricity, and all have fairly low investment costs. Indeed, a
range of individual alternatives should be promoted to improve the quality of rural life across the board
and create wealth, to commensurate with households’ desire and ability to pay for such services. Although
these services constitute a major step up from prevailing living conditions, several subsequent steps will
still be required to reach comparable conditions in the future as in urban areas.

Draft
3
Page

EnDev ending energy poverty ([email protected])

 candles, kerosene, firewood, B
a
Flashlights, radio c
k
micro lights, (car) battery charging,
t
more modern services h
e
pv systems n

isolated minigrids,
eventually interconnected

l
a
Extension national grid t
(absorbing many if not all minigrids) e
r

Figure 1 above shows the generic flows of development towards universal electricity access; the starting point for each
household, institution, or SME depends on their status of economic development. Households in deep rural areas mainly
outside of the cash economy are likely to start at the first rung, firewood, candles, kerosene, and dry-cell batteries.
Households that have several cash crops and already accumulated certain wealth may start at the second rung (car
batteries, with or without a PV system). In this way, all households will develop their own path towards better services
over time.

Want to end Energy Poverty? Look for quick fixes first, not final
solutions
A demonstration of the fact that the above concept could actually be acceptable is the only solution ever
adopted on a large scale - without donor intervention or subsidies and in many countries across the Globe:
car batteries (new or recycled) for household use. This is a poor solution from a technical point of view, yet
a popular one by rural households. Such batteries are in no way a substitute for grid electricity, but simply
a way to modernize living conditions in the absence of a grid connection, enabling households to watch
television and move out of the scaring dark. Statistics reveal that more rural households in e.g. Kenya and
Cambodia have battery-based electricity supply instead of from the grid. Recent surveys show that rural
households are ready to buy and use low-cost modern appliances, if only these were available, such as
battery chargers, electric lights, radios, and TVs, etc. The costs and quality of such appliances have greatly
improved over the past few years, making a roll out now possible on a large scale in rural areas.
Draft

Even though it will not be possible to realize universal access in the foreseeable future to grid electricity, it
will be possible to end energy poverty fairly quickly for the bulk of the rural and peri-urban population
through the use of low-cost equipment that considerably improves living conditions.

Rolling out rural energy poverty measures requires a paradigm

shift
4

A “one way fits all” mode is good for public infrastructure investment programs such as expansion of urban
Page

grid electricity by utilities. This is a fairly straightforward approach that however cannot be used in rural

EnDev ending energy poverty ([email protected])

 areas to end energy poverty. A paradigm shift is required instead: matching solutions need to be supplied
or rolled-out at multiple levels, to satisfy millions of beneficiaries, dispersed all over the country-side,
everyone desiring his own specific service levels. One utility - or even a number of utilities - will never be
able to cope with such situation.
Instead, a market-based approach disseminating multiple solutions simultaneously is likely to be better
geared to realize such roll-out quickly, particularly when in conjunction with a joint public-private effort to
facilitate, develop, and promote these ideas further among the energy poor and potential suppliers alike.
!!! For the first time, the constraint is no longer technology but logistics: having the right equipment there
where it is needed remains a considerable issue. An appropriate set of policy and framework conditions,
such as exoneration of import duty and VAT, subsidies, appropriate financing mechanism, etc, would
convince private companies to set up distribution chains for the supply of equipment to all potential
beneficiaries. Adding such conditions to the country’s Poverty Reduction Strategy document is necessary
but certainly not sufficient: sustainability and effectiveness will remain an illusion unless both capacity and
willingness to effectively enforce these new policies are realized.
Until population densities and/or urbanization rates increase considerably, grid extension is not likely to be
cost-effective for providing access to electricity for a large part of the population – and will therefore not
happen naturally unless large subsidies are provided. Even then, lower cost alternatives could more quickly
provide adapted solutions that could be quite different for different beneficiaries. These specific solutions
are likely to change over time, reflecting changing economic conditions in the area and/or the beneficiary,
effectively setting in motion a process whereby individual households, institutions, or firms follow their
own path towards ever increasing modern energy service levels and lower costs of service.
To overcome energy poverty for the entire rural and peri-urban population, development programs should
do more than pursue grid extension and a few decentralized electrification projects. They should take on a
more holistic approach and effectively roll out the various complementary solutions simultaneously, on a
large scale, addressing households, institutions, and SMEs. As said, this requires a conducive policy
environment whereby private and public players jointly deliver the needed services and equipment.
The main elements of a program to increase access to modern energy for the whole of the population are:
• Public grid to cover all main urban areas, through public and private utilities;
o Special effort to connect clients in surrounding peri-urban areas, through innovative
connection and pricing policies;
o Commercially run mini-grids in areas with pockets of high population density, which
could be interconnected to the public grid in due course;
• Market-based individual solutions for those households, social institutions and SMEs which are
not in close vicinity of the planned grid path or mini-grid locations; those solutions comprise of a
whole range of technological options, which over time could be upgraded through incremental
investments (such as adding a PV module to a battery-based system). This requires several
simultaneous interventions:
o Enable, promote and assist private sector driven sustainable market development for
quickly rolling-out equipment to all rural areas
o Develop the institutional environment for verifying the quality of a large variety of
technological solutions (markets, standards!) and allocating subsidies
• Support mechanism
o Awareness campaign to promote all solutions simultaneously
o Financial assistance for institutions and the poorest households (credit + subsidies)
o Support mechanism to develop technically and financially optimal solutions for public
institutions and community groups (such as mini-grids)
Draft

In fact, many of these ideas are really not so new and have individually already been tested under the
EnDev Program. However, they have not been implemented in the holistic approach that is warranted if
the whole of Africa is to enjoy the 21st century benefits of modern energy within one generation.

Subsidies
5

The aforementioned text made several references to subsidies. Normally subsidies would not be necessary
Page

in a perfect world. There is a lot of momentum to not subsidize infrastructure expansion but make this the

EnDev ending energy poverty ([email protected])

responsibility of the private sector, and let it be the outcome of market development. However, to realize
this in the presence of prevailing market failures, non conducive policy frameworks and lack of supportive
concessional financing, and generally low disposable incomes for the target population, much more
capacity development and institutional reforms are required than is realistically possible. It is nearly
impossible to address all these shortcomings simultaneously and expect markets to take care of public
service delivery. Hence, progress and achievements are greatly facilitated when subsidies can be used to
provide smart incentives.
Indeed, subsidies were considered not done by the international development community for a long time,
but momentum is slowly gaining in favor of newly applying limited subsidies. Such subsidies should be
designed to initiate markets and avoid supporting the costs of operation and/or consumption in the long-
run, or designed as emergency relief to quickly put in place mechanisms that are then supposed to become
sustainable after some time, and have a credible and transparent exit strategy. EnDev clearly incorporated
subsidies as an efficient and transparent mechanism to accelerate access to modern energy in rural and
peri-urban areas.
We should not forget that the infrastructure of industrialized countries was built with a high level of
subsidies or public support! Electricity has been around for more than 100 years now, but the bulk of the
SSA population still does not have access and remains in the dark. This in itself could be sufficient reason to
develop an emergency program to correct this and apply subsidies accordingly.
Sometimes pure logic or economic analysis fails to indicate that subsidies should be awarded after all:
modern technologies are often cheaper in terms of life cycle costs compared to prevailing traditional
technologies and services; data presented in the Annex clearly shows that 4-7 month payback times exist
for investing in modern lighting services. Yet, people are not investing on a large scale as could be
expected. An explanation is that poor households already spend one third of their income on traditional
energy like candles, kerosene and fuel wood and have no means to invest in modern technology, however
urgently they are willing to do so. Subsidies would really make a difference, for households to invest in the
modern technology and for private firms to develop the infrastructure to deliver the technology. It is
important too face and accept the basic trade-off: speed vs. sustainability. Access is long over due, so let us
focus first on realizing this, and then worry about making it sustainable.
It helps to make the subsidy support transparent; in most countries today, it is impossible to indicate the
contribution of subsidies in the supply of grid electricity. Most likely these subsidies are huge, compounded
throughout the supply chain: the generation capacity is subsidized, the fuel is subsidized or detaxed, the
expansion of the transmission and distribution network is subsidized, connection fees and tariffs are
subsidized, etc. In Rwanda1 the electricity Roll-out program intends to connect 250,000 new customers
over the next few years and applies a beneficiary contribution of less than 10% of the real cost of
connection. It is not phrased as a subsidy, but effectively it is; imagine what could have been done with this
amount of money in support of finding modern energy solutions for all rural households! In Senegal2
several concessions have been awarded for increasing access to electricity; the average subsidy for a new
connection in concession areas is about $350; alternatively, one could hand out for free 5-7 PV lanterns for
the same amount.
Only a transparent cost and price structure lays the basis for continued discussion and optimization of the
support given, the more so since subsidies always remain the outcome of a political process. Although this
process cannot or should not be stopped, the subsidies it awards should be made transparent so that
beneficiaries and non-beneficiaries alike understand where the support went.
The next Figure 2 shows from the top down, differences between the “before” and “after” situation;
before refers to the situation now, whereby most households have no access to electricity or to modern
Draft

lighting equipment. The situation after refers to the medium term future (5-10 years), when the options
are rolled out on a large-scale and available to potential beneficairies. The top of the figure shows generic
categories for wealth, location, cost of energy access, and obtained level of services from having access to
electricity.
6
Page

1
Lighting Africa Country Study (work in progress)
2
Lighting Africa Country Study (work in progress)

EnDev ending energy poverty ([email protected])

Figure 2: Situation “Before” and “After”

Draft
7
Page

EnDev ending energy poverty ([email protected])

EnDev results so far
Background
The Energising Development Programme (EnDev) is currently implemented in 23 countries in Africa, Asia,
and Latin America by GTZ through a Dutch-German Partnership (DGIS-BMZ); its original objective was to
provide sustainable access to modern energy services to 3.1 million people in developing countries. It is
now being scaled up and expectations are that more than 9 million people will have gained sustainable
access during the period of 2005 – 2013. Under EnDev several forms of sustainable access to modern
energy services are being pursued, including cooking, lighting, ICT; the majority of beneficiaries so far
obtained better access to cooking energy, which meant more efficient cooking stoves. This document only
addresses services that rely on the use of electricity, not cooking.
EnDev actively supports achieving the Millennium Development Goals through a strong impact-oriented
approach centered around three main pillars all dealing with improving access to electricity: (i) for lighting
and other low-power household appliances, (ii) for enabling social infrastructure3, and (iii) for productive
use4 in small and medium enterprises. The first is to improve living conditions and allow access to some
priority services mainly for poor households; the second is to improve the quality of public services for
public institutions, and the third is to generate more income in rural areas. The technologies currently
supported are: (i) grid densification, (ii) micro-hydro power, (iii) PV systems, and (iv) modern low-cost
lighting equipment. In most of the EnDev countries only one of these technologies is pursued at the same
time.
To realize these goals, GTZ: (i) provides guidance to local partners in these countries for carrying out the
work; (ii) monitors their progress; and (iii) allocates subsidies to these partners for having created
sustainable access; the subsidy is based on actual achievements. The total support costs include program
costs5 as well as subsidies6, and effectively present the cost of bringing sustainable access to new users for
a particular technology. EnDev’s support costs are subject to a maximum per type of technology, and are
expressed in “Euro per person who obtained sustainable access to modern energy services”; a range of €10
- €100 per person has been observed from the experience to date, see Figure 3. These numbers can be
compared to the cost of providing such access
under other support programs and determine Productiv Lighting/
relative efficiencies. e Use, Househol
0.33 d
Since EnDev monitors both the number of applicatio
Social n,
persons who obtained access as well as the
Institution 0.98
costs for realizing this, it can make additional s, 1.74
funds available to scale up those activities with
the highest impact and cost-efficiency. This not
only allows for competition between the
different technologies, but also between the
different countries where EnDev is operational.
Activities with the highest impact and the Cooking,
greatest cost-efficiency have the greatest 7.92
chance of obtaining scale-up funds. For
Governments and donors, EnDev thus provides
a transparent mechanism that shows with how
little support one can make an impact. Figure 3: Total Results EnDev, million people (Dec 09)
Finally, EnDev support is available only for
energy services that are reliable, affordable, socially acceptable, and environmentally sound. In addition,
Draft

supported activities would not be developed without the assistance of EnDev (i.e., would not occur
naturally or implemented by other organizations). EnDev aims at transparency of the intervention and its
costs. To enable this, it developed instruments such as a subsidy matrix7 and an outcome calculation sheet

3
Schools, mosques, churches, hospitals, dispensaries, community buildings, etc
4
Employment & income generation through private sector activities.
8

5
Among others: program management, staff, capacity building, transaction, indirect support costs.
Page

6
Direct subsidy to the beneficiary, e.g., for the purchase of equipment.
7
Energy Subsidies: Why, When, and How?; Kilian Reiche and Witold Teplitz, Feb08.

EnDev ending energy poverty ([email protected])

(OCS), providing an easy overview of the results, costs, and impact. Such instruments show the weaknesses
of individual subsidy schemes and lay the ground for improvement and exit of the provided support; these
instruments are constantly being updated and improved so that newly allocated funds can be used even
more efficiently.

Results
The number of people reached by EnDev until December 2009 is about 11 million, of which 8.9m people
obtained access to electricity or improved cooking technologies at the household level. Through
institutions (social infrastructure) an additional 0.8m people were reached via “improved cooking energy,”
and 0.9m people via electricity or other modern energy carriers; in addition, another 0.3m people were
reached through productive use (See Figure 3, above). With adjustment factors8, the total number of
people is reduced to about 6.6m, and this is still higher than originally planned for.
Figure 4 shows the current results of increasing
battery Grid access to electricity (Feb ‘10), so excluding
charging, densifica cooking technologies or modern biomass: so far
12,940 tion, about 1.3 million people gained sustainable
393,028 access to electricity under the EnDev program.
SHS, For about 60% of the beneficiaries access was
761,596 obtained through PV systems (Bangladesh,
Ethiopia, Honduras, Mozambique, Nicaragua,
Senegal, Uganda); for 30% grid extension or grid
densification (Benin, Bolivia, Ethiopia, Ghana,
MHPP,
Mozambique, Nicaragua, Peru); and the
131,544
remainder was for micro or pico hydro (Bolivia,
Ethiopia, Honduras, Indonesia, Mozambique,
Nicaragua, and Rwanda) or PV-based battery
Figure 4: EnDev electricity access by technological solutions charging stations (Mali).
Figure 4 shows that the average EnDev support
was about €15 per person for grid densification, €56 for micro hydro power, €70 for PV equipment, and
€83 for battery charging stations; these data are likely to change in the future, when more results are
accounted for. Also note that PV equipment promoted under EnDev includes a full range, from small solar
lanterns (< 10 W) to Solar Home systems (> 50 Watt). The average support of € 70 per person for PV
systems excludes Bangladesh as this would bias the results too much: in Bangladesh the dissemination of
PV systems in the range of 50 – 80 Watt was very successful with an average support of €9 per person. This
would bring the average support over all EnDev countries to about € 11 per person, which is unrealistically
low. The results in Bangladesh are excellent and should be replicated anywhere else, although this might
initially be difficult for a lack of a similar large infrastructure for PV systems already in place.

Main lessons learnt

Annex 2 shows a SWOT analysis for EnDev vs other energy sector development programs (generic). A few
lessons are emerging from EnDev’s short electricity access experience:
• EnDev complements ”classical” electrification,
o whereby governments on the one hand realize electrification efforts and traditionally
focus on subsidized urban electrification and/or pilot PV programs;
o EnDev on the other hand addresses issues that are generally not addressed by other
donors and governments and focuses on energy poverty issues for the poorest segment
of the population that generally live in peri-urban and rural areas;
Draft

• EnDev’s outcome-based approach contributes to effectiveness of the activities and promotes

innovation; the focus on realizing concrete outputs within a short period of time instead of on the

8
Including: (i) “sustainability adjustment factor,” which takes into consideration that the access provided to modern
energy technologies is not sustainable in all cases, (ii) “windfall gain factor” considering that some of households
9

supported by EnDev would have gained access to modern energy services even without this support, and (iii)“double
Page

energy factor which takes into account that some households and welfare institutions which received an improved
stove or other modern cooking energy technologies already had access to electricity.

EnDev ending energy poverty ([email protected])

 precise inputs required and the design to
launch new activities yields quick results
90
for relatively low costs;
80
• Only the most successful activities, in
70
terms of achieving concrete outputs, will
60
be scaled up so that there is a strong
50
impetus for being effective, for all
projects within the same country but also 40
between different countries; 30
• An holistic approach is needed if access 20
to electricity is to lead to accelerated 10
economic development; this requires 0
simultaneously focus on households, Grid MHPP SHS PV-BCS
public services, and SME to promote just
in time economic development, with for Figure 4: Average EnDEv Support per technology (€ per
every beneficiary changing and person)
temporary solutions over time, reaching
a more final solution in due course; and
• Subsidies are common for grid electrification, but less so for off-grid solution; however, they are
essential for
o unlocking private and public funds to jointly co-finance priority solutions for the various
beneficiaries, whether households, institutions, or private firms; and
o for accelerating the mechanisms to provide access to modern energy for all beneficiaries
EnDev has created an innovative package of intervention to put a halt to energy poverty at reasonable
costs. It invites governments and donors alike to join the effort and reach a zero energy poverty situation
within the next decade.

EnDev: What’s Next?

EnDev should consider scaling up and addressing the access problem in an even more holistic manner than
before, following the outline as presented in the Section “Rolling-out requires a paradigm shift”. This would
mean that in every country where EnDev is active, the “full menu of options” will be pursued: (i) extension
of the grid in urban and peri-urban areas and isolated high density areas; (ii) development of private
sector, market-based distribution of the various off-grid modern energy options, from low-cost lighting
systems to micro-hydro plants; and (iii) ensure that the necessary support systems are in place and
operate, including regulatory framework, transparent subsidies, and concessional financing. If adequate
support systems are not in place, EnDev should consider withdrawing its assistance.
It is clear that the EnDev budget will not be sufficient for carrying out this work in all 23 countries.
Proactive coordination and collaboration with other donors is suggested so that joint action plans are
developed for rolling-out the full menu of options. Asking for more resources from BMZ and/or MinBuza
might be difficult but should not be ruled out, and so is inviting more donors to join. The task at hand is
serious and long over due!
Finally, it should be understood that access to modern energy is not simply providing a one-off solution,
but creating several paths towards such solution. This also means that EnDev (or similar) assistance should
be available over a longer period of time, to guide beneficiaries all along the way of their own path to full-
scale modern energy access.
Draft
10
Page

EnDev ending energy poverty ([email protected])

 Annex 1
Costs & Benefits of Lighting
Introduction
The following provides a simplified analysis of the costs and benefits associated with the various
options for access to modern energy and ending energy poverty. Since lighting is one of the first
benefits for households when they start using modern energy in the form of electricity – albeit in
small quantities –, this analysis focuses on lighting. Better lighting would effectively end dark age
conditions and allow beneficiaries to enter the 21th century. This simplification is somewhat
biased against the higher-cost options of electrification, since these can provide additional
benefits than just lighting: i.e., with a PV system households can enjoy television but not water
heating, while with a grid connection there are no more constraints on the use of electricity. This
will be qualified somewhat in the analysis9.
Several generic options exist for households to enjoy certain levels of lighting benefits, each at
certain costs that are specific for that benefit level and that commensurate with their income
level. When economic conditions improve, households may opt for additional investments to
step up their benefit level. This Annex explores (i) costs and benefits for some of these options,
and (ii) the benefits of stepping up a service level. In sequence, traditional lighting options are
described, followed by the following electricity-based options; drycell batteries (plus charging
somewhere), PV micro-lights (with LED), PV lantern (with CFL), car battery, PV system, and grid
connection.

The conclusion of the analysis is that, not unlike other services such as cooking, the poorest
households using the lowest-cost lighting options pay the dearest!

Traditional lighting
Candle, kerosene
Wood fires are used for lighting in a number of countries, but this is not analyzed here. The first
and most simple device used specifically for lighting is the candle, one of the early references
that was established as a lighting unit (the Candela, or the lighting intensity of a single candle). A
candle burning at about 8 g/hr emits roughly 12 lumen of visible light, or 0.1 lumen per W. Some
60 candles are needed to give the equivalent light of a 60 W incandescent lamp. A candle is
therefore not a comparable to any of the electric lighting solutions; although it may be very
romantic, it is also very unpractical to have 60 candles in a room for a decent lighting level. The
reference for comparison that we will use here is a 60 W incandescent lamp, even though this is
past technology and will be phased out from the market in the near future. Such a lamp yields
about 700 lumen of light (at 12 lumen/W).
Draft

The candle, or the equivalent that uses kerosene, a lamp made from a tomato paste tin or other
tin can plus a wick, is the most basic lighting solution. Particularly the kerosene-based lamp is
used often as it allows the user to put in a little bit of fuel when he has the money, and although
it is not enough to read comfortably, it does allow users from bumping into furniture at night. A
typical consumption is 30 - 40 ml per day, or less than €0.05/day.
11
Page

9
Data used in the Annex are from detailed EnDev activity reports as well as from the author’s own experience, mainly
in Rwanda.

EnDev ending energy poverty ([email protected])

The hurricane lantern consumes more kerosene but also emits more light, making some (still)
uncomfortable reading possible. It emits about 40 lumen in total (or 3.3 times a candle), and
consumes roughly 300 ml per day (50-80 ml/hr), depending on the nr of hours used and the size
of the wick; a typical expenditure would be about €0.25 per day.
These two types of lamps give light and a lot of heat, but that’s all they do. There are no other
benefits that can be derived from their usage.

Modern electric lamps

Dry-cell batteries
Throw away dry-cell batteries are commonly used in radios, flashlights, and more recently
rechargeable batteries (NiCad, NiMH) are used in a number of “micro-lights”, that are configured
either as a desk lamp, a head lamp, ceiling lamp, or hurricane lamp replica. The micro lights
typically use LED lamps in low- and high setting, allowing for a set number of hours of usage
before recharging is needed. As an example, one of these micro-lights using 3 NiCad AA cells (600
mAh capacity each) has a 40 hr usage period at the low-setting, which would allow a 10 day
recharging period. A few hours are possible at the high setting. If better batteries were used, a 2-
3 week recharging period with a reasonable light output can be envisioned. Nevertheless, the
use of these devices is limited and restricted to lighting only.
Such lamps can be purchased in Africa in a number of countries for €15-25, and typical recharging
services cost €0.50-1.0. Micro lights have become feasible when the LED lamp “graduated” and
became commercially available, which is just 1-2 years ago. It is therefore amazing to observe
that LED flashlights have inundated the market and almost suppressed flashlights with an
incandescent lamp.

PV micro-lights
The sensible thing to do is to buy one of the battery operated micro lights in combination with a
1 W PV panel, so that 3rd party charging is not needed. A panel typically adds €10 to the system
costs (thus total costs: around €25-40). The incremental investment pays back in 5 years, well
before the end of the life of the equipment. Although this alone may not be enough to justify
investing in a panel afterwards, it certainly simplifies life by not having to go some place for a
recharge.
The lumens produced range from 7-70, which is between the level of a candle (although the light
is more focused/directed than a candle) and two hurricane lanterns.

PV lantern
PV lanterns are different from micro lights in that they have been around for a longer time and
there is much more choice of equipment. In terms of storage, typically gel type or wet-cell (lead-
acid, motor cycle) batteries are used with a capacity of 5-15 Ah, and more often than not a 5-11
W CFL lamp is used. In price, they range from €50 to over €200. It is probably a matter of time
before PV lanterns with LEDs become available; although LED lamps are more expensive, they
use much less power than a CFL and therefore battery capacity can be reduced for the same level
Draft

of service. CFLs have an efficacy of up to 50 Lm/W, whereas the latest “power LED’s” reach up to
75 Lm/W. Although LED’s are commercially available, their R&D continues and the end of the
development path has not been reached yet.

Car battery,
Car batteries (lead acid, wet cell, or gel cell) or ideally deep discharge “solar batteries” can be
12

used as a standalone option; in fact, for many African households this is the starting point of their
electricity system, as it allows more services to be powered than just lighting. In that case the
Page

battery is likely to be a 2nd hand car battery rather than a newly purchased deep discharge

EnDev ending energy poverty ([email protected])

battery. As an example, a television set can be used with such battery; there are not many other
low-cost off-grid alternatives that can do the same.
The battery typically holds 30-50 Ah and can be purchased for €50-75; depending on the level of
discharge, 300 recharges or less are possible. Households get a recharge often in a shop in the
next town once every 2-4 weeks, and pay something like €1-2 per charge.

PV system
The battery system can be expanded with a PV panel; a 20 W panel can usually be purchased for
€100 and allows users to forego further recharging at off-site locations. This adds to the
convenience of not having to lug around a heavy battery, and to the profitability as well as the
incremental investment is paid back in about 2 years.
A full PV system with 3-5 CFL lights and 50 W panel can be bought as a kit and is often promoted
as the starting point for off-grid electrification. An additional panel could be added later to
increase the level of services. It will still not be possible to use water heaters, rice cookers, etc.

Grid connection
This is still the preferred solution for a number of reasons. Grid connections are not available at
all locations, but if available they are a good solution: medium investment costs (unsubsidized a
minimum of €500) and in principle unlimited power – as long as the electricity bill is paid. The
cost of electricity is lower than for any of aforementioned the other solutions, with a typical cost
reflective tariff of €0.20 per kWh. For lighting grid customers still use incandescent lamps (10-15
Lm/W) rather than CFLs, mainly because of the purchase price differences: a bulb is much
cheaper than a CFL, although this is changing rapidly. In addition, a CFL sometimes has a poor
reputation, caused by the flood of poor quality Chinese CFLs of a few years ago. It is unlikely that
households will buy LED lamps for the time being.

Cost and Benefits of the various options

Figures 1 and 2 and Table 1 show the results of the analysis; Figure 1 shows that the cost per
kWh of electricity is reverse proportional to the investment costs (initial purchase) of the lighting
option. The more expensive the option, the lower the cost of electricity, and as can be seen in
Figure 2, the more lighting services (and other services) can be obtained. The poorest households
that barely can afford the first modern energy option, pay the highest per unit costs. Yet, in terms
of benefits, they also realize the largest gains: in terms of ease of use, quantity of lumens, costs,
health, and safety.
Investments range from a few Euro to over €500 for a grid connection, and the cost of a kWh of
electricity ranges from around €0.30 to not less than a whopping €35 per kWh for a battery-
based light with 3rd party recharging.
Draft
13
Page

EnDev ending energy poverty ([email protected])

 Figure 1: Investments & kWh costs
600 35
500 30

Cost per kWh (Eueo)

Investments (Euro)

25
400
20
300
15
200
10
100 5
0 0

Figure 2 shows for the same lighting options the average use of electricity per month and the
cost of lighting (in Euro per k.lumen.hours). A similar conclusion as for Figure 1 can be given: the
level of services to be obtained is reverse proportional with the cost of lighting. The contrast is
stark: those with the highest cost of lighting (> 1 Euro per klmh) have almost zero or effectively
zero kWh of electricity to spend; those with the lowest cost of lighting ( < 0.1 Euro per klmh) have
the largest nr of kWh at their disposal.

Figure 2: kWh use & cost per klmh

45 1.4
40 1.2
35

Euro per klumhr

kWh per month

30 1.0
25 0.8
20 0.6
15 0.4
10
5 0.2
0 0.0
Draft
14
Page

EnDev ending energy poverty ([email protected])

Table 1: Costs of lighting options

Candle/ Hurricane Battery PV PV car PV PV Grid

tomatopaste lantern only micro lantern battery system, system, connection
tin light 5W only 20 W 50 W
1W

investment 0.75 5 28 38 60 105 205 355 500

costs
O&M costs 1.3 8.1 2.8 0.5 0.9 6.3 2.3 3.0 8.0
(€/m)
kWh/month 0.1 0.1 0.7 3.5 3.8 6.7 40
klmh/month 1.4 6.0 6.5 10 33 174 188 336 1200
cost/kWh 33 7.2 2.3 3.0 2.3 0.9 0.3
cost/klmh 0.8 1.3 0.4 0.10 0.05 0.06 0.05 0.02 0.01

Table 1 shows that when the lighting options become more expensive to buy into (investment
costs), they yield more benefits and at lower costs; factor of 100 difference is no exception!
Appendix I shows assumptions and a breakdown of the calculations.
In what follows, the implications of upgrading from one lighting option to the next is analyzed.
The following steps are considered:
Hurricane lantern -> dry-cell battery
-> Micro-light with 1 Watt PV panel
-> PV lantern
-> car battery
Dry-cell battery -> Micro-light with 1 Watt PV panel
Car battery -> PV system 20 W
-> PV system 50 W

Table 2 shows the results; for the lowest-cost options it certainly makes sense to switch up to a
next level, and pay back times of 4-8 months appear acceptable. However, the higher the service
level, the less attractive it becomes to switch even further up. This is counter intuitive, as the
richest households are the most eager to step up to a next service level, and the explanation
must be that this is the limit of the simplified analysis: other benefits should also be accounted
for that are not valued in the current analysis, such as those accruing to society. The next chapter
deals with this issue in somewhat more depth.

Table 2: Payback time (months) for switching to the next level

Hurricane Dry-cell car battery

lantern battery
Dry-cell battery 4.3
micro light 4.4 4.5
Draft

pv lantern 5 W 7.6 17
car battery 54
PV system 20 W 35 386 25
PV system 50 W 69 77
Grid 4752
15
Page

EnDev ending energy poverty ([email protected])

 Community Benefits
The above analysis touched upon individual benefits, i.e., those accruing to the household.
Households most likely first obtain better lighting services and when economic conditions further
improve, they are likely to obtain battery charging, radio, TV, and possibly even water heating
services (rice cooker, kettle etc). However, one side effect of the higher service levels is that in
addition to household benefits also important benefits accrue to society.
An example is the collective wastewater and sewerage system: the difference between a village
with and without sewerage is mainly felt at the village level, as every household will have his own
individual solutions for getting rid of the waste materials, some with poor solutions (outhouses)
and some with quite comfortable solutions (septic tanks). The main benefit however is the fact
that hygienic conditions for the village as a whole improve as well, over and above any benefits at
the individual level. The same holds for electricity, when higher service level become available to
households, community benefits may accrue to society as well; until now, these have so far been
left out of the analysis in this document.
The following section briefly addresses community or societal benefits from the use of electricity;
this will mainly be done in a qualitative manner. The same service level options as presented in
Figure 1 of the main text are considered here in somewhat more detail, and individual as well as
community benefits are highlighted. The benefits are valued (mediocre: 0, or red; limited
benefits: 1, or yellow; full benefits: 2, or green). See Figure 3 below for a matrix with the service
level options and the benefits. The column at the right shows what percentage of the benefits
are community benefits (using the value quantification system as described above).
• The first community benefit from access to electricity is enhanced outdoor security,
which can be achieved by having more floodlights around individual houses and/or street
lighting. Enhanced security allows people to move more freely at night and reduces crime
such as theft and robberies.
• The second community benefit is better public services made possible by having access
to electricity; services such as electricity, water and sewerage delivery, but also health
services (dispensaries, doctor’s offices, hospitals), educational services (schools),
community services (administration offices, cinema), religious services (churches,
mosques). The whole community will benefit from enhanced service delivery, although
some households may benefit more than others.
• The third community benefit is the result of additional employment opportunities
provided by access to electricity and income generation. Having access to electricity
opens up possibilities for new businesses and thereby creates new jobs. Simple services
such as cell phone charging or car battery charging and flour milling are probably among
the first new services that will be set up. More productive activities, such as
manufacturing, processing, assembling may also occur. These activities have individual
benefits at the household level (higher income) and community benefits (larger tax base,
larger turn-over for other businesses through higher spending power for the inhabitants).
Draft
16
Page

EnDev ending energy poverty ([email protected])

 individual benefits community benefits

community benefits
security outside light

income generation
cellphone charging

social institutions

productive use
motive power
water heating
ambient light

street light
socializing

reading

radio

TV
candle, firewood 1 2 0%
kerosene 1 2 2 0%
microlight without PV 2 2 0%
micro light with PV 1 W 2 3 0%
car battery 2 3 3 2 1 0%
PV lantern 10 W 3 3 2 2 0%
PV system 20 W 3 3 3 3 2 0%
PV syste 50 W 3 3 3 3 3 3 1 5%
micro hydro 3 3 3 3 3 3 1 2 1 2 2 1 1 32%
mini-grid 3 3 3 3 3 3 1 2 2 2 2 1 1 34%
low-cost grid 3 3 3 3 3 3 1 3 3 3 2 1 1 41%
full grid 3 3 3 3 3 3 3 3 3 3 3 3 3 46%

Draft
17
Page

EnDev ending energy poverty ([email protected])

 Appendix
Candle/
tomatopaste Hurricane Battery PV lantern, PV lantern,
tin lantern only 1W 5W
investment costs
battery 15 15 25
panel 10 25
lights 0.75 5 8 8 5
other 5 5 5
total 0.75 5 28 38 60

hr use/d 4 5 3 4 5
Wh/d 2.9 4.5 22.3
kWh/month 0.09 0.13 0.67
Life (years)
battery 3 3 3
panel 20 20
lights 1 4 10 10 6
other 20 20 20

lumen/watt 75 75 50
lumen 12 40
k.lumen.hr/month 1.44 6 7 10 33
O&M costs (Euro/month)
fuel/recharge 1.2 8 2.3
battery 0.4 0.4 0.7
panel 0.0 0.1
lights 0.1 0.1 0.1 0.1 0.1
other 0.0 0.0 0.0
total O&M 1.3 8.1 2.8 0.5 0.9
Draft
18
Page

EnDev ending energy poverty ([email protected])

 battery PV system, PV system,
only 20 W 50 W
investment costs
battery 50 50 75
panel 100 200
lights 30 30 30
other 25 25 50
total 105 205 355

hr use/d 2.5 2.5 5

Wh/d 125 125 224
kWh/month 3.5 3.8 6.7
Life (years)
battery 3 3 4
panel 20 20
lights 6 6 6
other 20 20 20

lumen/watt 50 50 50
lumen
k.lumen.hr/month 174 188 336
O&M costs (Euro/month)
fuel/recharge 4.3
battery 1.4 1.4 1.6
panel 0.4 0.8
lights 0.4 0.4 0.4
other 0.1 0.1 0.2
total O&M 6.3 2.3 3.0

Draft
19
Page

EnDev ending energy poverty ([email protected])

 Grid connection
transformator
- kW 50
- costs € 30000
wire
- €/meter 5
- m per hh 20
- labor hr/installation 2
- ,, €/hr 25
household
- demand W 500
- demand kWh/m 40
nr hh 100
hrs/m (max) 80
trariff €/kWh 0.2
prepayment meter
- costs € 50
cost per hh
-trafo 300
-wire 100
- meter 50
- installation 50
500
lum/Watt 30
k.lum.h/month 1200
cost/k.lm.hr 0.010
Assumption: kWh price includes costs for
further capacity expansion of the network

Draft
20
Page

EnDev ending energy poverty ([email protected])

Annex 2: SWOT Analysis
EnDev & Electricity Sector Development assistance

ESD EnDev
Strength Simple approach for grid Complementary to ESD
extension, one donor Addressing rural areas, taking the pilot
(Government), one recipient process a step further than previous
(utility) activities
Simple procurement of services, Public private partnerships can deliver
equipment and works fast and to a wide audience
Easy sector planning through Low-cost decentralized solutions can be
Master Plan for Electrification promoted through market development,
Some pilot projects dealing with which is difficult through large-scale
decentralized electrification public projects
Weakness Only urban areas, cost-prohibitive Verification of results becomes more
in rural areas complex due to the multitude of actors,
Large-scale intervention requires both beneficiaries and private service &
lengthy social & environmental equipment delivery firms.
impact assessments Location of intervention is more difficult
Decentralized electrification to select
projects never went beyond the
pilot phase, or were taken
seriously indeed
Opportunity Grid extension is a low-cost Creating markets is part of development;
solution, if not the lowest cost this can be applied to more appliances
option providing the highest and services, which will benefit the whole
quality economy
Simple set up and limited number
of actors facilitated the use of
subsidy programs
Risk Cost over-runs, corruption, slow Weak private sector may result in delays
implementation due to or failure.
government rules

Draft
21
Page

EnDev ending energy poverty ([email protected])