What are the major barriers to increased use of modern energy services among the world s poorest people and are interventions to overcome these effective?

What are the major barriers to increased use of modern energy services among the world s poorest people and are interventions to overcome these effective?
Collaboration for
Environmental Evidence
Watson et al. 2011
CEE review 11-004
Systematic Review
School of Business, Management and Economics - SPRU – Science and Technology Policy Research The Freeman Centre - University of Sussex - Brighton BN1 9QE- UK
RAND Europe, Westbrook Centre, Milton Road, Cambridge, CB4 1YG, UK
Corresponding authors: Professor
ofessor Jim Watson (([email protected])
and Molly Morgan Jones (([email protected])
Draft protocol published on website: 30 May 2011 - Final protocol published on website:
te: 10 August 2011 nd
Draft review
iew published on website: 7 February 2012 - Final review posted on website: 2 October 2012
Cite as: Watson, J., Byrne, R.,, Morgan Jones, M., Tsang, F., Opazo, J.,, Fry, C. & Castle-Clarke,
S. 2012.
What are the major barriers to increased use of modern energy services among the world’s poorest
people and are interventions to overcome these effective? CEE Review 11
004. Collaboration for
Environmental Evidence: www.environmentalevidence.org/ SR
Background: A lack of access to modern energy services among the world’s poor is widely
recognised to have negative impacts on their health, education and quality of life, as well as a
lack of growth of income at a national and individual scale, further deepening and entrenching
their poverty. However, despite the long-standing efforts of many national and international
organisations to improve the accessibility of the poor to modern energy services, progress has
been slow. Given the uneven record of interventions over many years, there is a large body of
literature that attempts to identify what is preventing success (i.e. what are the barriers), and what
policies might be implemented to realise widespread access to modern energy services for the
world’s poor.
Methods: The review was performed in order to answer the question: (i) “What are the major
barriers to increased use of modern energy services among the world’s poorest people”, and (ii)
“are interventions to overcome these effective?” A structured, systematic review of academic and
grey literature focussed on developing economies, including the BRICS (Brazil, Russia, India,
China and South Africa) countries, was conducted according to a detailed search protocol. This
included five sets of search terms relating to modern energy services, modern energy
technologies, barriers, interventions, and effectiveness measures. Semi-structured interviews were
also conducted to triangulate the literature searches and findings. Retrieved papers were then
systematically reviewed and included in the study according to a detailed set of criteria related to
relevance to the topic of barriers to, and interventions for, modern energy services. In particular,
papers were included if relevant to barriers, interventions, lessons learned and geographic focus.
Furthermore, each paper was assessed in terms of the quality of its evidence base and foundation
for conclusions. Papers which met these criteria as well as a rigorous set of quality criteria for
methodological robustness were qualitatively analysed and synthesised into a single narrative.
Results: Despite the large body of work analysing barriers to, and interventions for, using
modern energy services, there is a highly uneven spread of coverage and a significant lack of high
quality research. Much of the literature focuses on financial barriers and interventions, electricity
services, a few technologies, and a small number of developing countries. Within the limited high
quality research available, there are constraints to the certainty with which conclusions can be
drawn about what barriers are the most important, and what is the relative effectiveness of
interventions to overcome them. While this is problematic for policy makers seeking to intervene
to increase the use of modern energy services amongst the world’s poorest people, there are areas
where evidence is conclusive. It is also important to note that the particular method of conducting
a systematic review may exclude significant amounts of evidence from the practitioner literature,
which might be reporting valuable knowledge relevant to the focus of this review.
Conclusions: Most of the evidence on economic and technical barriers to energy access is
consistent and strong. Specifically, this evidence relates to high upfront costs of energy
conversion technologies and grid-connection charges, cost-recovery difficulties, poor
performance of equipment, and technical capacities for operation and maintenance. However,
evidence for interventions to overcome these is less robust. The weakest evidence concerns
political and cultural barriers and associated interventions, despite frequent references to their
importance. Moreover, our review highlights the interactions between different types of barriers
and interventions. To understand these interactions, and increase the chances that the poor can
gain access to modern energy services, analyses of barriers and implementation of interventions
should be more systemic. The review concludes with implications for policy, management and
research that flow from these conclusions.
Modern energy services, Modern energy technologies, barriers, interventions, developing
countries, Sub-Saharan Africa
1. Background
Although there is no simple definition, modern energy services are often identified in terms that
contrast them with traditional energy services such as those derived from the burning of biomass
in open fires (UN-Energy 2005; Brew-Hammond 2010). As such, the notion tends to combine
both energy carriers and associated technologies, together with the benefits to users that these
afford: lighting, cooking, heating, transportation, and so forth (UNDP 2005). Examples of
modern energy services, therefore, include (among others) electricity from solar home systems
(SHSs) for lighting, natural gas burned in modern stoves for cooking and petroleum-based
engines for motive power to enable agro-processing (Modi et al. 2005; Practical Action 2010).
However, as this review will show, there is not always complete agreement or clarity within the
literature about how modern energy services are defined. The overall review strategy therefore
aimed to be as broad as possible so as to enable a robust analysis of the concept. However, further
comments are made in the results section about refining this definition.
A lack of access to modern energy services among the world’s poor is widely recognised to have
negative impacts on their health, education and quality of life, as well as a lack of growth of
income at a national and individual scale, further deepening and entrenching their poverty (DFID
2002; Modi et al. 2005; UNDP-WHO 2009; Bazilian et al. 2010). However, despite the longstanding efforts of many national and international organisations to improve the accessibility of
the poor to modern energy services, progress has been slow (Modi et al. 2005). One notable
exception to this pattern is the case of China’s electrification programme, which has achieved
about a 99% electrification rate (although 1% of the Chinese population is still a large number of
people) (Urban 2009). Nevertheless, the record in China is not entirely one of success: there are
significant problems with, for example, the reliability of China’s electricity supply (Cherni and
Kentish 2007). If progress elsewhere continues along current trends, the world’s energy-poor will
remain so, with the current 1.4 billion without access to electricity only falling to 1.2 billion by
2030 and the 2.7 billion who rely on traditional biomass today rising to 2.8 billion over the same
period (OECD-IEA 2010). Some interventions have had limited beneficial impacts, while others
have worsened the situation for the energy-poor (Karekezi and Sihag 2004)1.
Given this uneven record of interventions over many years, there is a large body of literature that
attempts to identify what is preventing success, and what policies might be implemented to
realise widespread access to modern energy services for the world’s poor. One of the abiding
concepts in these analyses is that of barriers to access or to the adoption of technologies that can
deliver modern energy services. The UK Department for International Development (DFID)
commissioned this systematic review in order to “neutrally collect, critically appraise and
synthesise” the evidence provided in the literature on barriers to the use of modern energy
services among the world’s poorest people, and interventions to remove those barriers, as part of
its drive for evidence-based policy making (DFID et al. 2010:1). The final review provides an
analysis of the evidence base so as to inform DFID’s policy and practice. DFID has asked the
review team to focus on sub-Saharan Africa, but the team has conducted the searches in as broad
a manner as possible in order to capture lessons learned from lower-middle income countries in
other parts of the world.
The review focuses on the barriers affecting or interventions targeting poor people in developing countries, particularly
those of Sub-Saharan Africa, who lack access to modern energy services. We acknowledge that the terms poor or poorest
people are rather simplistic and general. However as poverty levels and thresholds are defined by the particular conditions of
every country or territory we have remained cautious in adopting a particular definition or parameter (such as a predefined
average income level) to assess inclusion/exclusion of papers. We have nonetheless adopted World Bank categorisation of
“developing country” (WDI 2009) as one of the relevance criteria (see section 3 below).
The main body of this report is structured in the following way: Sections 2 and 3 set out the
research protocol that was used to collect, analyse and synthesise the available evidence.2 Section
4 details the findings and analysis. Section 5 discusses limitations of the study and Section 6
presents final conclusions, including implications for policy and knowledge gaps. Sections 7-10
cover final aspects of the review, including a full list of references in Section 9. Appendices A
and B list the search terms used for the academic and grey literature, respectively. Appendix C
provides the quality review criteria, and appendix D gives any conflicting interests of the authors.
The results from pilot testing are given in appendix E, and a list of articles excluded at full text
stage is presented in Appendix F. Appendices G and H provide quantitative analysis of the grey
literature and appendices I and J give the systematic maps for this grey literature.
2. Objective of the Review
The objective of this review was to address two primary questions: (i) “What are the major
barriers to increased use of modern energy services among the world’s poorest people”, and (ii)
“are interventions to overcome these effective?”
The following sub-research questions immediately fall out of the primary, overarching question.
What are the different types of modern energy services, and associated technologies for
their delivery and/or use, which are used, or could be used, by the world’s poorest
What are the major economic, technical, political, cultural and social barriers to the use of
modern energy services, and associated technologies, amongst the world’s poorest
people? What criteria are used to define a ‘major’ barrier?
What are the different types of interventions that have been used to promote or increase
uptake of modern energy services and technologies? What are their characteristics and
What are the different measures of effectiveness which are used to assess the
interventions and how do these vary depending on the intervention, the modern energy
service and the relevant barriers in different contexts?
DFID requested that the review concentrated on modern energy services other than cooking, as
another systematic review has been commissioned to understand the specific issues related to
cooking technologies, such as kerosene or gas which are often categorised as modern fuels for
cooking. The review did not include these (and other modern cooking fuels) and the associated
technologies, nor did it include transport-related technologies. The specific list of terms used to
exclude such technologies is given below in section 3.1.1.
3. Methods
3.1 Searches
3.1.1 Search terms and languages
A broad range and combination of search terms were used. An initial list of search terms, divided
according to the four sub-questions, is given below. These were continually refined during the
See Watson, J., Byrne, R., Opazo, J., Tsang, F., Morgan-Jones, M. & Diepeveen, S. 2011. What are the major barriers to
increased use of modern energy services among the world s poorest people and are interventions to overcome these
effective? CEE protocol 11-004. Collaboration for Environmental Evidence: www.environmentalevidence.org/
search, and after a pilot search stage (see below) and the list here represents the final search terms
which were used. As far as possible, the terms allow for variants of word beginnings and endings.
For example, *phone charg* should capture phone and telephone as well as charger, chargers
and charging. The full set of search terms are provided in Appendix A.
The search terms were applied to TITLE-ABSTR-KEY searches in all cases for the academic
literature databases. However, this was not possible for all of the grey literature as this depended
on the capacity of individual databases. A full set of search terms for each grey literature database
is given in Appendix B.
Some of the terms on the above lists have generic meanings in addition to their specific meanings
in the field of energy. For example, the pilot test search revealed that the terms “light*” and
“illuminat*” resulted in many irrelevant hits which have the terms in their abstracts to mean “to
show clearly”. Therefore, in order to improve the relevance of the search, it was judged that it
would be necessary to ensure that either the terms “energy” or “electricity” were in every search.
To ensure that the search retrieved a manageable number of hits, the parameters of the search
were also defined by language and location. Only English language publications were included,
given the skills of the research team members and the resource constraints of the current review.
In addition, the main emphasis of the review was on studies conducted in low-income countries
in Sub-Saharan Africa. For the most part, only these studies were included past the title review
stage, although it is recognised that some studies conducted elsewhere will include important
comparators with sub-Saharan African contexts. In light of this, a targeted search was done of the
retrieved literature covering Brazil, Russia, India, China and South Africa (otherwise known as
BRICS) for specific interventions and electrification programmes. Further information about this
search and geographic and topical coverage of the review is discussed in sections 3.1.7 and 3.2.
3.1.2 Search strings and/or combinations of searches
The primary review question is comprised of two parts, with the first part focussed on the barriers
and the second part focussed on the interventions. It was recognised that papers that are relevant
to barriers may not necessarily discuss interventions, but papers that are relevant to interventions
should cover the specific barriers which are being addressed. Three searches were therefore
undertaken; one each to identify barriers, interventions and effectiveness measures. All three
searches included the terms for energy services, technologies and geographic regions. Many of
the papers appeared in more than one of the searches and duplicates were removed prior to
undertaking the title and abstract review to avoid any duplication in effort.
Search #1 to locate papers that identified the energy services/technologies and barriers in the
selected geographic regions:
(energy OR electricity)
(List 1a and List 1b search terms “OR-ed” together)
(List 2 search terms “OR-ed” together)
(List 5 search terms “OR-ed” together)
Search #2 to locate papers that focussed on the energy services/technologies and interventions in
the selected geographic regions:
(energy OR electricity)
(List 1a and List 1b search terms “OR-ed” together)
(List 3 search terms “OR-ed” together)
(List 5 search terms “OR-ed” together)
Search #3 to locate papers that focussed on the energy services/technologies and effectiveness
measures in the selected geographic regions:
(energy OR electricity)
(List 1a and List 1b search terms “OR-ed” together)
(List 4 search terms “OR-ed” together)
(List 5 search terms “OR-ed” together)
We undertook pilot searches to test the effectiveness of these keywords. Key pilot search results
which influenced our search strategy were recorded in Appendix E.
3.1.3 Comprehensiveness and effectiveness of the search
While a systematic review methodology is designed to be comprehensive, it is always possible
that relevant articles are missed. In this review, librarians 3 formally trained in conducting
comprehensive literature searches using technical search strategies and data retrieval methods
were used to help minimise this risk. All of the searches, including pilot testing, were conducted
by a RAND librarian with oversight and continued iteration with members of the review team so
as to ensure the retrieved articles were relevant to the purposes of the study.
In addition, steps were taken to limit the potential for bias of some geographic locations over
others. Though the task was to focus specifically on sub-Saharan Africa, the searches were
conducted in as broad a manner as possible and geographic exclusion terms were only applied to
the search at the abstract review stage.
In addition, there is a risk that a literature search will not capture relevant findings in books or
edited books where individual chapters are not indexed. Additional interviews were undertaken
with a selected group of seven experts, identified in consultation with DFID, to minimise this risk
and identify relevant literature our search might have missed. The interviews were also used to
gain key insights into barriers and interventions and these were triangulated against the findings
from the literature analysis.
3.1.4 Publication databases searched
A range of scholarly database sources were searched, such as:
subscription databases including JSTOR, Web of Science, WorldCat, Business Source
Premier, GreenFILE, U.S. Department of Energy (U.S. DOE) Energy Citations Database,
and EconLit.
in particular, individual journals such as Energy Policy, Renewable and Sustainable
Energy Reviews, Global Environmental Change, World Development, and Rural
The librarians are trained in providing customised research support by assisting with development of search strategy,
literature searches, citation management, cited reference searches and document delivery. In addition, RAND librarians are
knowledgeable in locating and retrieving difficult to find and elusive sources for researchers and have expert knowledge in
rapid searches of internet sources and stakeholder organisations in order to retrieve grey literature.
Sociology were included in the subscription databases as these journals were particularly
relevant to our review.
3.1.5 Specialist search for grey literature
Grey literature was retrieved through searches conducted in the following non-subscription
databases and websites of key stakeholder organisations:
the International Energy Agency, United Nations databases, Organisation for Economic
Cooperation and Development, the World Health Organisation, the World Bank, the
African Development Bank, the Renewable Energy and Energy Efficiency Partnership,
the Global Village Energy Partnership, and the African Energy Policy Network.
Similar search terms to those used in the subscription databases were applied to the internet
search, although depending on the database we were often limited in our ability to apply all the
search terms. We also applied additional, specific inclusion/exclusion and quality review criteria
to the grey literature which are further described in Section 3.2 below.
3.1.6 Literature provided directly by stakeholders, including peer reviewers
We sought advice from DFID on an appropriate list of key stakeholders to contact and conducted
interviews with them as discussed above. When literature was provided to us by stakeholders, we
included it in the review along with the other articles and evaluated it according to the established
study inclusion criteria. Similarly, when expert views were provided to us by stakeholders, we
considered it as a single source of evidence and assessed its quality and relevance to the review
during the data analysis stage (described below in Sections 3.4-3.6).
3.1.7 Targeted search of BRICS literature
As mentioned above, though the main emphasis of the search was on literature in Sub-Saharan
Africa, it was important to capture some lessons learned from interventions in other countries. In
consultation with DFID, and after an initial assessment of the academic literature selected for full
text review (see sections below for information on this process) for gaps, the team conducted a
targeted search of the retrieved literature on the BRICS countries using the EndNote search
function. The terms used in these searches are in Annex A, but broadly covered the following
topics: rural electrification, poverty alleviation, solar home systems, biogas digesters, and
decentralised energy. These terms were selected because they were believed to cover
interventions and programmes which were likely to indicate clear lessons learned and messages
about ‘what works and why’ in the BRICS countries. The papers identified in this way were
added to the review and evaluated according to the review criteria.
Study inclusion criteria
Inclusion criteria were defined and used to determine which studies were included for further
review out of those retrieved in the searches (see section 4.1 for totals). Inclusion criteria were
organised by: relevance to the subject of the review; by the types of intervention; by the types of
comparators in the studies; by the types of outcomes discussed (effectiveness); by the types of the
study; and by geographical relevance. Further detail about each type of criterion and the review
stage (title, abstract or full text) is given below.
Two sources of literature were used: peer-reviewed academic articles and grey literature. The
inclusion criteria which were applied were the same for both sources, although we first focussed
on academic literature as this set of papers was more likely to include highly relevant and
methodologically good quality articles4. The results from the inclusion assessment of academic
literature allowed us to gain an initial understanding of the types of barriers present and
interventions implemented to increase the use of modern energy services. We then applied the
inclusion criteria to the grey literature, focusing in particular on the lessons learnt from
interventions implemented to overcome barriers to increased use of modern energy services. We
adopted this focus here as we thought the grey literature was likely to have particular relevance to
lessons learned about interventions, for example through evaluations of major World Bank
programmes. However, it was also introduced due to resource constraints and the need to narrow
the scope of the initial body of grey literature which was returned. For this reason, we also
introduced two other inclusion criteria for this set of literature: i) studies published after 2005 and
ii) studies where the geographic focus was Sub-Saharan Africa. Further information about the
quality review criteria for the grey literature are discussed in Section 3.4 below.
The full set of articles was first reviewed by title according to the first two broad categories of
inclusion criteria below (relevance by subject and geographic coverage). The criteria were
applied in a tiered fashion, that is, if an article passes the first stage of relevance of subject, we
then applied the criteria for geographic coverage. However, many papers had ambiguous titles. If
there was no direct information about barriers, interventions or geographic coverage in the title,
we erred on the side of caution and included the article for further abstract review.
3.2.1 By relevance of the subject(s)
The first inclusion criterion applied was relevance to the main subject of the review. This
criterion was applied at the title review stage. If the article did not meet these basic relevance
criteria, and there was no ambiguity in the title leading it to be included out of caution, it was
immediately excluded from the review. These criteria were:
Discussion of modern energy services or energy technologies
Barriers to using energy services (including but not limited to those listed in List 1a and b
in Appendix A), or interventions to using energy services (including but not limited to
those listed in List 3 in Appendix A)
Population, or subject of the review, as including ‘the world’s poorest people’. Only
studies which examine the situation in at least one “developing country” as defined by the
World Bank (WDI 2009) were included. At the title stage inclusion assessment both low
and middle income countries were considered and then the geographic relevance criteria
described below were applied.
3.2.2 Geographic coverage
Each article was reviewed at the title stage for geographic relevance. If it was not possible to
deduce geographic coverage at the title review stage, the article was included for abstract review,
at which point it was feasible to deduce geographic coverage of the article according to the
criteria discussed below.
The team recognised it was possible that barriers to uptake of modern energy services, and
interventions to overcome them, have been studied in one part of the developing world but not
another. Although the review analysis focuses on sub-Saharan Africa, this is not to imply there is
not value to be gained by including evidence from Asia and Latin or South America, or even
The authors recognise that publication in an academically peer-reviewed journal is not an automatic indication of relevance
or quality, and for this reason the reader will note that a detailed quality review stage was carried out and is described in
Section 3.4.
articles looking at pockets of extreme poverty in higher-income countries. However, as discussed
above, due to resource constraints a full geographic review which would have allowed for this
comparison was not possible. The team did, though, conduct a targeted search of the BRICS
literature which enabled some comparative analysis of interventions and lessons learned.
3.2.3 By types of intervention
The intervention had to be a public policy intervention or type of involvement to encourage or
address increased uptake of energy services. If at least one of the intervention(s) discussed in the
article was a public intervention, then the article was selected for inclusion in the full text review.
‘Public intervention’ was taken to mean an activity undertaken by an actor (or actors) not
working through normal market means. This does not mean that private actors were precluded.
Private actors may well be involved, but it was expected that there would be some kind of public
dimension to the activity: for example, donor-support, NGO involvement or government role.
The rationale for this criterion was that if normal market means are able to deliver modern energy
services then there is no need for an intervention to address a ‘market failure’, and so the study
will have little relevance to the needs of this review.
3.2.4 By types of comparator
Each article was reviewed for inclusion of the following comparators. If one of the comparators
below was listed, the article was selected for full text review. The comparators used to include
studies were:
Use of modern energy services before/after an intervention to target barriers to use: This
was the most obvious comparator to select and the presumption here was that the
intervention successfully targets the intended barrier (or barriers) and any change in
outcome can be attributed to the intervention. However, it may not have captured
compounding factors (e.g. changes in the broader context that might have had more impact
than the intervention), and it would not have necessarily captured technology substitution
(e.g. if the energy service is new to the user).
Use (and associated barriers to use) of modern energy services/technologies versus prior
use of traditional fuels: This comparator captured technology substitution.
Use of modern energy services versus use of traditional fuels used at the same time within a
geographical region/community: This could have revealed something about other factors
such as individual decision making, or revealed something about the relative importance of
different barriers/interventions.
Analysis of different communities: This could have revealed something about factors in the
contexts of different (localised) cultures.
Analysis across different time periods: Different time periods are likely to consist of more
differentiated contextual factors and so could have revealed the relative importance of
context versus barriers/interventions.
Use of two or more different types of modern energy services/technologies: There could be
two or more modern energy services/technologies being used within the same household; or
different modern energy services/technologies used across different households; or the same
modern energy service being realised with a range of technologies either within the same
household or across different households. In all cases, the relative importance of barriers and
interventions could be different across these combinations.
3.2.5 By types of outcomes
Studies were not excluded from the review based on the types of outcomes. The review question
was interested in studies looking at barriers to the use of modern energy services, and in the
effectiveness of interventions to address barriers. As such the team was interested in two
outcomes: first, use of modern energy services, and second, effectiveness of interventions. Thus,
the team have looked primarily for studies that included some type of an indicator of the
type/level of energy services being used and evidence of increased use. The analysis explored the
robustness and variation across different types of indicators and drew conclusions about what and
how they are conceived of, how this varies in different contexts, and the implications for
judgements about effectiveness.
3.2.6 By types of study
Articles or reports were not excluded from the review on the basis of study type, and both
quantitative and qualitative studies were included in the review. For quantitative analysis, only
studies that have controlled for time periods and geographic location were selected. For
qualitative studies, this control criterion was problematic. In such cases, the review team assessed
whether the analysis was appropriately sensitive to contextual factors and the extent to which
lessons could be translated to other contexts. These criteria helped ensure that only the highest
quality quantitative and qualitative research was selected for inclusion.
3.2.7 Kappa test(s) for consistency of decision regarding inclusion/exclusion, at title,
abstract, and full-text level
After the first review of titles, two reviewers reviewed the abstracts of all accepted articles and
made decisions about inclusion or exclusion for consideration for full text review. In both stages,
reviewers kept a record of whether articles were accepted or rejected on the basis of just the title
or both title and abstract. Reasons for acceptance or rejection based on the abstract were recorded
in a template.
Prior to the full review of all the articles retrieved, (of which there were 22,957 as shown in
Figure 1 below), a Kappa test was performed to ensure consistency between reviewers. This
involved two members of the review team making independent decisions on accepting/rejecting a
small sample of 500 of the relevant articles. This number was based on the recommended sample
size, which is a minimum of 10% of the full list, up to a maximum of 500 references (CEE
guidelines, p. 37). The Kappa test was passed with a level of 0.55 which indicated a moderate
level of agreement between the reviewers and after discussing and resolving reasons for
discrepancy, the team was able to proceed to reviewing all titles and abstracts independently.
3.3 Potential effect modifiers and reasons for heterogeneity
The majority of studies were qualitative in nature and heterogeneity of study type and output
prevailed as can been seen in the analysis sections 4.1-4.3 and as discussed in more detail in
Section 5. Contextual factors of each study were considered as these may have had an impact on
outcomes or findings of the study (see Section 5 below for a discussion). Difficulties that may
have arisen were anticipated because of many factors in study design and analytical approach. In
quantitative research, these problems are usually discussed in terms of reliability and validity
(both internal and external). These terms are less appropriate for qualitative research; instead
‘equivalents’ used were dependability/auditability in place of reliability, credibility/authenticity
in place of internal validity, and transferability/fittingness in place of external validity (Miles and
Huberman 1994; Lewis and Ritchie 2003). There were, indeed, many possible ways to attempt to
address these difficulties and many of them are covered in more concrete terms in the next
section on how study quality was assessed.
3.4 Study quality assessment
In order to determine which papers were given a greater emphasis in the full text analysis and
final data synthesis, each of the selected papers was assessed according to their topical and
methodological relevance, as well as for their quality and potential risk of bias. This process
proceeded in two stages. First a rapid quality assessment was made of the paper for basic quality
criteria (listed below). If any of these were not met, the paper was excluded. After this, the paper
was then subject to a more detailed quality assessment and data extraction process whereby the
information and data in the text was systematically reviewed and summarised in a separate data
template. This allowed the reviewer to determine the quality of the different types of data,
analysis and conclusions. A full list of the questions the reviewer considered in this process is
given in Appendix C.
The same quality criteria were applied to the grey literature, along with additional criteria
concerning date of publication (after 2005) and geographic scope (Sub-Saharan Africa only). In
addition, we only included papers which had a specific focus on interventions as this would
provide information on barriers and would address pragmatic solutions to overcome them.
However, it is important to note that the grey literature often omitted methodological information,
and therefore it was difficult to assess the quality of such papers. Since methodological
information and clarity about data collection is a crucial indicator of quality, particularly for
qualitative papers, when this information was not available in a grey literature paper we
automatically excluded it. Further comments about the implications of this are made in the
discussion section below (Section 5).
Rapid quality assessment
An initial set of criteria was used for a rapid quality assessment of the paper. If any of the criteria
were not met, the paper was automatically excluded from the full text review. These criteria were:
Does the paper describe the study design?
Does the paper describe the form of original data?
Does the paper describe the barriers/interventions in the context of at least one
Does the paper have a clear methodology, including information/data collection
methods, analysis and conclusions?
Does the paper define the focus of the study/population or community or region?
If the initial, rapid quality assessment was passed, the paper was further assessed for topical and
methodological relevance.
Topical relevance
To assess topical relevance, two sets of criteria were used, one for studies focussing on the
identification of barriers, and the other for studies examining intervention effectiveness. Some of
the papers fell into both categories, in which case both sets of criteria applied. The two sets of
criteria are as follows:
For studies which focus on the identification of barriers
The first half of the primary question was “what are the major barriers to increased use of
modern energy services among the world’s poorest people?” The keyword here was “major”,
and the challenge here was to separate the important barriers from the less important ones.
While it was difficult to objectively determine what qualifies as ‘major’, the team did require
that in order for a barrier to be identified and included in our synthesis, the barrier needed to
be clearly supported by evidence in the paper and, ideally, linked to analysis of an
intervention. Studies which simply listed barriers without providing the reasons why they are
presented, or how they have been identified, were less useful in contributing to the
conclusions on “major barriers” and were not extracted for analysis.
For studies that examine intervention effectiveness
The second half of the primary question was “are interventions to overcome these effective?”
This was a question about whether the policies, programmes or other interventions have
achieved the intended outcomes. A highly relevant study for addressing this question had to
include the following elements:
o a measurable and appropriate indicator of outcome
o a measurement of how the outcome has changed
o a baseline/counterfactual against which the change in outcome can be measured
We exercised our expert judgement on what an appropriate outcome indicator would be, as
they are often open to debate. The fact that baselines/counterfactuals are difficult to define
with qualitative studies was also taken into account. That they are often contentious anyway,
and that many studies in this field are good and rigorous but do not necessarily include
counterfactuals was noted. Therefore, the criteria list above was only used as a guide rather
than a set of rules.
Methodological relevance
To assess methodological relevance, quantitative and qualitative evidence was appraised by
looking at methodological quality and relevance and topic relevance, with specific criteria
and questions within these broad assessment categories to fit the nature of the evidence. For
quantitative studies (which we expect would represent only a small proportion of the relevant
papers), their methodological rigour was assessed by considering their study design,
including sample size, use of control variables, methods of measuring and assessing different
variables, and potential sources of bias.
For qualitative studies, the factors which contribute to methodological rigour were not so
easily identified because of the diversity of disciplinary approaches present in qualitative
research. However, the team adapted and refined the recommendations in the Weight of
Evidence Framework (Gough 2007) published by the Evidence for Policy and Practice
Information and Co-ordinating Centre (EPPI-Centre) and used in previously developed
RAND frameworks used in rapid evidence reviews5 to define a list of core quality criteria.
Once papers were evaluated as having both topical and methodological relevance, the data
extraction phase began (described below). During this reading a further set of quality criteria
were considered and used to make a final judgement about the quality of the paper. Low
quality papers were not included in the final analysis, and medium papers were used where
there were particularly well-evidenced points about barriers or interventions, even though the
paper as a whole may have been of lesser quality. All high quality papers were included in the
final analysis.
These frameworks are not publicly available, but are based on EPPI Centre and Cabinet Office rapid evidence assessment
The full set of quality criteria included several criteria for review and a full list is provided in
Appendix C. However, they fell under the main headings of: credibility of findings; breadth
and depth of study findings; extent to which the study addresses the original aims and
purposes; defensibility of study design; defensibility of sample design and coverage; quality
of data collection; quality of the approach and formulation of the analysis; clarity and
coherence of reporting; clarity of assumptions that have shaped the form and output of the
3.5 Data extraction strategy
Information management software (EndNote) helped ensure transparency and replicability
throughout the review. This was used to record bibliographic information for the studies retrieved
during the searches. Descriptive information was recorded for each paper in a separate Excel
database, the template for which was:
Full bibliographical reference
Publication type (peer review journal article, institution working paper, etc.)
Study design
Country or region of the study
Type of energy service or technology analysed
Type of barriers assessed
Characteristics of community assessed (including the reason why the community was
selected and how the study subjects were recruited)
Time period
Sample size and characteristics
Outcome/s or type of intervention under investigation
Effectiveness measures (of interventions to overcome barriers) discussed
Findings (quantitative and qualitative).
Biases, effect modifiers/confounding variables identified and/or measured by authors
(including in their discussion if a posteriori to acknowledge any flaws in their results)
3.6 Data synthesis and presentation
The synthesis is presented as a structured empirical narrative alongside several mapping and
summary tables (presenting descriptive details of each study included in the review). A statistical
meta-analysis was ruled out as it was anticipated that the majority of studies reviewed would be
The synthesis is structured according to the sub-research questions that fall out of the main
review question, as indicated in section 2. Within this structure, the secondary questions given in
section 2 were used to guide the analysis and develop the synthesis. This approach promoted
transparency and the possibility for repeatability.
As evidence was reviewed, the strength of the conclusions that could be derived from it based on
assessments of the methodologies employed to gather and analyse that evidence were discussed.
The range of comparators included supported these assessments further by enabling a discussion
of any effect modifiers, such as contextual factors, and the extent of their influence over
outcomes. These were dependent on whether there are sufficient studies to reveal these factors.
Intervention outcomes were kept purposely broad and inclusive in the search stage of the review;
the synthesis appraises the scope of outcomes analysed in the evidence, and considers any
outcome indicators potentially neglected. To the extent possible, we discuss variability in
outcomes and how such variability might be explained according to different barriers, types of
modern energy services, or contextual factors. Finally, the synthesis addresses gaps in the
evidence base on the systematic review question.
3.6.1 Selection of studies and outcome data for synthesis
Studies were selected for synthesis based on the criteria for study quality and relevance as
described in earlier sections. All studies selected are summarised in Section 4.2.
As previously mentioned, the definition of effectiveness was broad, and the potential range of
outcomes of primary studies was thus large. However, not all outcome data from primary studies
were relevant for analysis in this systematic review. Outcome data synthesised in the review only
includes those which specifically address how the interventions increase uptake of modern energy
services and technologies. This includes both quantitative and qualitative findings, with
qualitative findings including, for example, narrative subject reports of outcome differentials of a
given intervention effectiveness indicator.
3.6.2 Process used to combine/synthesise data
The synthesis of data was guided by the following key questions:
What is the overall evidence on the barriers to uptake of modern energy services in the
developing world?
What is the evidence on different types of intervention providing differential outcomes?
Following from the steer provided by these two questions, the sub-research and secondary
research questions were used to focus the analysis, as has been discussed in greater detail in
earlier sections. However, it is worth nothing that one hypothesis of the authors was that the
evidence on barriers to the use of modern energy services among the world’s poorest people, and
interventions to increase uptake, has not been gathered or analysed in an integrative way.
Therefore, a fully integrative synthesis of the evidence of barriers and their interactions, as well
as interventions and their impacts, was problematic since such a synthesis would require different
methodological approaches to those used in the studies themselves. Therefore, an integrated
synthesis of barriers and interventions was not possible within the limited resource constraints of
this review.
3.6.3 Deriving conclusions and implications
Implications and conclusions were derived from the synthesis of findings from the review. Within
the review team, it was considered how the question fits within broader development strategies
and the key literature identified through the search strategy on modern energy services access and
uptake. In the study design, the team has chosen to include a wide range of barriers and
interventions, including evidence on economic, technical, and cultural barriers and their
interaction in different local contexts. Necessarily, then, the conclusions and implications are
emergent and context-dependent.
4. Results
4.1 Review descriptive statistics, systematic map
A systematic study search was undertaken based on the protocol set out in previous sections. This
resulted in 22,957 articles from the academic database and 1,425 articles from the grey literature
database. A title and abstract screening and relevance assessment were then undertaken
identifying 130 relevant articles from the academic literature and 15 from grey sources for full
text review. In addition to these 145, there were 1,039 academic and 173 grey articles identified
that were not directly relevant to the review and so were set aside as low priority. The purpose of
doing this was to be able to return to these if needed and resources permitted. In the event, these
were not examined further and so were finally excluded as a low priority. After a detailed review
of quality and relevance of the 145 articles, 41 were included in the final synthesis (29 peerreviewed academic papers about sub-Saharan Africa, 6 grey literature articles and 6 peerreviewed academic papers about BRIC countries). The results from each stage of the review are
summarised in Figure 1.
Figure 1: Results of each stage of the review (academic literature and grey literature)
22,957 from academic databases
1,425 from grey literature databases
excluded after title and abstract
excluded after title and abstract
excluded as low priority papers
excluded as low priority papers
130 for full text review
(124 on sub-Saharan African
countries and 6 on BRIC countries)
35 for inclusion in synthesis
(29 on sub-Saharan African
countries and 6 on BRIC countries)
15 for full text review
10 excluded because of no access
0 excluded because of no access
85 excluded after full text review
9 excluded after full text review
6 for inclusion in synthesis
The rest of the descriptive statistics focuses on the 41 highly relevant articles that are of
high/medium quality. Table 1 shows the different types of modern energy service or technology
examined in this set of articles. As shown, electricity/electrification is the most commonly
discussed modern energy service, followed by solar or photovoltaic. It is also interesting to note
that they are often examined alongside one another in the same paper.
Table 1: Type of energy service or technology analysed
Number of articles
Grey literature on
Type of energy service or technology analysed
academic papers
on sub-Saharan
Electricity or electrification
Battery or battery charging
Heat or thermal comfort
Lighting or illumination
Mobile phone charging
Hydro power
Biofuel, biogas, bioethanol or waste
Solar or photovoltaic
Wind power
Gas turbine
Nuclear power
Fuel cell
Improving energy efficiency
Sustainable/renewable/modern energy services in general
Note: total count would not equal the number of studies reviewed, as some papers look at multiple energy
academic papers
on BRIC countries
The barriers described in this set of papers were also examined. Based on a close examination of
the data extracted by individual reviewers, five key themes/categories were drawn out:
financial/cost, cultural, technical, institutional/regulatory, and capacity (skill). The barrier that is
featured most prominently in the articles is financial/cost barrier (see Table 2). These categories
will be defined and discussed in greater detail in the section 4.3.
Table 2: Types of barriers assessed
Number of articles
Type of barriers assessed
Capacity (skill)
Macro socio/demographic factors (geographical dispersal and
academic papers
on sub-Saharan
Grey literature on
academic papers
on BRIC countries
The main types of interventions are: financial (e.g. subsidy, grant or other forms of assistance),
technical assistance, and capacity building (see Table 3). Financial interventions are slightly more
commonly discussed in articles, but not by much. There are a few other interventions that were
not grouped under any of the headings in Table 3, including: voluntary actions by companies,
government intervention, ESCO project-a service, 'market oriented' approach, infrastructure
provision, and project implementation. Additionally, it should be noted that some papers did not
discuss interventions at all (12 out of 28 academic papers).
Table 3: Types of intervention under investigation.
Number of articles
academic papers
on sub-Saharan
Type of intervention under investigation
Financial subsidy/grant/assistance
Institutional/market reform
Technical assistance
Capacity building
Grey literature on
academic papers
on BRIC countries
Finally, the distribution by country within sub-Saharan Africa was examined (Table 4). There is a
high concentration of papers in Kenya, Zimbabwe, and eastern and southern African countries in
Table 4: Distribution by country
academic papers
Burkina Faso
Eastern and southern Africa in general
South Africa
Sub-Saharan Africa in general
Grey literature
4.2 Narrative synthesis including study quality assessment
For each included study a summary of quality and findings along with other key information such
as barriers, interventions and effectiveness measures examined is provided. Before the summaries
of individual papers, Table 5 below provides the full bibliographical reference, an overall rating
of quality and the type of publication. This set covers 41 articles in total, with 35 from academic
literature (including the 6 peer-reviewed academic papers about BRIC countries) and 6 from grey
Table 5: Full bibliographical reference of the included papers and their overall quality rating
Full bibliographical reference
Acker RH, Kammen DM: The Quiet (Energy) Revolution: Analysing the
Dissemination of Photovoltaic Power Systems in Kenya. Energy Policy
1996, 24:81-111.
Adkins E, Eapen S, Kaluwile F, Nair G, Modi V: Off-grid Energy Services
for the Poor: Introducing LED Lighting in the Millennium Villages
Project in Malawi. Energy Policy 2010, 38:1087-1097.
African Development Bank Group: Tanzania Rural Electrification Project:
Project Performance Evaluation Report (PPER). Tanzania Rural
Electrification Project: Project Performance Evaluation Report (PPER)
African Development Bank Group; 1996.
Campbell B, Vermeulen S, Mangono J, Mabugu R: The Energy Transition in
Action: Urban Domestic Fuel Choices in a Changing Zimbabwe. Energy
Policy 2003, 31:553-562.
Clough L: Marketing Challenges and Strategies for Micro and Small
Energy Enterprises in East Africa. In Book Marketing Challenges and
Strategies for Micro and Small Energy Enterprises in East Africa, GVEP
International; 2011.
Collings S: Phone Charging Micro-businesses in Tanzania and Uganda.
Phone Charging Micro-businesses in Tanzania and Uganda, GVEP
International; 2011.
D'Agostino AL, Sovacool BK, Bambawale MJ: And Then What Happened?
A Retrospective Appraisal of China's Renewable Energy Development
Project (REDP). Renewable Energy, 2011. 36(11): 3154-3165.
Diniz ASAC, Machado Neto LVB, Camara CF, Morais P, Cabral CVT,
Oliveira Filho D, Ravinetti RF, Franc ED, Cassini DA, Souza MEM, Santos,
JH, Amorim M: Review of the Photovoltaic Energy Program in the State of
Minas Gerais, Brazil. Renewable & Sustainable Energy Reviews, 2011. 15(6):
Dube I: Impact of Energy Subsidies on Energy Consumption and Supply in
Zimbabwe. Do the Urban Poor Really Benefit? Energy Policy 2003,
Duke RD, Jacobson A, Kammen DM: Photovoltaic Module Quality in the
Kenyan Solar Home Systems Market. Energy Policy 2002, 30:477-499.
Foster V, Steinbuks J: Paying the Price for Unreliable Power Supplies: InHouse Generation of Electricity by Firms in Africa. Policy Research
Working Paper 4913. World Bank: African Sustainable Development Front
Office; 2009
Gaunt CT: Meeting Electrification's Social Objectives in South Africa, and
Implications for Developing Countries. Energy Policy, 2005. 33(10): 13091317.
Green JM, Wilson M, Cawood W: Maphephethe Rural Electrification
(Photovoltaic) Programme: The Constraints on the Adoption of Solar
Home Systems. Development Southern Africa 2001, 18:19-30.
Green JM, Erskine SH: Solar (Photovoltaic) Systems, Energy Use and
Business Activities in Maphephethe, KwaZulu-Natal. Development
Southern Africa 1999, 16:221-237.
Guimarães L, Caracaleanu C, Sy B, N'Dongo A, Sankaré O: Energy
Affordability in the Sahelian Region. Applied Energy 2003, 76:9-13.
Gumbo R, Katsvairo L, Asai K: State of Photovoltaic Home Systems in
Zimbabwe. IEEE; 2003, 2643:2640-2643
Gustavsson M, Ellegård A: The Impact of Solar Home Systems on Rural
Livelihoods. Experiences from the Nyimba Energy Service Company in
Zambia. Renewable Energy 2004, 29:1059-1072.
Haanyika CM: Rural Electrification in Zambia: A Policy and Institutional
Analysis. Energy Policy 2008, 36:1044-1058.
Hofmeyr IM: Towards Adequate and Equitable Energy Provision for
Farmworker Families. Development Southern Africa 1995, 12:273-279.
Ilskog E, Kjellström B, Gullberg M, Katyega M, Chambala W: Electrification
Co-Operatives Bring New Light to Rural Tanzania. Energy Policy 2005,
rating of
Type of
Acad. literature
Acad. literature
Grey literature
Acad. Literature
Grey literature
Grey literature
Acad. Literature
Acad. Literature
Acad. Literature
Acad. Literature
Grey literature
Acad. Literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Karekezi S, Kimani J: Status of Power Sector Reform in Africa: Impact on
the Poor. Energy Policy 2002, 30:923-945.
Karekezi S, Kithyoma W: Renewable Energy Strategies for Rural Africa: Is
a PV-Led Renewable Energy Strategy the Right Approach for Providing
Modern Energy to the Rural Poor of Sub-Saharan Africa? Energy Policy
2002, 30:1071-1086.
Ketlogetswe C, Mothudi T: Solar Home Systems in Botswana-Opportunities and Constraints. Renewable and Sustainable Energy Reviews
2009, 13:1675-1678.
Kirubi C, Jacobson A, Kammen DM, Mills A: Community-Based Electric
Micro-Grids can Contribute to Rural Development: Evidence From
Kenya. World Development 2009, 37:1208-1221.
Lee KS, Anas A, Oh G-T: Costs of Infrastructure Deficiencies for
Manufacturing in Nigerian, Indonesian and Thai Cities. Urban Studies,
1999. 36(12): 2135-49.
Mapako M, Network BU: Provision of Long-term Maintenance Support for
Solar Photovoltaic Systems - Lessons from a Zimbabwean NGO. Journal
of Energy in Southern Africa 2005, 16:21-26.
Mulder P, Tembe J: Rural Electrification in an Imperfect World: A Case
Study From Mozambique. Energy Policy 2008, 36:2785-2794.
Mulugetta Y, Nhete T, Jackson T: Photovoltaics in Zimbabwe: Lessons
From the GEF Solar Project. Energy Policy 2000, 28:1069-1080.
Murphy JT: Making the Energy Transition in Rural East Africa: Is
Leapfrogging an Alternative? Technological Forecasting and Social Change
2001, 68:173-193.
Nygaard I: Institutional Options for Rural Energy Access: Exploring The
Concept of the Multifunctional Platform In West Africa. Energy Policy,
2010. 38(2): 1192-1201.
Peng WY. Pan JH: Rural Electrification in China: History And Institution.
China & World Economy, 2006. 14(1): 71-84.
Pereira MG, Freitas MAV, da Silva NF: Rural Electrification and Energy
Poverty: Empirical Evidences From Brazil. Renewable & Sustainable
Energy Reviews, 2010. 14(4):1229-1240.
Rehman IH, Kar A, Raven R, Singh D, Tiwari J, Jha R, Sinha PK, Mirza A:
Rural Energy Transitions In Developing Countries: A Case of the Uttam
Urja Initiative In India. Environmental Science & Policy, 2010. 13(4): 303311.
Schut M, Slingerland M, Locke A: Biofuel Developments in Mozambique.
Update and Analysis of Policy, Potential and Reality. Energy Policy 2010,
Steel K: Dynamics of Growth and Investment in the Kenyan Electric
Power Sector, Power Engineering Society General Meeting, 2007. IEEE,,
pp.1-5, 24-28 June 2007 doi: 10.1109/PES.2007.385906
Thiam DR: An Energy Pricing Scheme for the Diffusion of Decentralized
Renewable Technology Investment in Developing Countries. Energy Policy
2011: 39(7). 4284-4297
van der Plas RJ, Hankins M: Solar Electricity in Africa: A Reality. Energy
Policy 1998, 26:295-305.
Walekhwa PN, Mugisha J, Drake L: Biogas Energy From Family-Sized
Digesters in Uganda: Critical Factors and Policy Implications. Energy
Policy 2009, 37:2754-2762.
Winkler H, Spalding-Fecher R, Tyani L, Matibe K: Cost-Benefit Analysis of
Energy Efficiency in Urban Low-Cost Housing. Development Southern
Africa 2002, 19:593-614.
Wolde-Ghiorgis W: Renewable Energy for Rural Development in Ethiopia:
The Case for New Energy Policies and Institutional Reform. Energy Policy
2002, 30:1095-1105.
The World Bank Independent Evaluation Group: The Welfare Impact of
Rural Electrification: A Reassessment of the Costs and Benefits. World
Bank; 2008
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Grey literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Acad. literature
Grey literature
The summaries below provide further detail on the papers in Table 5 above. The summaries are
structured as follows: full bibliographic reference of the paper; indication of paper quality
assessment; short summary of the paper covering analytical focus or topic, methods used or
sources of data, main findings focusing on barriers or interventions and conclusions.
Acker RH, Kammen DM: The quiet (energy) revolution: analysing the dissemination of
photovoltaic power systems in Kenya. Energy Policy 1996, 24:81-111.
Quality: high
Summary: This paper analyses the growth of the SHS market in Kenya and the performance of
systems in the field. The study focuses on small-scale PV systems ranging from 10 to greater
than 100 watt-peak (Wp), within a range of income levels and income sources amongst users.
Technical, financial and capacity (skills, knowledge, etc.) aspects of SHSs, the market and
potential adopters are investigated using a case-study approach. The approach included
interviews with 40 SHS adopters and a few local experts, as well as a set of measurements of the
technical performance of the 40 systems investigated. The field research was conducted during
July and August 1994, and the systems were clustered around Nakuru, Meru and Bungoma. The
paper also provides a detailed contextual discussion of the Kenyan economy, energy profile,
energy policy environment and the history of the SHS market. A range of variables is reported in
the paper: system size, age, technical condition, cost of system components, appliances used and
distance from the grid at time of purchase; dispersion of systems across income groups; systemcost compared with user-income; channel through which the adopter became aware of SHSs;
perceived benefits and disbenefits of SHSs.
Based on the empirical analysis, the paper offers many recommendations for policy, which are
categorised under small-scale approaches, financing schemes, capacity building (on both supply
and demand sides), and domestic production of system components.
Small-scale approaches (promotion of solar lanterns and PV battery-charging stations) would
respond to the market demand for small systems and extend services to poorer users. Financing
would help overcome the high initial cost barrier, and examples are given from experience
elsewhere (Bangladesh, Dominican Republic, Mexico, as well as two Kenyan dealers). Capacity
building would include education and training on the supply side, and information and advice for
potential adopters. Domestic production could begin with encapsulating solar cells while
developing the skills necessary to begin manufacturing the cells, with the hope that this would
lower costs to the consumer. Again, examples are given from elsewhere: Dominican Republic,
Zimbabwe, China and India.
Adkins E, Eapen S, Kaluwile F, Nair G, Modi V: Off-grid energy services for the poor:
Introducing LED lighting in the Millennium Villages Project in Malawi. Energy Policy 2010,
Quality: high
Summary: The study examined a pilot programme which aimed to introduce solar light-emitting
diode (LED) lanterns in Malawi. The program was market-based, relying on local vendors and
cooperatives to sell lanterns to villagers without subsidy.
The study surveyed lantern buyers as well as non-buyers (54 and 43 households respectively) to
investigate the factors underlying households’ decisions to buy LED lanterns. It found that cost
was a determining reason, based on the following observations:
Households that purchased the lanterns early on reported significantly higher incomes and
greater ownership of assets than those who did not
Non-buyers reported monthly incomes that were about half of the price of the lantern
Feedback from non-buyers further confirmed cost was a determining factor (although the
number of non-buyers who provided this feedback was not clear from the article).
Additionally, the authors investigated on the impacts of LED use. They found that lantern
purchases had a positive effect on households’ lighting use and expenditure. Households obtained
longer hours of light (from 2.7 hours per day from kerosene in the week before lantern purchase,
compared with 4.4 hours per day from the LED lantern in the week after the purchase, i.e. a 63%
increase). At the same time, households enjoyed reduced lighting expenditure (an annualised
saving of $47.06 on average, i.e. enough to cover the cost of the lantern ($29.87) after the first
The study also offers some preliminary results on whether the lanterns had an impact on income
generation. When asked “do you see the lantern providing opportunity for economic
development”, 53 out of 54 respondents replied “yes”, with a few (less than 10%) offering the
explanation that the lanterns provided extended business opportunities by allowing more time to
work at night.
Finally, the paper also identified lessons learned about programme implementation. It is found
that community engagement was crucial. Early vendor interviews allowed the programme
implementation team to assess vendors’ willingness and capacity to sell LED lanterns. Vendors
were educated about the benefits of LED lighting and discussed the proposed business models
(e.g. single payment vs. instalments). Community interactions informed villagers of the technical
features of the LED lanterns, the expected savings, and the business model and process, e.g.
African Development Bank Group: Tanzania Rural Electrification Project: Project
Performance Evaluation Report (PPER). Tanzania Rural Electrification Project: Project
Performance Evaluation Report (PPER) African Development Bank Group; 1996.
Quality: medium
Summary: This paper is a project completion report on an electrification project in the Arusha
and Kilimanjaro regions in Tanzania, executed between 1982 and 1989. The project financed the
foreign exchange costs of expanding the national high voltage grid to facilitate transmission of
cheaper hydroelectricity, medium voltage distribution systems and rural electrification networks
in certain regions.
The report follows the Bank Group’s guidelines and regulations to assess the effectiveness of
project implementation through a number of indicators and measures such as: rate of
electrification; institutional performance (electric power sector reform, tariff policy and structure,
accounting and budgeting, auditing, billing and collection); management and organisational
effectiveness (staff recruitment, training and development); performance of consultants and
contractors; economic performance; financial performance; social impact and environmental
impact. The review also looked at project sustainability, including the restructuring of TANESCO
(Tanzania Electricity Supply Company), increased investment in the power sub-sector, operations
and maintenance of the electricity systems and power system losses.
The report highlights several lessons learnt from the project, which are mainly referred to as
administrative, procedural and coordination issues: closer co-ordination and synchronisation by
the Bank of activities with those of other donors would have reduced the delay in loan
effectiveness that occurred on this project; delays during implementation would have been
reduced had the Borrower been better appraised of the Bank’s procurement rules and procedures;
the Bank ought to have ensured that the observed lapses in the audit reports were rectified; more
rigorous examination of loan conditions and, where necessary, provision of technical assistance,
would ensure appropriateness of the conditions. The report also investigates on the structural
reform processes implemented in the country and recommended that an earlier study and
appreciation of the liberalisation process by the Borrower would have minimised the negative
impact on TANESCO. Finally, it is recognised that a baseline study on socio-economic profile of
the project areas would have provided a benchmark for evaluating the effectiveness of the project
after completion.
Campbell B, Vermeulen S, Mangono J, Mabugu R: The energy transition in action: urban
domestic fuel choices in a changing Zimbabwe. Energy Policy 2003, 31:553-562.
Quality: high
Summary: This paper describes a comparative case study looking at the modern energy transition
from wood to kerosene to electricity in urban Zimbabwe. The research is based on 2 surveys of
communities carried out in 1994 and 1998 to generate quantitative data. The study investigated
the factors determining access to fuel such as income, prices of fuels and appliances changes
during that period. Through the use of a logistic chi-square test the study found that electricity use
increases with income (pp557), but there were no differences in the rate of change of electricity
use over time with income. Access to electricity is a key determinant of whether a household uses
electricity, with only 0-3% of households reporting to have a connection to electricity but not use
it. The study shows that, amongst those without a connection in 1999, 51% cited the price as a
reason, 21% due to non ownership of their home, 18% because they had a preference for
kerosene, and 5% because electricity was not available in their area (pp558). The paper concludes
that government policy in Zimbabwe of subsidising electricity and liquid fuels has enabled
poorest households to use the same fuel, however the policy has failed to keep fuel prices in pace
with inflation, and has favoured wealthier households. ‘A series of foreign exchange crises and
breakdown in law and order have meant that fuel has become far more erratic in both availability
and price’ (pp561).
Clough L: Marketing Challenges and Strategies for Micro and Small Energy Enterprises in
East Africa. Marketing Challenges and Strategies for Micro and Small Energy Enterprises in
East Africa, GVEP International; 2011.
Quality: medium
Summary: This GVEP (Global Village Energy Partnership) report examines the marketing
challenges and strategies for micro and small energy enterprises in East Africa. Fieldwork
research undertaken in Kenya, Tanzania and Uganda, where 37 entrepreneurs and 22 customers
were interviewed was used to inform the report. The study investigates the effectiveness of
promotion means and consumer awareness about modern energy products.
The report found that the greatest proportion of the interviewed entrepreneurs (10 out of the 28
responded, i.e. 36%) suggested word of mouth was their most effective promotion technique.
This was followed by hawking (14%), product displays (10%) and promoting at community
meetings (10%). Only 7% cited linking with a local community based organisation. The reliance
on customer referrals to increase their market means that product quality and price are of crucial
importance; however entrepreneurs also face competition from sub-standard goods that are underpriced. One highlighted exception is phone and battery charging, in which the price of the service
is fairly fixed with little variation. For these businesses location is key and entrepreneurs need to
be located within 5 km of their customers.
Concerning consumers’ familiarity with certain energy products – namely briquettes, improved
cook stoves and solar lanterns – all the 22 consumers interviewed were aware of improved cook
stoves, 73% were aware of solar lanterns and only 45% were aware of briquettes. These
responses varied across the countries visited – no Tanzanian consumers were aware of briquettes,
while there was awareness among Ugandan consumers. The report implied that the level of
awareness correlates with areas where the GVEP International programme Developing Energy
Enterprise Project East Africa (DEEP EA) operated.
The report also elaborates on barriers that prevent customers from purchasing these energy
products and refers to finance and availability as the most often cited reasons.
Although this study is a publication of GVEP, it is not an evaluation of the success or failure of
the GVEP programme; therefore the risk of bias is low.
Collings S: Phone Charging Micro-businesses in Tanzania and Uganda. Phone Charging
Micro-businesses in Tanzania and Uganda, GVEP International; 2011.
Quality: medium
Summary: This GVEP report examines phone charging micro-businesses in Tanzania and
Uganda with the aim to understand (i) the market dynamics and the potential for growth and
possible diversification into sales of solar lanterns and lighting systems, and (ii) the impact that
the Developing Energy Enterprises Project (DEEP) has had on these businesses. The data used in
the study were collected through interviews with 15 phone charging businesses (3 in Uganda and
12 in Tanzania) and 14 customers. As the sample size is very small, this report should be viewed
as a qualitative study for generating in depth insights rather than a characterisation of the overall
phone charging micro-business market. The study’s key findings are summarised in the following:
Phone charging businesses have a high potential to grow, based on the author’s observations that:
In all cases there were more customers than could be serviced. Almost all of the
entrepreneurs interviewed intended to buy another solar panel to expand their service.
Some charging businesses had no immediate competitor (nearest service was more than
3km away). The majority had 1-3 competitors offering similar services in the immediate
None of the entrepreneurs expressed major concerns about competition.
Phone users reported significant economic and social benefits from use of their phones, including:
The benefit of having a service locally, which saves considerable amounts of time and
money. While previously they had travelled considerable distances (typically 7-15km) to
charge a phone, they now have a provider within 3 km of their home. One customer
reported having to spend UGX 6000 (US$2.20) a week on transport in order to charge her
phone – she now spends the money on airtime.
Customers felt that they received higher quality of service (e.g. full charge) and reduced
risk of theft or damage.
The author suggested that a major constraint on growth is lack of access to funds for the
businesses to purchase additional panels and accessories. Most businesses were considering
applying for a loan.
GVEP manages a programme which supports micro-businesses engaged in servicing the energy
needs of poor communities in Kenya, Tanzania and Uganda – Developing Energy Enterprises
Project (DEEP). Although this study is a publication of GVEP, it is not an evaluation of the
success or failure of the GVEP programme; therefore the risk of optimism bias is low.
D'Agostino AL, Sovacool BK, Bambawale MJ: And then what happened? A retrospective
appraisal of China's Renewable Energy Development Project (REDP). Renewable Energy,
2011. 36(11): 3154-3165.
Quality: high
Summary: The paper is a peer-reviewed, retrospective and qualitative analysis of the long-term
impacts of China’s Renewable Energy Development Project (REDP) in the areas of Beijing,
Xining, and Shanghai. Though the REDP aimed to support wind technology improvement and
photovoltaic technologies, this study focuses on evaluating the photovoltaic (PV) component of
the programme. The PV component had the priorities of improving product quality (with
penalties for noncompliance), reducing production costs and installing a total of 10 MWp of SHS
capacity. The REDP was supported by the World Bank and though it had conducted its own
evaluation, this is the first independent study to qualitatively understand the long-term
implications of the programme for all relevant stakeholder groups. The methodology included a
literature review and semi-structured interviews with thirty stakeholders, including project staff,
employees at renewable energy companies, retailers in direct contact with end users, and end
The REDP set up a sub-grant for each SHS unit sold of $1.5/Wp. The scheme also incorporated a
quality assurance component where compliant products were branded with a ‘Golden Sun’ to
indicate high quality. The study found that sales figures for PV companies increased after the
REDP and several companies have won additional contracts for similar projects. However, there
was a concern that the domestic market was now saturated. SHS retailers claimed sales had
increased and that quality was the most important criterion customers used to make decisions.
However, this contradicted feedback from consumers which indicated they made their decision
on price alone. None of the retailers actively promoted products based on certified quality, such
as products which had the ‘Golden Sun’ branding, so the effects of this aspect of the PV
programme were unclear. End-users also reported benefits of SHS ownership which in some
cases contradicted those of the World Bank evaluation and cited different reasons for not
undertaking repair and maintenance of their systems, including distance to shop, and previous
poor experiences with equipment reliability.
Overall, the authors concluded that the following lessons could be learned:
Rural credit constraints have impeded end user and corporate investments in new SHS
stock and the REDP should not have discontinued its credit access programme.
More robust results are needed to make definitive conclusions about the effects of SHS
ownership on poverty and randomised controlled trials are needed.
Improving the quality of after-sales service at township and provincial level would help to
improve user uptake and there is a need for research into which after-sales service models
are most pragmatic.
SHS can be a transition technology while people are waiting for grid connections.
The REDP had a life cycle approach to quality improvement (starting with manufacturing
standards and practices through to product certification and introduced randomised testing
regime which penalised non-compliance at production-line and retail stages). This
contributed to strengthening the quality of SHS at multiple points in the supply chain.
Diniz ASAC, Machado Neto LVB, Camara CF, Morais P, Cabral CVT, Oliveira Filho D,
Ravinetti RF, Franc ED, Cassini DA, Souza MEM, Santos, JH, Amorim M: Review of the
photovoltaic energy program in the state of Minas Gerais, Brazil. Renewable & Sustainable
Energy Reviews, 2011. 15(6): 2696-2706.
Quality: medium
Summary: This paper presents a summary review and an ex post policy evaluation of the
photovoltaic energy programme which was implemented in the state of Minas Gerais, Brazil, as
part of the ‘Light for All’ government programme, whose main objective was to supply electricity
to all communities by 2010. The specific energy programme in Minas Gerais was implemented
by the Energy Company of Minas Gerais (CEMIG) between the 1990s and the late 2000s. The
authors give a detailed background of the context of rural electrification in Brazil and provide a
summary of many of the political, technical and economic factors which have influenced the
development and execution of the Light for All Programme (p. 2696-2700). There were three
alternatives which could be pursued: extension of the distribution grid, autonomous off-grid
generation using micro-grids, and individual electricity supplies. After local research, a solar rural
electrification programme was pursued in Minas Gerais, called the Sunlight Program. It had three
components: a SHSs Sub-programme; Rural School and Community Centres Electrification with
PV systems; and a PV Solar Energy Training Programme. The paper reports on the following
effectiveness measures, which subsequently are presented as key criteria to consider when
developing a solar rural electrification programme and its relative merits over other rural
electrification technologies. These include:
Geographic location: Isolated and remote communities, as well as those on
environmentally protected land should be considered for use of PV or hybrid systems;
Cost analysis: When the cost of electrification of the community is at least double that of
photovoltaic systems, then PV technology is competitive and can be chosen;
Market analysis: potential consumers are small rural producer, or rural producer typical,
and consumers must be categorised as residential low income, agriculturally based and
earn 2 minimum wages per month.
The authors conclude that their experience using these criteria has demonstrated both
technological reliability and electrification cost effectiveness, and so future solar electrification
projects could benefit from using these criteria in ex-ante policy analysis.
Dube I: Impact of energy subsidies on energy consumption and supply in Zimbabwe. Do the
urban poor really benefit? Energy Policy 2003, 31:1635-1645.
Quality: high
Summary: This paper discusses subsidy policies for grid connection in Zimbabwe. Data were
gathered using a survey of members of the community in major cities in Zimbabwe and data from
the Department of Energy on subsidies and energy sources.
The paper examines the effect of subsidies on the affordability of electricity, and the distribution
of subsidies among different urban household income categories and other economic sectors,
measuring if subsidies were decisive for the affordability of electricity by the urban households.
The study found that different household income categories pay for electricity differently, with
the urban poor paying a larger proportion of income than non-poor (pp1639). Furthermore, the
author concludes that ‘the energy costs incurred by the poor on non-electrical energy sources
could cover the current subsidized electricity costs’ (pp1639).’ It was found that urban
households are currently paying much more for kerosene than the stipulated subsidized price.
This means that subsidies are not decisive for the affordability of energy by kerosene users
(pp1641). The existence of electricity subsidies is also not decisive for the affordability of
electricity by the urban household, but the removal of such subsidies will impact more negatively
to the more vulnerable groups than the affluent groups (pp1641). Subsidies are not entirely
decisive for the affordability of energy due to different household wealth and other factors, such
as upfront energy costs (or initial costs) for poor households (pp1642). Finally, regarding factors
that create difficulties in access to energy, the study concludes that ‘the upfront electricity costs,
could be considered, a significant barrier towards the access to electricity by the poor households,
compared to affordability of the recurrent electricity costs even without subsidies’ (pp1642).
Duke RD, Jacobson A, Kammen DM: Photovoltaic module quality in the Kenyan solar home
systems market. Energy Policy 2002, 30:477-499.
Quality: high
Summary: This paper examines the information failure problem of the solar home systems (SHSs)
market in Kenya. The study uses a review of literature to inform the qualitative analysis.
The paper describes PV technology in the country of origin of prevailing brands. Most SHSs
purchased in Kenya since the 1990s use small 10-14 Wp amorphous silicon (a-Si) modules. Two
of the three main a-Si brands marketed in Kenya performed adequately; however, a third brand
with severe quality problems maintained a substantial market share, despite a far higher price per
delivered Wp. The problem was that SHS owners are only able to gauge PV module quality after
purchase, making PV a classic candidate for quality information failure.
The paper found that Kenyan households suffered major financial losses simply because they
have had the misfortune to purchase the wrong brand of PV module. In addition to this direct
harm, the inability of consumers to discern the relative quality of different module brands reduces
overall confidence in PV and creates an important market failure. "Pooled quality" assessment
pushed public perception of module reliability towards the performance level of the worst brand
sold, while unfair competition from low quality brands forced to fulfil these pessimistic
expectations by pressuring better performers to over-rate their modules as well. The authors
conclude that this perverse dynamic constrains the market for SHS below the socially optimal
Foster V, Steinbuks J: Paying the Price for Unreliable Power Supplies: In-House Generation
of Electricity by Firms in Africa. Paying the Price for Unreliable Power Supplies: In-House
Generation of Electricity by Firms in Africa, Policy Research Working Paper 4913. World Bank:
African Sustainable Development Front Office; 2009.
Quality: high
Summary: This study investigates the reasons behind electricity generation by private firms in
sub-Saharan Africa (SSA) covering the period 1990-2006. It explores factors such as the
unreliability of public power supply, regulations of international export markets and firm
characteristics in more than 26 countries in SSA utilising original data from two sources: the UDI
World Electric Power Plants Data Base (WEPP), containing an inventory for 2,843 operating
plants and 941 plants under construction of all sizes and technologies in 47 SSA countries and the
World Bank’s Enterprise Survey Database, which gathers 8,483 operating firms in 25 North and
Sub-Saharan African countries sampled from the universe of registered businesses.
Although the study does not focus on barriers to the use of modern energy services among poor
communities or public interventions (it rather covers investment decisions by private firms), the
results can be clearly linked to factors hindering the effective use of electricity in poor SSA
countries. The paper also estimates costs and benefits of personal electricity generation by private
firms, which can be used as a measure to estimate net benefits compared to public power supply.
The results of the study show the extent to which personal electricity generation by firms
contributes to the national electricity generation mix in the countries studied, which can be used
as an effectiveness measure of energy service provision.
Overall, there are at least 4,000 MW of installed capacity owned by firms in SSA, which accounts
for about 6% of total generation capacity in SSA. This figure rises in low-income countries, postconflict countries, and West Africa (from around 12% on average and more than 20% of total
generation capacity in a few countries). The conclusions of the report demonstrate that the causes
of personal generation are due not only to unreliable public power supply, but also other factors
including: firm characteristics, such as size (the probability of own generation doubles in large
firms relative to small ones), sector (agriculture, construction and tourism are highlighted),
corporate structure, and export orientation of firms (emergency back-up and export regulations
are also crucial factors).
Gaunt CT: Meeting electrification's social objectives in South Africa, and implications for
developing countries. Energy Policy, 2005. 33(10): 1309-1317.
Quality: medium
Summary: This paper presents a literature review and analysis of rural electrification in South
Africa during the period from 1994 to 2005. The sources of data are not always clear, but the
paper does contain useful insights into how rural electrification can be assessed in relation to how
well it meets the social objective of poverty alleviation.
The paper describes interventions to enable rural electrification including funding for grid and
non-grid connections to electrify rural areas; subsidies for the rural poor; and technology change
to meet demands and differences in costs of different types of technologies, in particular PV. The
paper emphasises that effectiveness measures for rural electrification programmes need to address
social objectives such as poverty alleviation, health, education, etc. and that the initial objectives
of rural electrification were too focussed on economic and socio-economic objectives and so were
unable to meet these social objectives. In particular, the author claims that restructuring the
electricity industry with clear social objectives in mind can help meet demands of poverty
alleviation by enabling electrification programmes to have suitable tariffs for different social
classes and can be used to indicate more socially appropriate capital investment decisions. The
author concludes that decision-makers need to be aware of the differences between economic,
social and socio-economic objectives and how the different understandings of these shape
electrification plans and investment decisions.
Green JM, Wilson M, Cawood W: Maphephethe rural electrification (photovoltaic)
programme: the constraints on the adoption of solar home systems. Development Southern
Africa 2001, 18:19-30.
Quality: high
Summary: In this paper the authors examine a pilot programme aiming to investigate and test the
process of introducing solar home systems (SHSs) in Maphephethe (a rural community in South
Africa). The project has a focus on local capacity building, involving the establishment of a
community-based organisation responsible for all aspects of solar energy acquisition in the
community. It is found that the rate of SHS dissemination was slow - only 5 percent of the
community have adopted the new technology after 3 years of project implementation. In
consumer theory terms, the process has not moved beyond the innovators' stage. Various reasons
for the slow dissemination were explored through group discussions held in the community and
during informal conversation throughout the course of the project. The most pertinent reasons
identified were: financial constraints, a lack of experiential knowledge of solar power and a lack
of opportunities to try out the system before making a purchase. The project also had limited
success in building up the capacity of the local community. Since no accreditation is provided for
those who are trained to work with solar technology, and the remuneration for their roles was
poor, the author observed that there exists a danger of losing all locally trained people to more
lucrative employment in urban areas.
Green JM, Erskine SH: Solar (photovoltaic) systems, energy use and business activities in
Maphephethe, KwaZulu‐‐Natal. Development Southern Africa 1999, 16:221-237.
Quality: medium-high
Summary: The paper describes the implementation of a PV SHS project in Maphephethe,
KwaZulu-Natal, led since 1996 by Solar Engineering Services (SES), which is the South African
branch of SELF (a Washington DC-based non-for-profit organisation). Using focus groups as
primary data and secondary sources from a survey carried out by an external organisation (EU)
the study offers an analysis of the factors explaining the success and difficulties in promoting
local participation and a business approach to rural electrification.
The paper elaborates on uptake levels to measure successful diffusion/dissemination of
innovations, based on theoretical contributions from consumer studies literature (Schiffman &
Kanuk, 1987; Grundy & Grundy, 1999). Under the rationale of these authors, it is concluded that
the Maphephethe project has reached very low levels of PV SHS purchase and installations,
accounting for 5% of potential market (all rural households in the area).
Barriers to greater technological diffusion are characterised as financial, cultural, technical and
Financial barriers are characterised by: inability of rural population to access finance and pay
upfront capital, dependency on donors' financial support; insufficient working capital for
manufacturers and distributors and lack of financing options for investors.
Cultural factors associate the expectations of getting access to grid electrification instead of PV
SHS; low awareness levels about PV electrification; increased levels of vandalism and theft and
the perceived status of owning a PV SHS.
Technical limitations are explained as low levels of technical capacity in the community to
operate and maintain SHS and dependency on donors' technical support.
Organisational barriers refer to the limited capacity to implement local management and
maintenance schemes. This is linked to the low uptake of PV systems which leads to financial
struggles of local technicians, who prefer to move to more vibrant markets. Another mention is
that of payment systems that does not fit local needs (people have to travel long distances to pay
monthly bills).
The main findings of the study are:
1. The Maphephethe project aimed at building local capacity for management, maintenance and
overall administration of the project. However, the project has not been able to create the
necessary conditions for a local business to grow and become independent from donors'
2. It is not clear the extent to which marketing and promotional activities have been well
designed so rural inhabitants are encouraged to buy PV SHS. It seems that the project has
devoted more efforts in technical aspects related to the technology rather than engaging a
broad set of actors that can contribute to create the adequate institutional and market
structures that can sustain the development of the project.
Guimarães L, Caracaleanu C, Sy B, N'Dongo A, Sankaré O: Energy affordability in the
Sahelian region. Applied energy 2003, 76:9-13.
Quality: medium
Summary: This review looks at financial interventions to sustainable energy introduction in the
Sahel region, covering the period from 1997 to 2001. It is argued that most measures only look at
the economic and technological aspects of new energy systems, ignoring acceptability, cultural
gaps, education and sustainable implementation. This study focuses on energy affordability and
observes that in order to assess population's willingness and ability to switch energy types, it is
important to look at their present energy expenses and their willingness and ability to pay more.
The study found that affordability of energy depends on income, spending (surveys have shown
that the average household is ready to spend around 10% of its income on energy), indirect costs
of energy (such as deforestation, land degradation, time consumed by energy production), and
energy subsidies (such as fee-for-service-public sector subsidy per household connected, tax
exemptions for renewable energy technologies). The paper concludes that there are measures to
achieve more sustainable energy structures that are more affordable for customers: 1) frameworks
to promote private actors for electrification: legislative and regulatory framework, public subsidy;
2) framework of organisational issues: village cooperatives to coordinate and combine efforts
locally, promotion of productive activities at village level; 3) framework of technical issues:
promotion of improved cooking stoves, building local energy-equipment manufacturing
Gumbo R, Katsvairo L, Asai K: State of photovoltaic home systems in Zimbabwe. IEEE; 2003:
2640-2643 Vol. 2643.
Quality: high
Summary: This conference poster discusses a study looking at various solar delivery strategies in
Zimbabwe. The study gathered data using a structured survey questionnaire, a literature review,
visual observations, measurements of battery characteristics and some in-depth discussions with
owners or users of PV systems. Four different PV implementation projects were compared with
regards to technical barriers to their implementation: 1) the hire purchase approach, (‘The terms
of project involved loan facility from a revolving fund for system purchase at 15% deposit, 15%
interest rate; 3 year payback period; accredited companies doing the installation with limited
back-up for a year’ (pp2641)); 2) the energy service company approach, (low cost and low
maintenance systems installed with a user monthly fee); 3) the do it yourself approach; 4) the
company approach (cash up front system). Comparing the 4 strategies using effectiveness
measures perfectly working, not working, meeting GEF standards, battery failure rate, charge
controller failure rate, and light failure rate, it was found that: ESCO has the highest number of
perfectly working systems (due to local technicians) while DIY has the lowest number of systems
that are functioning without a fault; there are more hire purchase (29%) systems that are not
working (due to lack of technical backup) than the other three delivery modes; DIY delivery
mode has the lowest number of systems meeting standards (lack of technical knowledge of
installer and owner). The highest rate of battery failure (81%) is under the DIY delivery mode,
(most systems do not have charge controllers for battery protection monitoring the system) and
ESCO had the highest rate of charge controller failure (14%), and light failures (43%) (ESCO
system lights are used often).
It has been noted that future projects should have awareness campaigns on the importance of
using quality equipment, and it is also important to educate the clients on the importance of each
component that constitute the solar home system, to make components affordable to reduce the
usage of less efficient components and to have consumables accessible to the users of the
Gustavsson M, Ellegård A: The impact of solar home systems on rural livelihoods.
Experiences from the Nyimba Energy Service Company in Zambia. Renewable energy 2004,
Quality: high
Summary: This paper is a study on a solar energy service intervention (i.e. provision of a package
providing maintenance and repair as well as solar panels) in Zambia and assesses financial,
technical and socio-economic barriers to implementation. A survey of users and interviews of non
users was conducted to gather information about the provision of a service package by a solar
energy service company (ESCO) in Nyimba town, Zambia. The ESCO was a government
intervention to test a ‘market oriented’ approach. Amongst the indicators monitored was a change
in livelihood as a result of access to electrical services such as light.
The survey results illustrated that ESCO clients were paying substantially more than their
neighbours for energy services (paying a monthly sum rather than cost per use) (pp1063). ‘The
expense on energy services per household was ZMK 28,000 for clients and ZMK 19,000 for
neighbours’ (pp1063). However, most clients (63%) find the service fee reasonable, but they are
found to be the upper middle-class rural society, with most clients (75%) reporting a steady
household income over the year. There are reported problems with the systems (61% of clients
reporting problems), including blackouts (reported by 55% of users), low capacity of systems
(45%) (pp1065), and seasonal variations in solar radiation can be a problem, as in the rainy
season the battery may not be fully charged in one day (pp1066). One feature of the programme
is that servicing is carried out by the company, with technicians visiting each client monthly.
Clients think that this works well, as maintenance, servicing and client contacts are found locally
(people know them, know who to ask if there is a problem) (pp1070-1071), but 86% of clients
would like to own the systems themselves and not have to pay for servicing (pp1066).
In general, a wide range of benefits was reported from the systems, with half of the respondents
reporting benefits for children. Two thirds of client respondents reported changing their daily
routine and entertainment (TV, radio etc) (pp1066-1067). The number of hours of light used per
household did not change significantly, but there was improved quality of light reported, with 89%
reporting that children could study at night without complaints about the light (pp1068).
Haanyika CM: Rural electrification in Zambia: A policy and institutional analysis. Energy
Policy 2008, 36:1044-1058.
Quality: medium-high
Summary: The paper presents a general analysis of developments in rural electrification policy in
Zambia from 1994 to 2005. The study covers grid extension and decentralised micro-hydropower
and solar PV, supported by examples. The analysis describes financial, skills related, and
institutional difficulties in achieving policy objectives.
The main barriers to increased use of modern energy services are characterised as financial
constraints (high cost of the infrastructure due to long distances from existing power stations to
targeted rural areas and low population densities and inability of rural users to pay for both
capital and service due to high poverty levels). From a technical and institutional perspective, low
skills availability is mentioned as one major barrier, as well as the need for less stringent
technical standards and service levels that take into account local conditions. In addition, it can be
inferred that the institutional developments have faced important pitfalls and drawbacks due to
the lack of capacity to fully implement a well intended set of policy objectives. Main reasons are
explained as a lack of accountability, monitoring and evaluation. Moreover, a lack of mechanisms
to translate central government guidelines and policy directives are apparent, although this is not
explicit in the paper.
The main institutional and policy developments are described. Among these, the following are
highlighted: creation of a National Energy Policy (1994); enactment of the Electricity and
Electricity Regulation Acts (1995); establishment of the Rural Electrification Fund-REF (1995);
enactment of the Rural Electrification Act (2003). However, there seems to be a disconnection
between policy objectives and measures to implement them. An example of this is that although
the Rural Electrification Fund (REF) (to which all electricity consumers contribute with 3% of
their billed electricity) is considered a key governmental strategy for rural electrification, the
actual available funds are only a small proportion of the estimated annual budget, due to
bureaucracy in the transmission of the levy (in 2000, out of K12 billion budgeted in the fund,
only K700 million were available (pp 1054)).
Hofmeyr IM: Towards adequate and equitable energy provision for farm worker families.
Development Southern Africa 1995, 12:273-279.
Quality: medium
Summary: This paper reports the main findings of a study conducted during the period from 1992
to 1993 of energy access and use by farm-worker households on farms across South Africa. It
combines a postal survey of farm owners (575 respondents) in 14 regions, interviews with 36
farm-worker households in the western Cape and secondary sources. The study looks at
electrification rates but also reveals energy-use patterns for fuel wood and commercial
hydrocarbons, as well as other sources of energy services (candles and dry-cell batteries).
The author finds wide variety in electricity-access rates and consumption levels across the
country (11% access in the northern Cape, 71% in the western Cape, only 32% of households on
electrified farms across the country have access), and wide variety in fuel wood shortages (8% to
42% of households face shortages in Natal and Transvaal), while farm-owners perceive shortages
at much lower levels (9% to 12%). Where farm-worker households have electricity access, the
paper reports that consumption is constrained in a range of ways by farm-owners: limits on
appliances, restricted hours of connection, limited free units of electricity, rationing of meter
cards. Farm-owners, on average, pay 81% of farm-workers’ electricity consumption. The analysis
of these patterns suggests that barriers to electrification are financial (cost of connections and
consumption) and what could be described as political-legal or cultural, depending on how power
relations between farmers and farm-workers are conceptualised (noting the still-strong legacy of
the Apartheid regime at the time of the study).
The paper proposes a number of recommendations to overcome barriers. With regard to costs of
access to electricity, support for connections is recommended for both unconnected households
on electrified farms and PV systems in non-electrified farms. Additionally, the paper
recommends suitable tariffs to avoid low consumption levels and damage to the viability of the
utility. Payment systems should “suit low or intermittent cash incomes” (p277), and delivery
systems would need to “place control of electricity use in the hands of the consumer” (p277).
Other recommendations include strong central policy development for integrated energy planning,
minimum standards, allocation of responsibility for implementation and maintenance,
mechanisms for administering funds for service provision and structured roles for cooperation
between farm-workers, farmers, government, NGOs and energy suppliers.
Finally, the paper elaborates on democracy and inclusion. Farm-worker families should be
viewed as rural residents in their own right and their dependence on farmers lessened:
electrification was seen to depend on the goodwill of farmers, and impediments such as trespass
laws exist. In addition, any subsidies in effect at the time of the study were controlled by farmowners not farm-workers. More generally, governance institutions need to be more democratic
and efficient to ensure equitable opportunities for disadvantaged communities.
Ilskog E, Kjellström B, Gullberg M, Katyega M, Chambala W: Electrification co-operatives
bring new light to rural Tanzania. Energy Policy 2005, 33:1299-1307.
Quality: medium-high
Summary: This paper looks at a rural electrification intervention in Urambo village, Tanzania,
between 1985 and 2002, with an emphasis on the period from 1993 to 2002. The agriculture
dominated district, in which tobacco is the most important cash crop, has a population of
approximately 80,000 inhabitants, from which around 20,000 live in the village.
The paper does not have a clear methodology and design. However, the study provides useful
insights about the main barriers faced by, and lessons learned from, the establishment of a local
electrification co-operative. This operates a diesel-based generation and isolated mini-grid
distribution system providing electricity to 10% of households in Urambo township.
With regard to barriers to the increased use of electricity, the authors elaborate on financial,
managerial and technical problems. Financing for capital investment seems to be the main barrier
to increased access to electricity in rural areas. However, there seems to be high willingness to
pay for electricity services despite high poverty levels, which suggests unnecessary subsidised
tariffs in other areas served by the national utility. The case of Urambo co-operative demonstrates
that electricity supply can be managed locally when adequate technical, managerial and financial
support is provided. This support has involved highly committed international assistance from the
Stockholm Environment Institute (SEI) and the Swedish Development Cooperation Agency (Sida)
and national technical assistance from TANESCO (para-statal national utility). In order to
overcome financial struggles the authors highlight the importance of transparent management, so
clients improve their understanding and trust towards the business model (tariffs and cost
structures) and thus payment default is reduced.
According to the authors, important reasons for the success in Urambo have been (pp 1307):
Strong local leadership and commitment;
Training of the co-operative’s staff;
Utilisation of well proven technical solutions;
Initial financial support for investments and covering of initial problems with recovering
operational costs;
5. Ability and willingness to pay for at least the full operating cost of the service;
6. Availability of an organisation that is prepared to provide technical support, when needed,
without much delay.
Karekezi S, Kimani J: Status of power sector reform in Africa: Impact on the poor. Energy
Policy 2002, 30:923-945.
Quality: medium
Summary: Based on a regional study by the authors of eastern and southern Africa, and
supplemented with experiences from across other countries in sub-Saharan Africa, this paper
reviews the status of power sector reforms and their impacts on the poor.
The paper provides a detailed description of the status of power sector reforms. The average
electricity consumption per capita in Sub-Saharan Africa is found to be very low with a tradition
of a monopolistic hold over national electricity industry in the power sector. For most countries in
sub-Saharan Africa, power sector reform was the focus of intense effort during the 1990s,
attempting to address the poor performance of electricity services via the grid arising from
dissatisfaction over its poor performance. This poor performance was assumed to be the result of
inefficient and corrupt government-owned and vertically integrated utilities. Popular options of
reform are structural change and privatisation. The hope was that privatisation and unbundling of
electricity services would increase their efficiency and reliability, reduce levels of corruption and
remove political influence from electrification programmes. It was expected that electrification
rates would improve and the power sector would become more economically sustainable. Many
countries have attempted this corporatisation and commercialisation and many have introduced
contract management to improve efficiency in the power sectors. Independent power projects aim
to meet the demand of populations in Sub-Saharan Africa and recently a number of legal and
regulatory changes have occurred in some countries to encourage private participation, minimise
government intervention, redefine public utility from welfare agency to commercial entity.
The authors find that the record of power sector reform is patchy, varying at least by country, and
has not necessarily resulted in the electrification gains and reduced corruption expected. There
has been controversy surrounding the reforms, in particular surrounding the independent power
producers with allegations of corruption between government and private sector companies. Even
though a number of African countries have implemented independent power projects, the state
utilities still remain the only buyers of the power. There has been an observed reluctance to
establish independent regulatory bodies, with multi-sector regulation potentially being more
effective (as in the example of Ghana). Similarly, the impacts on the poor are context-specific.
Many countries have seen dramatic tariff increases as private companies seek full cost recovery.
The extent to which these price rises affect access to and use of electricity from the grid by the
poor depends on the extent to which the grid reaches poor households. Many of the poor in Africa
are not connected to the grid and so tariff increases do not affect them directly. However, some
countries do have large numbers of poor households electrified and here there could be serious
Not least amongst such impacts are the reactions that tariff increases can cause, such as riots in
Ghana and protracted political debate in Kenya. With regards to electrification, ‘[n]otable
examples of failed rural electrification programmes can be found in Kenya and Zambia. Several
factors contributed to the failure of rural electrification in the two countries. Firstly, poor billing
and revenue collection led to non-remittance of rural electrification levies; secondly, reallocation
of rural electrification remittances by the utility to the Treasury to other uses. Thirdly, costly
undertakings by the utilities and the failure to utilise least-cost electrification options’. On the
basis of their wide-ranging and detailed analysis, the authors offer a number of suggestions for
ways forward.
Karekezi S, Kithyoma W: Renewable energy strategies for rural Africa: is a PV-led
renewable energy strategy the right approach for providing modern energy to the rural
poor of sub-Saharan Africa? Energy Policy 2002, 30:1071-1086.
Quality: medium
Summary: This paper examined the role of PV for providing modern energy to the rural poor of
sub-Saharan Africa. They drew on data from various sources (including the World Bank,
AFREPREN and other academic publications) and found that the cost of a typical low-end PV
(i.e. 40-50 Wp) household system is 130% to 360% of the GNP per capita of most sub-Saharan
African countries, or 100% to 200% of the annual income of a rural household. Hence, they
concluded that high cost was one of the most important barriers to greater dissemination of PV
technology to households in sub-Saharan Africa. Additionally, they examined the energy use by
small and micro enterprises (i.e. enterprises that rely primarily on family/household members).
Many of these small and micro-enterprises are involved in agro-processing. Their activities
require equipment with a minimum power output of well above 1000 Wp, but a PV system of this
size was too costly for most rural households. Therefore, PV is limited to low-power application,
such as lighting, powering radios and black-and-white TVs. The paper argued that rural energy
policies that emphasize a broader range of renewable energy and focus more on incomegenerating activities were likely to yield greater benefits.
Ketlogetswe C, Mothudi T: Solar home systems in Botswana–Opportunities and constraints.
Renewable and Sustainable Energy Reviews 2009, 13:1675-1678.
Quality: medium
Summary: This paper looks at solar energy in 3 villages in Botswana – Kudumatse, Lorolwana,
Motlhabaneng – in a fee-for service project run by the Government of Botswana and the Japanese
International Cooperation Agency between 2002 and 2005. With the majority of populations in
Sub-Saharan Africa living in rural areas, with low incomes and scattered clusters of households,
the development of electricity infrastructure and therefore rural electrification is poor. This paper
describes the electrification state in Botswana, and the state of reforms to expand the use of
photovoltaic lighting systems in rural communities. It is apparent that this is the ideal market for
photovoltaic systems. The authors examine data from monitoring of revenue collected from
participants and it is found that reasons for low use of solar home systems by rural communities
in Botswana include: 1) low income status of most rural inhabitants, resulting in them being
unable to pay for the systems, and being repossessed due to a lack of payments and; 2) migration
of house-owners from village status to lands, or cattle posts, having an effect on regular payments.
The authors conclude that there is economic variation among rural communities and energy
policy should focus on this variation, and also that from observed results of revenue, the fee-forservice model seems unsuitable for rural communities. The authors make several
recommendations to stimulate sustainability of solar home systems in rural communities in
Kirubi C, Jacobson A, Kammen DM, Mills A: Community-based electric micro-grids can
contribute to rural development: evidence from Kenya. World Development 2009, 37:12081221.
Quality: high
Summary: This paper presents a detailed case study of a community-led, donor supported dieselbased micro-grid electric system in Mpeketoni, Kenya (Mpeketoni Electricity Project -MEP). The
study covers the period between 1994 and 2007 and is informed by interviews held with a broad
set of actors. It analyses the socio-economic effects of improved access to electricity, highlighting
the “mechanisms through which rural electrification can contribute to rural development” (pp.
1208), such as increased productivity per worker and income levels of SMEs, improved public
service delivery at schools (lighting and electrically-powered equipment used for educational
activities), markets (improved banking and communication services)
and agricultural
productivity (pumped irrigation and maintenance of tools and tractors). The paper describes
general barriers to improve access to modern energy services in rural contexts from a review of
literature, highlighting obstacles such as low population densities, limited ability to pay (hence
low demand), high capital and operating costs, low levels of cost recovery and political
interference. The study also assesses the financial performance of MEP analysing the levels of
cost recovery over a period of 12 years. The paper concludes that:
1. Community-led rural electrification has a high potential to reach adequate levels of cost
recovery, which is fundamental in the success of such interventions (through cost-reflective
tariffs charged and enforced and linked to the promotion of productive uses that can generate
revenues and improve the load factor).
2. The provision of capital subsidy (in the form of concessionary interest rates or grants) in rural
electrification programmes may be justified.
3. Rural electrification policy must be coordinated with other development efforts (mainly
4. Specialised institutions/agencies in charge of coordinating and developing criteria/
methodologies and rules for the process of rural electrification seem to be vital.
5. Alternative management and ownership models seem to be promising in accelerating rural
electrification in SSA given the status of development of electricity systems, but the
mechanisms to enhance collective action deserve more detailed research.
Lee KS, Anas A, Oh G-T: Costs of Infrastructure Deficiencies for Manufacturing in Nigerian,
Indonesian and Thai Cities. Urban Studies, 1999. 36(12): 2135-49.
Quality: medium
Summary: This paper presents an economic analysis of infrastructure service self provision by
private manufacturing firms based in main and secondary cities of Nigeria, Indonesia and
Thailand. Based on data collected through a survey implemented in 1988 the study assesses onsite electricity generation and other infrastructure service provision (water supply; transport and
communications), which are self managed and generated due to deficiencies in public provision
of these services.
With regard to effectiveness measures, the paper highlights that public service provision from the
supply side (infrastructure provision) is not a good metric to analyse effectiveness. Demand-side
satisfaction of service needs should also be considered when assessing effectiveness of services
The main findings of the study, in the context of Nigerian firms, can be summarised as follows:
There is a high dependence on self provision of electricity generation.
As firms grow in size they tend to rely more on self provision of infrastructure services.
Costs of service (self) provision decline as firm size grows, due to economies of scale.
Regulatory environments influence the extent to which firms are able to produce
infrastructure services on-site. Nigeria has a strict regulatory framework inhibiting the
development of private infrastructure service delivery, despite the inefficiencies of state
infrastructure provision, while in Indonesia and Thailand governments have opened up the
market and the incidence of private service delivery is much higher than in Nigeria.
Mapako M, Network BU: Provision of Long-term Maintenance Support for Solar
Photovoltaic Systems - Lessons from a Zimbabwean NGO. Journal of Energy in Southern
Africa 2005, 16:21-26
Quality: medium
Summary: This paper compares two different interventions in Zimbabwe, which both aimed to
ensure maintenance of the solar home systems (SHSs) throughout their lifetime. These two
interventions were guided by different approaches. One is the facilitator approach, illustrated in
the GEF Solar Project implemented between 1993 and 1999. This approach uses an independent,
trained, local person who can respond to maintenance issues in the SHS on a case by case basis,
with payment from the SHS owner per visit. The other is the fee-for-service approach as
exemplified by the Japan International Cooperation Agency Energy Service Company (ESCO)
study which ran from 1999 to 2003. In this case, the customer pays a regular fee to the company
for installation and maintenance of the SHS.
The authors found that the clustering of installations is important for facilitators as they have
access to more business, and also for the ESCO as it is able to have shorter response times, and
lower transport costs. The facilitator approach can be viewed as more quality based and
competitive, as customers can choose which facilitator to use, and there is an incentive to
compete with other facilitators in the area. There is little competition under the ESCO approach,
but the user can withhold fee payment if they are unsatisfied. The main problem with the ESCO
approach is that it has not been shown to cover operational costs: e.g. fee payments are disrupted,
foreign exchange rates are volatile. The facilitator approach does not suffer from an erratic
payment problem and so may be more suitable for the developing world. In summary, the
facilitator approach shows ways to disseminate SHSs commercially, but the ESCO approach has
not shown financial viability.
Mulder P, Tembe J: Rural electrification in an imperfect world: A case study from
Mozambique. Energy Policy 2008, 36:2785-2794
Quality: medium-high
Summary: This study presents a cost-benefit analysis (CBA) of a rural electrification project,
grant-funded by a donor – Sida (Swedish Development Cooperation) – at a cost of USD 4 million
in the mainly agricultural Ribáuè district in Nampula province in northern Mozambique. The
project included a grid extension of 160 km plus installation of a distribution network providing
electricity to households, as well as commercial customers and other organisations.
The CBA is conducted in two parts: an analysis of a period of operation (2000 to 2005) and a set
of scenarios (2005 to 2020). A combination of reported and assumed values is used for the CBA
together with a sensitivity analysis using different discount rates. These values are derived from
electrification rates, consumption patterns, direct and indirect costs and benefits: household
energy savings, lower production costs and increased demand (cotton production and maize
milling), new businesses (shops, bars and restaurants), education and health gains, more tax
revenue and increased access to ICTs. However, some of the benefits and costs are not included
because of methodological difficulties.
The paper argues that institutional barriers explain why such projects are difficult to implement.
The study finds that rural electrification can be commercially viable if there is at least one major
consumer for productive purposes. This leads to the recommendation that projects should look for
such productive opportunities so that the positive externalities (many of which are difficult to
quantify) are captured – those such as educational and health gains, as well as the development of
economic linkages into other parts of the locality. However, complementary developments need
to be in place to maximise these externalities – e.g. roads and stronger institutions at the national
level. Many of the positive externalities are not appropriable by the private sector and so there is
likely to be an underinvestment in such projects even where they might be commercially viable,
adding weight to the case for government and donor support.
However, donor support can be a problem because it can introduce complexities into policy
development and implementation – donor fads, for example, could mean shifting priorities that
undermine the long-term needs of infrastructure development. The paper adds that other
institutional changes could be necessary such as utility reform but these are likely to be resisted
by vested interests – the Mozambican case illustrates this where reforms have not been funded by
the government and so have in effect been blocked.
Mulugetta Y, Nhete T, Jackson T: Photovoltaics in Zimbabwe: lessons from the GEF Solar
project. Energy Policy 2000, 28:1069-1080.
Quality: high
Summary: The paper critically analyses a GEF-funded PV project implemented in Zimbabwe
between 1992 and 1998. It starts by characterising the country-wide intervention to promote the
diffusion of PV at the household level and its main objectives. It then analyses the effects that the
project had on the PV market in the country. The paper investigates the extent to which a donordriven rural lighting project achieved its intended goals of promoting the creation of a solar
market in the country while reducing poverty and limiting GHG emissions. The main findings
can be summarised as follows:
1. Donor-driven technology-push initiatives have little likelihood of long-term sustainability
(retention of technological capacity at the local level, adequate operation and maintenance of
PV systems, growth of service and technology markets), so there is a need to generate
adequate policy frameworks beyond limited project lifetime;
Projects aimed at delivering rural electrification solutions should be conceived in the
framework of defined rural electrification strategies and national energy polices;
Such strategies should respond to critically assessed demands and needs informed by
continuous monitoring, analysis and evaluation from national policy makers in order to
ensure donors focus on national priorities;
There is a need to improve accountability and transparency of policy interventions, allowing
greater participation of a diverse range of stakeholders and defining clear monitoring and
evaluation criteria;
Donor-driven projects providing market mechanisms and incentives have a tendency to
distort already feeble local renewable energy markets. It is not clear the extent to which
market creation-based projects are able to target poor populations in low-income rural areas
of developing countries.
Murphy JT: Making the energy transition in rural East Africa: Is leapfrogging an
alternative? Technological Forecasting and Social Change 2001, 68:173-193
Quality: medium-high
Summary: The paper looks at three energy technologies in East African households: electricity
grid extension; renewable electrification (solar PV and biogas); and efficient cooking stoves.
Using data from financial and technical cooperation projects from international and multilateral
donors implemented in the 1980s up to the late 1990s in Kenya, Tanzania and Uganda, the study
analyses the potential for leapfrogging to modern energy services. This paper presents an
interesting approach, drawing on ideas of technological change, socio-technical transitions,
technology transfer, leapfrogging and capabilities. It is argued that the problem of lack of access
to modern energy services requires a systemic approach in which the context, characterised by
social, economic, cultural and political institutions, is of crucial importance.
Thus, the paper describes several cultural factors hindering the diffusion of energy technologies,
such as gender dynamics, local power structures and socio-economic realities of poor rural East
Africans. The author argues that most of the barriers arise due to an over-emphasis on technoeconomic viability of the technologies being promoted in detriment of an appropriate
consideration of socio-cultural and institutional factors.
Although the study refers to economic, social, political, and cultural factors, barriers could be
categorised as financial, technical, organisational (or managerial) and cultural.
There is also mention of effectiveness measures. In that regard, the study indicates that most
interventions focus on technologies filling a gap in energy services from a top-down approach
relying on the automatic creation of markets and provision of supply-side services, ‘(h)owever,
adoption levels tell us little about the sustainability of these technologies or the efficacy with
which they are utilized’ (pp187).
The paper concludes that "energy transitions in rural areas are incremental processes" (pp 173),
therefore leapfrogging is a misconception because there is insufficient attention to the
dependence of the accumulations of technological capabilities at the individual and institutional
levels in different scales (local, regional), which requires long time-scales for the technology and
associated practices to be designed, tested, redesigned or redeveloped and widely disseminated, a
rather "hard slog" process (pp 188).
Nygaard I: Institutional options for rural energy access: Exploring the concept of the
multifunctional platform in West Africa. Energy Policy 2010, 38:1192-1201.
Quality: high
Summary: This paper is the analysis of two case studies to examine the implementation and
barriers to implementation of multifunctional platforms (MFP), a small diesel engine turning one
or several types of milling machinery and a generator for production of electricity, under the
management of women in Mali and Burkina Faso. The study included a literature review, Mali
programme internal reviews (1994-2005), informal talks in Mali in 2007-2008, and in Burkina
Faso in 2007.
The paper found that installations were subject to a number of technical and organisational
problems. Barriers included the hard to install multifunctional nature of the technology, financial
losses from sequential running of machines (therefore using the same time as two separate
machines), the risk of repair and associated down-time, and reduced flexibility with regards to
location (pp1196-1197). About 35% of installations were not in operation by the end of 2005 and,
according to the review, 60% of non functioning MFPs were a result of socio-organisational
problems, 26% ceased due to technical problems, and 14% due to economic problems (pp1197).
It seems that women’s organisations faced difficulties in operating and managing the platforms,
with a greater share of functional than non functional platforms changing their organisational
modality. Specifically, although donors had stipulated that the MFPs were to be controlled by
women’s groups, ‘men are almost systematically present in the women’s groups, and in a
considerable number of cases operation has been taken over by men’ (pp1198). In 43% of cases,
the women had handed over the operation of the platform to a concessionaire, illustrating that
women’s associations faced difficulties in operating and managing platforms, with only 50% of
platforms having women responsible for day to day management. While this would suggest that
women’s groups were poor managers, the argument actually appears to be that any community
groups would have faced the same problems. That is, the author appears to argue that communitygroup management is not a familiar form of business operation and so, in the cases studied, led to
social conflicts. Consequently, the argument is further elaborated to say that development aid
should build on local structures rather than impose new and complicated institutions.
Peng WY, Pan JH: Rural electrification in China: History and institution. China & World
Economy, 2006. 14(1): 71-84.
Quality: high
Summary: This paper reviews China’s rural electrification program in relation to its three stages
of development. The first stage was from 1949-1977 and established a comprehensive
management network vertically from the national level. Investment was primarily in rural
electrical irrigation and agricultural production, in addition to a small programme of hydropower
production. Rural electrification in this stage was slow and 245m people did not have electricity
access. Rural power consumption was 13.3%, whereas people in rural areas accounted for 70% of
the country's population. Line losses due to inefficiency were 25-30% and there was a total
capacity of 6.33 GW.
The second stage was from 1978-1997 and was characterised by central government transferring
management of local electricity systems to the local government. Three main investment projects
were also undertaken in this stage: i) rural small hydropower investment enabled rural
electrification in 653 counties; ii) poverty reduction was the second project and significant
investment was put into rural electrification in order to reduce poverty; and iii) hydropower
station construction in rural electrified counties was initiated and government subsidies were
introduced. At the end of this stage, electricity consumption in rural areas increased to 495.5bn
kWh, which was about 40.53% of the national total. Capacity of generating units at county level
and below was 44.15 GW and accounted for 26.7% of total electricity consumption. Rate of
access was 99.2% in townships, 98.1% in villages and 96.87% in rural households. However,
8.81 m rural households were still without access.
The third stage of policy for rural electrification was from 1998-2002 and this stage promoted
commercial operation of the utility market. This stage saw institutional reform which aimed to
facilitate the commercial operation of the utility market and a rural grid renovation programme in
order to overcome technical barriers to electrification. All effectiveness measures improved
during this stage. Electricity voltage was more than 90 percent; 12% higher than before
renovation of the grid. The reliability rate was up by 8 percentage points to 95% and grid losses
were down by 25-30% to about 12%. Costs to some rural areas were reduced by over RMB
0.13/kWh, resulting in savings of over RMB 42 billion in electricity costs.
The authors conclude that it is easy to simply attribute the success of China’s rural electrification
programme to government investment and favourable policies for rural electrification, although
this is not the case. Instead, the main driving forces of rural electrification in China are related to
a diversity of funding channels, modes and institutional structures. Rural electrification drives
development and must be supported by a mix of technical, social and economic factors, within a
supportive institutional context.
Pereira MG, Freitas MAV, da Silva NF: Rural electrification and energy poverty: Empirical
evidences from Brazil. Renewable & Sustainable Energy Reviews, 2010. 14(4): 1229-1240.
Quality: medium
Summary: This paper reports on a large survey which aimed to generate empirical evidence on
the impact of rural electrification on energy poverty in rural areas of Brazil between 2000 and
2004. The paper contains a summary of the state of knowledge about rural electrification and
energy poverty in Brazil, in addition to a presentation of the findings of a survey and analysis of
23,000 rural households.
The authors argue that although energy poverty is widely discussed as an objective of rural
electrification, there are few studies which empirically examine their relationship and the
dynamics of energy poverty over time. Two groups were surveyed for the study: those that were
in the electrified sample were part of the national rural electrification program, Light for All, and
the second group was a control sample (non-electrified) comprised of a set of rural households
that were not part of national rural electrification programme. Households were sampled prior to
and after access to regular electricity. The control sample was interviewed at similar time periods
to detect natural changes in energy consumption/poverty during the time period.
They show that use of energy after electrification indicates a dramatic drop in energy poverty as
energy use increased significantly in rural households. A small drop in energy poverty was seen
in the non-electrified sample, over the same time period, but it was not as significant. The authors
conclude that the fact that there was a statistically significant change in energy usage indicates
that there was a change in behaviour in families which had access to electricity, though this level
of usage is still far from that indicated as necessary in the literature in order for a significant
improvement in life quality to be observed.
Rehman IH, Kar A, Raven R, Singh D, Tiwari J, Jha R, Sinha PK, Mirza A: Rural energy
transitions in developing countries: a case of the Uttam Urja initiative in India.
Environmental Science & Policy, 2010. 13(4): 303-311.
Quality: medium-high
Summary: This paper is a case study of the Uttam Urja initiative, a project that used an
alternative business model to disseminate photovoltaic lighting technologies in rural areas of
India from 1999 to 2006. The initiative aimed to create a market for renewable energy technology
services and to build a supply chain which would meet market demand and would promote
delivery and management of energy services through a local supply chain. The main features of
the Uttam Urja project included the articulation of expectations which refers to the project's
attempt to differentiate itself in the market by creating awareness and brand recognition, social
network development which targeted dealer and after-sales service provision which 'piggybacked' on existing entrepreneurs, learning through after-sales services and from the market in
order to develop new products which met specific user requirements and created new niches, and
protection to entrepreneurs and end-users through financial innovations.
The paper describes the initiative and analyses its outcomes using the Strategic Niche
Management framework. Though the paper is well-written, and the arguments are coherently
structured, it is at times unclear how the evaluation data were collected and what the sources were.
Nevertheless, the paper describes a range of technological, social, environmental and financial
barriers to uptake of photovoltaic lighting technologies. For example, technologies can be deeply
embedded in societal norms, values and cultures, and therefore social acceptability is important to
examine. Other barriers to renewable energy technologies can include accessibility, reliability and
maintenance issues; inconsistent levels of funding through government subsidies; the fact that
products do not always cater to regional/local needs; financial barriers for end-users, in particular
access to funding for rural households with an agrarian lifestyle; and the nature of programmes
which are target-oriented and have low scope for learning, monitoring and evaluating.
The authors found that advertising schemes led to increased sales. Before branding, sales were
INR 0.46 million and INR 0.59 million in Bikaner and Rishikesh, respectively. After branding,
sales were INR 3.61 million and INR 2.43 million, respectively. Dealer response was also high,
with products crossing into 7 adjoining districts and at the end of the project 12 new products
were introduced, many of which were more affordable for lower income groups. Finally,
innovative financing allowed for increased uptake: group financing systems, soft loans and credit
cards all contributed to this. Success of the experiment could be attributed to the following
elements: development of a local business network; improved accessibility of products to rural
consumers; products were designed to meet local needs which led to greater acceptability; and the
portfolio of products on offer meant they could reach a broader market.
Schut M, Slingerland M, Locke A: Biofuel developments in Mozambique. Update and
analysis of policy, potential and reality. Energy Policy 2010, 38:5151-5165.
Quality: high
Summary: This paper analysed various sources of secondary data, including biofuel proposals
submitted from 2006 up to December 2008, in order to examine barriers (policy, biophysical,
social, economic and legal) to biofuel use in Mozambique. The review of all dynamics, using
recent biofuel development proposals as case studies to understand biofuel development in
Mozambique, found that there are numerous interconnected factors relating to development, and
particularly the location of developments.
Some of these factors are listed below. Barriers to policy: trade agreements (tariffs opening up
markets, e.g. SADC trade protocol eliminating tariffs on intra-regional trade within 11 SADC
members opens up the market), land law, investment law (government incentives are location
specific, affecting location of activities), national biofuel policy and strategy (‘aims to contribute
to energy security and sustainable socio-economic development by exploiting agro-energetic
resources through stimulating the diversification of the energy matrix, contributing to the wellbeing of the population and promoting socio-economic development’ (pp5154)). Biophysical
potential: natural resources (land zoning recently summarized potential on a national scale)
(pp5155). Social and economic factors: labour force availability and quality, access to
infrastructure and services, existing processing and storage facilities (‘efforts to provide
electricity to rural Mozambique are mainly concentrated around urban centres...’ due to poor
transport infrastructure, service provision, and higher population density and literacy rates in
cities) (pp5156).
Steel K: Dynamics of Growth and Investment in the Kenyan Electric Power Sector, Power
Engineering Society General Meeting, 2007. IEEE. 1-5. doi: 10.1109/PES.2007.385906
Quality: high
Summary: This institution working paper describes preliminary findings of a study on
electrification (grid vs. off grid) in Kenya, looking at barriers to electrification including
technical, socio-economic and financial barriers. The study uses a literature review, then
ethnographic methods of interviews and observations, surveys, coupled with system dynamics.
It was found that people are moving off grid due to the unreliable nature of the power, long waits
for connections, common outages, and high fees. Many industrial and commercial customers
already generate some of their own power. Residential people may be off grid as they are not
close enough to the grid. The paper concluded that if people start to move off grid, the grid could
have decreased finances to invest in new infrastructure, therefore creating a cycle (which has
already been seen in the telecom sector in Kenya and the takeover of mobile phones).
Thiam DR: An energy pricing scheme for the diffusion of decentralized renewable
technology investment in developing countries. Energy Policy 2011: 39(7). 4284-4297
Quality: high
Summary: This paper is an empirical analysis of pricing mechanisms of renewable energy
technologies (particularly wind and photovoltaic) in Senegal. Data were gathered for the model
from local interviews and secondary sources.
The barriers to implementation of renewable energy technologies in a country were found to be
related to availability of renewable resources (i.e. wind and solar), externalities (tax policies,
subsidies, tradable permits) and command and control (standards and regulations).
The study suggests that alongside a shift to market oriented approaches to renewable energy, an
improvement in energy governance and the use of local stakeholders to include local constraints
and socio economic characteristics is required. The financial conditions of a country could be an
obstacle for deployment of renewable technology as many price mechanisms require borrowing
from international organisations.
Finally, the paper concludes that investing in renewables requires strong institutional reliability,
and therefore depends on political and social structures, and the technology selected should be
socially accepted by local end users.
van der Plas RJ, Hankins M: Solar electricity in Africa: a reality. Energy Policy 1998, 26:295305.
Quality: high
Summary: The paper describes the dissemination of SHSs in Kenya, focussing on adopters of PV
systems from rural areas in 12 districts across 3 regions of the country: Mount Meru, Western
Kenya and Rift Valley – all “high potential areas”, meaning areas with fertile soils and mainly
export-oriented agriculture. The study attempts a pseudo-random sampling method to establish a
survey group that includes 410 SHS adopters who were investigated during December 1996 to
February 1997.
Technical, financial and capacity (skills, knowledge, etc.) aspects of SHSs and a brief history of
the SHS market is given. A range of indicators is reported: user-satisfaction, including by SHS
size; technical quality and performance of systems, using a range of measures; and market prices
for systems and components.
Based on the observations that many users buy the smallest modules available (10 or 12 Wp) and
build their systems piecemeal, the paper suggests that there is a need for the PV industries to
address this low-power market – small systems (1 to 2 lights) and small deeper-discharge
batteries as well as smaller modules. Credit is suggested as a way to overcome the cost barrier of
higher quality systems “but less so for initial market entry purchases” (p304). Costs are seen as
the main constraint to accelerated use of SHSs. The paper also suggests that access to SHSs
means that customers build up their demands for further electrical services and so could be a
strategy to develop enough demand to make grid extension economically attractive.
In summary, the paper suggests promotion of a modular approach to technology
diffusion/adoption, importers and retailers should provide more information to potential adopters
of the real costs and benefits of such components, and training of installers (perhaps with
certification) should be implemented to raise the quality of installations (the survey found that
those systems installed by a technician or solar company were generally in better technical
condition than those self-installed).
Walekhwa PN, Mugisha J, Drake L: Biogas energy from family-sized digesters in Uganda:
Critical factors and policy implications. Energy Policy 2009, 37:2754-2762
Quality: high
Summary: This paper looks at biogas using family sized digesters in Uganda, assessing socio
economic and demographic barriers to their implementation. The data were gathered via a survey
between November 2007 and April 2008 from 220 households characterised as 'users' and 'non
users', individual interviews and focus group discussions with households and key informants.
It was found that the adoption choice of a technology is a complex set of interactions between
comparable technologies and the user's socio-economic and demographic characteristics.
Variables affecting biogas adoption are: age of household head, education, household size, size of
land of household, number of cattle owned, household daily fuel wood cost and kerosene, gender
of head, location of house, monthly household income. A logistic regression using primary data
on households was used to find the empirical results which are listed below.
Factors positively promoting biogas development: younger/male head, increased farm income or
cattle owned, increased cost of traditional fuels. Factors with a negative impact on biogas
adoption include household location and farm size (pp2759). ‘On average, biogas users were
older, had more years of formal education, owned a larger area of land and spent more on
kerosene for household lighting purposes than their counterparts. However, the biogas non-users
had larger households, reared more livestock and incurred more expenses for fuel wood for
cooking purposes’ (pp2758).
This paper shows that characteristics of households can provide good indicators as to reasons for
uptake or not of technology, and consideration should be given to economic, social and human
behavioural characteristics as well as technical questions when considering biogas technology
uptake (pp2759).
Winkler H, Spalding-Fecher R, Tyani L, Matibe K: Cost-benefit analysis of energy efficiency
in urban low-cost housing. Development Southern Africa 2002, 19:593-614.
Quality: high
Summary: This paper presents the results of a cost-benefit analysis of energy efficiency in urban
low-cost housing in South Africa (Cape Town, Johannesburg, and Durban). The authors
considered a number of energy-efficiency improvements in low-cost housing, including: ceiling,
roof insulation, partition, wall insulation, window, shared walls, compact fluorescent lighting and
solar water heating. The method used was discounted cash-flow analysis. All interventions are
considered over 50 years, as this is (optimistically) assumed to be the standard economic life of a
low-cost house. Affordability was measured by the capital subsidy that would be required to
induce consumers to invest in energy efficiency on their own.
The study founded that most of the energy efficiency improvements do not yield a net benefit
when a 30 percent discount rate is used, as the future energy savings simply have much less value
to consumers with high discount rate. For the 30 square-meter Reconstruction and Development
Programme (RDP) house, a capital subsidy of around ZAR 1000 is required. According to the
authors, this is a modest amount.
Wolde-Ghiorgis W: Renewable energy for rural development in Ethiopia: the case for new
energy policies and institutional reform. Energy Policy 2002, 30:1095-1105
Quality: medium
Summary: This paper reviews rural energy policy in Ethiopia. In rural areas in the country,
energy is supplied by traditional energy sources. Fuel supply is mainly biomass-based with a very
low level of grid electrification. However, biomass supplies are dwindling and it is deemed
imperative to find an alternative such as renewable and other non-renewable energy sources. It is
found that in Ethiopia, renewable energy technologies are unattainable for rural communities due
to high import and customs duties or indecision by manufacturers to invest (pp1098). There are
arguments that the reason for low levels of electrification have been due to a low demand for
modern fuels, which perhaps is due to an inability to pay for such energy. High costs of rural
electrification are also an issue (pp1099) and rural energy development has received a low
proportion of funding in comparison to education, road construction and health (pp1100). It can
be argued that these problems stem from a discourse in budgetary allocations, with rural energy
initiatives under the responsibility of regional governments, therefore not allocated in federal
level budgets (although the regional governments receive their allocation from the federal
treasury), or that rural energy can be seen as a waste as it is not used for industry or agriculture,
only for household use. It is observed that Ethiopia has delayed action with regards to energy
infrastructure development in comparison with other African countries. Wolde-Ghiorgis
recommends several policy issues that may lead to improved performance of rural energy
initiatives (pp1102), including a revision of the national energy policy to include rural energy
initiatives, clearly state the responsibilities of the authorities in the policy, and the need to draft
new regulations and a new legal framework with regards to deforestation. Aside from these issues,
careful consideration of the financial implications and the role of human resources and technical
capability are needed.
The World Bank Independent Evaluation Group: The Welfare Impact of Rural Electrification:
A Reassessment of the Costs and Benefits. The Welfare Impact of Rural Electrification: A
Reassessment of the Costs and Benefits, World Bank; 2008
Quality: high
Summary: This report is a re-assessment of costs and benefits of rural electrification in the World
Bank’s projects. It reviews recent methodological advances made in measuring the benefits of
such projects. The study examines the extent to which changes in the Bank portfolio have
addressed concerns regarding the limited poverty impact of rural electrification.
The projects assessed fall into three categories: dedicated rural electrification projects, energy
sector projects with rural electrification components, and multi-sector projects with rural
electrification components. The study highlights a recent trend in the growth of off-grid
electrification, which mostly relies on renewable energy technologies. Most of the projects aim to
improve energy supply or to support institutional development. Only 7 percent of projects
explicitly address poverty-reduction objectives.
From an economic perspective, the study shows that rural electrification can generate sufficient
benefits and that willingness to pay (WTP) for electricity is high so cost recovery tariff levels are
achievable. Conventional grid electrification is found to be more efficient from an economic
perspective that off-grid, which is more costly and benefits more limited. It is argued that “the
economic rationale for funding off-grid alongside grid extension when the latter has the higher
economic rate of return is far from clear” (p.17). Social concerns or technology development
through learning by doing and expected cost reduction are suggested as arguments for off-grid
electrification support.
The study also shows that projects have been more effective in providing infrastructure (although
technical problems are still critical) than in supporting institutional development (mainly linked
to the lack of financial sustainability). However, the sole provision of infrastructure does not
guarantee outcomes (benefits on poor rural populations) are yielded automatically. Specific
project design components connecting outputs (infrastructure or number of connections) to
outcomes (welfare/development impacts) need to be carefully considered.
One of the main conclusions of the study is the widely recognised fact that the larger share of
benefits is captured by the non-poor. From a supply-side, these benefits depend on which
communities get connected: normally richer communities close to towns and commerce are
prioritised for grid extension. From a demand-side, the benefits depend on which households can
afford the connection: when electricity is available, the better-off households get connected to
grids and the poorest in those villages or towns remain unconnected. For off-grid SHSs, the
better-off families can afford expensive systems. This is also true for off-grid electrification,
which has typically been carried out through a private business model. Moreover, lack of
information to customers has meant that they do not get the full potential of their access to
electricity because they normally restrict consumption on the wrong assumption that they will
save money, even when they normally benefit from a fixed charge “lifeline tariff”.
The study concludes that the evidence base remains weak for many of the claimed benefits of
rural electrification. From a normative perspective, project designs that would enhance project
benefits, such as financing schemes for connection charges, consumer education and support for
productive uses have to be explored.
4.3 Analysis of cross-cutting issues (qualitative synthesis)
It was not possible to undertake a quantitative meta-analysis, as the studies reviewed used a wide
range of methodologies which makes them non-comparable in most cases. Nevertheless, a
qualitative meta-analysis was undertaken to provide a discussion on the cross-cutting issues. In
this section, the results of the qualitative analysis are presented. The key themes of barriers and
interventions are broadly classified into: economic, technical, political/institutional and
cultural/social. This classification is to provide a framework to understand them, although it is
recognised that the barriers and interventions do not always neatly fall under a single category
(these cases are noted).
Economic barriers
Economic barriers refer to several factors linked to the ability to afford the installation,
maintenance and operation of technologies that facilitate energy services. These barriers can be
categorised as:
• Demand-side barriers, such as high costs of technology (capital investment and operational
costs), customer preferences, complexity of financial flows and lack of access to finance
• Supply-side barriers, such as low capacity for adequate cost recovery, high cost of
Upfront costs are reported strongly and quite consistently in the literature as barriers, irrespective
of the technology. Operation and maintenance costs are also discussed regularly, although
whether they act as barriers is more contested. Some authors report that low income users are
unable to pay for operation and maintenance (Haanyika 2008; Kirubi et al 2009), while other
evidence suggests there is high willingness to pay for electricity services, even in poor areas
(Ilskog et al 2005). The evidence and arguments for these demand-side factors are of different
kinds: for instance, Acker and Kammen (1996) and van der Plas and Hankins (1998) show how
Kenyan customers prefer the smallest PV modules available in the market; Kirubi et al (2009)
calculate financial flows and cost recovery for a diesel-based mini-grid in Kenya; Winkler et al
(2002) report on upfront cost of energy efficiency measures in South Africa; and this type of
barrier is also mentioned in empirical research for rural electrification projects in Tanzania
(Ilskog et al 2005); Mali (Diarra and Akuffo 2002); Zambia (Haanyika 2008); India (Rehman et
al 2010), and mentioned in many interviews (interviewees Collings, Hunt, Bess, Brew-Hammond,
Yadoo, Barnett). These demand-side economic barriers are normally linked to a lack of access to
finance at several levels: poor rural customers unable to access credits, unreliable or non-existent
government subsidies or dependency on donor support.
On the supply side, low population densities and high poverty levels make for unattractive
markets because of low energy demand and difficult cost recovery. Supply is further hindered by
lack of working capital and financing options for investors and utilities, even where there appears
to be high willingness to pay for electricity (Green and Erskine 1999; Ilskog et al 2005). The
aforesaid was also mentioned by one interviewee 6 : small enterprises are constrained in
“supplying modern energy services to poor people [where economics is concerned, by] lack of
equity rather than debt finance, the high transaction costs in developing viable enterprises…and
the limited deal flow of viable schemes”. Remoteness, cost recovery difficulties and transaction
costs were also identified as barriers in three interviews (interviewees Bess, Brew-Hammond,
However, it is also recognised that economic factors do not act independently but in conjunction
with other factors. This confers a systemic nature to the barriers analysis presented here, which is
often neglected in the research reviewed. For instance, Schut et al (2010) offer evidence from
biofuels developments in Mozambique, where infrastructural links, access to goods and services,
processing and storage, and skilled labour appear to be more important than the biophysical
potential or government policy. In a similar vein, one of the interviewees, in a supplementary
note to the interview (see footnote 6), refers to the availability of “complementary inputs”7 for
achieving poverty reduction with modern energy services; and Murphy (2001) argues that techno-
The quote is taken from a supplementary note provided by the interviewee, Andrew Barnett: see Barnett (2010) in the
The note gives the example of pumped irrigation having more impact if irrigation ditches are already in place.
economic viability is often overemphasised, detracting from attention to socio-cultural and
institutional factors, which are reviewed later in this section.
Technical barriers
The discussion in the literature about technical barriers is consistent in general terms but diverse,
ranging from the quality and performance of technological equipment to the capacities to install,
use and maintain energy systems at different scales (users, service providers, regulators,
institutionalisation of energy service provision), to the social adequacy of the technologies.
Those barriers involving hardware failures or poor performance create direct problems for the use
of the energy service and indirect problems by creating perceptions of risk for their uptake. These
problems are common for off-grid electrification technologies (renewable and fuel technologies)
and grid connected electricity. In some cases this is a problem of technical quality of the
components available on the market, as Acker and Kammen (1996) show for Kenyan SHSs in
which the most common problems occur with batteries and lights. One interviewee referred to the
low quality of components and inflated claims about technical performance creating huge risks
for investors (interviewee Barnett) and another reinforced this view mentioning a lack of quality
control of imports of PV components in Kenya (interviewee Yadoo). Indeed, Duke et al (2002)
report how Kenyan households suffer major financial losses simply because they have had the
misfortune to purchase the wrong brand of PV module. Consumers are unable to discern the
relative quality of different module brands which brings as a consequence a reduced confidence
in PV, creates an important market failure, and generates problems of operation and maintenance.
But even more relevant, ‘software’ problems of the capabilities for designing, installing and using
technologies are commonly and consistently discussed in the literature. Green and Erskine (1999)
refer to low levels of technical capacity in the community to operate and maintain SHSs and to a
dependency on donors' technical support, which reduce the understanding of users about the
limitations of SHSs and result in failures or poor performance. For non-traditional technologies
(such as PV and other renewables) the lack of maintenance services and formal operational
contracts is an indicator of very low skills levels leading to poor performance and failure
(Haanyika 2008; Rehman et al 2010; interviewees Collings, Yadoo, Lockwood). For some
established technologies, these problems are not so acute. For example, for those technologies
around traditional commercial fuels such as kerosene, diesel or petrol, or services such as batterycharging, there are informal networks that can deliver maintenance. However, generally,
management capabilities are weak such that systems (particular installations, but also systems of
delivery, maintenance and service) are insufficiently organised and supported. Furthermore,
where technical capabilities are developed in rural areas, the skilled people so trained are often
attracted to urban areas where there are better job prospects and markets are more established
(Green et al 2001).
Another technical limitation to the effective use of modern energy services appears in grid
connected areas where unreliable power supply is the norm. In such cases, some users seek their
own solutions by using alternative generating technologies, and so face higher costs than
otherwise would be the case. For poorer users, this may just mean no service at all. Lee et al
(1999) and Foster and Steinbuks (2009) studied the causes of self-provision of electricity by
private firms, finding that there is considerable installed capacity in sub-Saharan Africa (on
average 6%, but in some countries private self-generation accounts for more than 20% of total
installed capacity) and that it depends on firm size, sector and export orientation. In the case of
households, Steel (2007), van der Plas and Hankins (1998) and Acker and Kammen (1996) found
that some SHS users in Kenya are actually very close to the grid or are moving off-grid. There
may be many reasons for this but one suggested by Acker and Kammen (1996) was the slow
process of establishing a grid connection (in which local politics can be a frustrating factor)
compared with the speed of buying and installing a SHS. This evidence suggests that off-grid or
decentralised power generation could play an important role in energy access in the future,
particularly in Sub-Saharan Africa where generation, transmission and distribution infrastructure
is weakly developed (World Bank 2008), although, as Murphy (2001) concludes, “energy
transitions in rural areas are incremental processes” (p173), highly dependent on the
accumulations of technological capabilities at the individual and regional scale, a “hard slog”
(p187) process.
Political/Institutional barriers
The politicisation of energy service delivery and its impact is discussed in some of the literature,
and was mentioned by a number of our interviewees. Kirubi et al (2009) talk of political
interference, while elsewhere there is reference to corruption and the behaviour of vested interests.
In the case of Mozambique, Mulder and Tembe (2008) attempt to explain some of the difficulties
in implementing institutional changes such as power sector reform with reference to the
resistance of vested interests, including some in government. The government, they claim, has not
funded power sector reform and so it has, in effect, been blocked. Two interviewees referred to
the case of India and the free provision of electricity to farmers, who form a powerful voting bloc
(interviewees Lockwood, Barnett). The flipside of this observation is that where there are few
votes to be gained, such as in sparsely populated rural areas, there may be little political incentive
or will to drive energy-service provision. Two interviewees expressed the view that lack of
political will is an impediment to extending energy services to rural areas (interviewees BrewHammond, Yadoo). In the case of Apartheid-era South Africa, Hofmeyr (1995) found that the
lack of political power amongst farm-worker households, resident on commercial farms, was an
important reason why they were either constrained in their electricity consumption by the farm
owners or were not connected to the grid even when it was available on the farm. And, at the
local level, politics and conflict are reported as constraints to the use of energy services. A study
of multi-functional platforms in Mali and Burkina Faso, for example, reports that 60% of nonfunctioning platforms were the result of social or organisational problems that were caused by
conflicts at different levels (Nygaard 2010, p1197).
At a more technical level, the literature reports many of the difficulties posed by institutional
weaknesses or dysfunctional constraints of various kinds. Some of these relate to capabilities, as
discussed above under technical barriers, but others arise from weak policy implementation
mechanisms or the complexities of policy interactions, unstable policy environments, or the
influence of shifting donor priorities. A study in Zambia, for example, implies that a long
administrative chain from policy objectives to implementation to set up a Rural Electrification
Fund resulted in drastically reduced funds. Instead of the budgeted ZMK 12 billion in 2000, only
ZMK 7 million was made available (Haanyika 2008, p1054). Mangwengwende (2002),
discussing power sector reform in Zimbabwe, lists a number of tariffs – lifeline tariffs and special
agricultural tariffs – and explains that attempts at their rationalisation were constrained by the
failure to establish a predictable and sustainable tariff-setting process. Mulder and Tembe (2008)
argue that shifting donor priorities can introduce complexities into the policy process, particularly
where they conflict with the long term needs of infrastructural development. And Lee et al (1999)
compare the self-provision of energy services among firms in Nigeria, Indonesia and Thailand;
responses to poor-performing electricity supply. They find that Nigerian firms are highly
constrained by strict regulations that do not allow private provision of infrastructural services
(although many firms generate their own power in any case), while the Indonesian and Thai cases
are much more open with consequently much higher rates of self-provision.
Cultural/Social barriers
Often under-considered as determinants of success or failure in the uptake of modern energy
services, social and cultural barriers arise in regard to the ways in which technologies fit with
social practices, and can inform understanding of other types of barriers.
As already mentioned in the discussion on economic barriers, the dispersed nature of rural
populations (Kirubi et al 2009), and behaviour such as theft of grid electricity or PV modules
(Green and Erskine 1999), poses difficulties for the expansion of energy markets into rural areas.
Dispersed and often low demand can create unfavourable conditions in which to provide reliable
and affordable service. In addition, the remoteness of rural livelihoods makes it difficult to
prevent theft, further undermining service (interviewee Collings).
However, cultural practices, status of owning a particular technology (such as PV), acceptability
of technologies and design versus needs, are claimed to be particularly important for whether an
energy technology will be adopted by users. Some authors suggest there can be misalignment
between technological design, on the one hand, and practices and needs on the other (e.g.
Rehman et al 2010). According to one informant, this is particularly relevant with cooking
practices and thermal needs, but less so for electricity (interviewee Brew-Hammond). Some
technologies can be seen as status-enhancing while others are the opposite, or the same
technology can be viewed positively by some and negatively by others. Green and Erskine (1999)
discuss the preference for accessing grid electricity instead of SHSs but also find that, for some
rural dwellers, owning a SHS creates enhanced status, which helps to explain increasing levels of
theft and vandalism.
Economic interventions
Finance and credit are suggested by many as ways to help make energy services cheaper for
poorer people in order, mainly, to overcome the upfront costs of conversion technologies and, in
some of the literature, to help with grid connection charges. For other energy services, subsidies
are usually suggested. There are also suggestions for coupling income-generating activities to
energy-service provision in order to increase the commercial viability for the service provider as
well as create opportunities for energy users to generate income, thereby helping them afford
better services. There is little strong evidence in this category of intervention, although some of it
is suggestive of the potential for success. Some of the studies are based on empirical research into
market dynamics; some are modelling or simulations; and others investigate specific
Acker and Kammen (1996), reporting on the Kenyan SHS market, give examples of where credit
was extended to dealers who were then able to extend it to SHS buyers and increase sales. They
supplement this with examples from other developing countries, using these to suggest that the
Kenyan utility could become involved in credit schemes for PV. Also discussing the Kenyan
experience, van der Plas and Hankins (1998) suggest that credit could be given for buying higher
quality systems, having found that these were more likely to continue to provide consistent
services than the lower quality or smaller systems. However, they suggest that the private sector
should also focus on providing smaller systems that are specifically designed to meet low power
needs as this was clearly an important entry point for many people. Unfortunately, no high quality
studies were found in the academic literature on sub-Saharan Africa of consumer-financing
experiences. One study in India does provide some discussion of consumer-financing. Rehman et
al (2010) investigate a market development project in which soft loans were provided from stateowned and cooperative banks, and a credit card was introduced, under schemes that were
sensitive to the variable incomes of farmers. They report that there was increased uptake of SHSs
as a consequence of these financing schemes. In a more general assessment of rural electrification,
the World Bank (2008) reports that credit and consumer financing schemes have been limited in
its interventions, suggesting that the evidence remains weak.
A few studies investigate the experiences of fee-for service or cost-recovery electrification
schemes, though the findings are mixed. The Global Environment Facility (GEF) supported a
project in Zimbabwe that intended to create a market for solar lighting systems targeting low
income rural inhabitants (Mulugetta et al 2000). The authors question the extent to which these
goals were achieved, suggesting that well-off rural dwellers were more likely to have benefited
from the fee-for-service approach. Also, without any integration of the project into a national
policy strategy or other measures of accountability and monitoring, the project was focussed on
creating incentives to meet its own target of installing 9000 systems, which led to marketdistorting behaviour. This resulted in little attention to longer-term efforts at local capability
building or ensuring adequate management, operation and maintenance for after-sales services.
Once the project finished, any capabilities that had been developed were quickly lost. The World
Bank report (2008) also highlights that the larger share of the benefits rural electrification yields
are captured by the non-poor. However, it is also suggested that cost recovery is achievable and
that there is a high willingness to pay for electricity in rural contexts.
Ilskog et al (2005) and Kirubi et al (2009) report on cost-recovery electrification schemes that
appear to have been more successful. Discussing a cooperative-led project in Urambo in Tanzania,
Ilskog et al identify a range of factors to explain success. Financial support for investments and
covering the initial problems with recovering operational costs, ability and willingness to pay
among consumers for at least the full operating cost of the service, and committed international
assistance from the Stockholm Environment Institute (SEI) and Swedish International
Development Cooperation Agency (Sida) were all important. Kirubi et al (2009) describe a
community-led diesel-powered micro-grid electrification scheme in Kenya, which received a
capital grant and technical assistance from Deutsche Gesellschaft für Technische
Zusammenarbeit (GTZ) and German Assisted Settlement Project (GASP). The scheme included a
number of productive elements. Based on their findings, they argue that such schemes have high
potential to reach adequate levels of cost recovery, particularly where they enforce cost-reflective
tariffs and link to productive uses that generate income and improve the load factor. They also
argue that the provision of capital subsidies in such programmes can be justified.
Subsidies are discussed by others, some for interventions other than electrification. Winkler et al
(2002), for example, model subsidies for energy efficient measures in low cost housing in South
Africa. They find that the high discount rates of the poor mean most energy efficiency subsidies
do not yield net benefits. Where benefits do appear, they are for measures that are already
attractive and so do not require subsidies. Hofmeyr (1995), however, calculated that costeffective electrification subsidies for worker households on electrified farms in South Africa were
possible, assuming low cost technologies and the use of on-farm labour and equipment. For unelectrified farms, PV systems could also be cost-effective and were cheaper per household than
the utility’s urban electrification costs. Dube (2003) attempts to model the effect of a subsidy
policy on the affordability of electricity in Zimbabwean cities. The analysis suggests that the
existence of subsidies is not decisive for the affordability of energy, although the removal of
subsidies would impact more negatively on vulnerable groups. A more significant barrier to
accessing electricity is the upfront cost, not recurrent costs (even without subsidies).
Finally, by contrast with the experiences in electrification in much of sub-Saharan Africa, Peng
and Pan (2006) describe China’s rural electrification programme. This began in 1949 and
developed through three stages. Initially, investment was primarily in irrigation and agricultural
production, and electrification was slow. The second stage saw the transfer of local electricity
systems to local government management. Alongside this transfer, small hydropower investments
enabled the electrification of 653 counties; and significant investment was directed to rural
poverty reduction. By 1998, the rate of access was above 96% in rural households, although more
than 8 million rural households still had no access. The final stage included construction in rural
electrified counties of hydropower stations, a rural grid renovation programme and the
introduction of government subsidies. Indeed, one interviewee mentioned that there were huge
amounts of transfer from rich to poor in China via government subsidies (interviewee Barnett).
Technical interventions
Interventions to address technical challenges in increasing the use of modern energy services can
be categorised into those dealing with hardware options and problems, those concerned with
capabilities of various kinds (‘software’), and those to do with regulations. The literature covers
these issues either by developing recommendations based on the problems identified in the field
or with reference to the experiences of specific interventions.
The World Bank (2008) assessment of rural electrification found that interventions leading with
hardware provision tended to be more successful than interventions related to institutional
development. However, the study shows that technical problems of infrastructure provision are
still critical and that the benefits of rural electrification do not follow automatically from
hardware availability.
On interventions dealing with software issues, studies from Kenya, once again, feature
prominently. Acker and Kammen (1996), as well as van der Plas and Hankins (1998),
recommend education and training on the supply side. The study by van der Plas and Hankins
(1996) found that SHSs installed by trained technicians or solar companies tended to be in better
condition than those that were self-installed and so suggest that training, which could be certified,
is likely to raise the quality of systems and hence service to the consumer. Acker and Kammen
include a number of other recommendations as well. They suggest that demonstration systems
would raise awareness about the benefits of PV, something that had helped to prime the early
Kenyan market, but a measure no longer well supported by development agencies. They also
speculate that Kenya could be helped to develop manufacturing capabilities in PV, beginning
with the encapsulation of cells. The hope would be to lower the price to the consumer, and they
cite experiences from other countries to support the view that this approach could work. Ilskog et
al (2005) identify training and other technical support as important factors in the success of the
project they studied in Tanzania. The staff of the cooperative managing the electrification scheme
was given training and further technical support whenever needed, including from the national
utility. In contrast, Mulugetta et al (2000) describe the GEF Zimbabwe SHS project as having
paid little attention to local training; focussing, instead, on short term project targets. As
mentioned in the previous section, once the project finished, any capabilities developed during
the project quickly vanished. Thus, although the evidence is consistently reported in the literature
that training and capacity building are needed, it is not clear what kinds of training and capacity
building these should be in different contexts and how these should be assessed.
But even if capabilities are developed, Haanyika (2008), discussing electrification by grid
extension or decentralised micro-hydro and PV in Zambia, notes that technical standards and
service levels might need to be lowered to take account of local conditions. This contradicts other
voices in the literature who call for the development and enforcement of standards, or point to
perceived benefits of standardisation. D’Agostino et al’s (2011) analysis of China’s Renewable
Energy Development Programme (REDP) claim that manufacturers perceived benefits from
the emphasis the REDP placed on standardisation and compliance for higher quality products.
However, these did not appear to trickle down to dealers or end-users who reported buying
products solely on price. Duke et al (2002) recommend performance testing of PV components
and public disclosure about their quality as a way to address information failure in the Kenyan
market. And a more general information failure issue was raised by an interviewee (interviewee
Barnett). Data on the field and context specific performance of technologies are not available to
the consumer (or others such as investors) making it difficult to make choices when faced with a
number of options. It is well-known, for instance, that a PV module loses efficiency as its
temperature rises but understanding what this would mean for a consumer in their particular
circumstances is practically impossible.
However, according to some authors in the literature, addressing capabilities or technical
performance issues in isolation is not sufficient. The discussion has already mentioned the
cautionary note from Murphy (2001) that an emphasis on the techno-economic viability of
particular options can detract from other important factors. Green and Erskine (1999), studying a
PV project in South Africa, observe that it was not clear the extent to which issues beyond the
technical aspects were addressed; issues such as engaging a broad range of actors who can create
the institutional and market structures needed to sustain the development of the project. Each of
these actors could be expected to have certain capabilities and access to information that could
facilitate learning from the market, as Rehman et al (2010) report from their study of a PV project
in India. As a result of such linkages, learning from the market and after-sales service facilitated
the development of 12 new products, many of which were more affordable by lower income
groups. The findings from the analysis of China’s Renewable Energy Development Programme
by D’Agostino et al (2011) also reinforce the need for retrospective learning and research into
robust and pragmatic life-cycle approaches to renewable energy programmes. Diniz et al (2011)
also report on their experience in implementing a PV rural electrification programme in Brazil
and conclude that a range of interrelated criteria addressing costs, capabilities and social factors
are most likely to lead to successful outcomes.
Political/Institutional interventions
Interventions to address political and institutional barriers are wide ranging, although not
exhaustive. There are many to do with strategic aspects of policy making, and also concerned
with policies and institutions themselves from national down to the level of the firm. The weakest
part of the literature appears to be at the political level where it is presumably more difficult to
implement interventions.
From the strategic perspective, the interventions seem to be derived mainly from observations of
problems in the field and not necessarily systematic empirical studies. These recommendations
centre around the ideas of integration and coordination. The most general of these is that there
needs to be stability in the institutional policy making environment; a recommendation arising
from the difficulties experienced in rationalising Zimbabwe’s tariffs (Mangwengwende 2002).
Mulugetta et al (2000) suggest that projects should be aligned with national frameworks and
policies in order to raise the chances of their impacts being sustainable. Kirubi et al (2009), and
two interviewees, suggest that energy service policies should be integrated or coordinated with
other development policies (interviewees Brew-Hammond, Barnett). These might include policies
for agriculture, health, education and commerce, and there should be more sector-wide
approaches involving coordination between bilateral donors and national governments. One other
suggestion is that it would make sense to start with local institutional structures that are seen to
work and then build on these (interviewee Lockwood). In terms of starting with local institutions,
the example of China’s electrification programme is perhaps instructive. It is commonly
understood that China’s governance is heavily centralised and it was from the centre that the
electrification programme began. However, over time, management responsibility was
successfully transferred in phases from central government to the local level (Peng and Pan 2006).
At the level of policy itself, power sector reform receives some attention. However, it has a
patchy record. Karekezi and Kimani (2002) discuss some of the impacts such reform can have on
the poor, especially where tariffs are increased steeply. They cite examples from Ghana, where
tariff increases caused riots, and Kenya, which experienced protracted political debates about the
reform. The World Bank (2008) also found that interventions operating at the institutional level,
primarily regarding financial sustainability concerns, have been less successful. One interviewee
observed that much of power sector reform has been focussed on ownership to the neglect of
governance; that private ownership will not, in itself, necessarily change the problems of
politicisation and patronage that the energy sector has attracted in so many countries (interviewee
Lockwood). Nevertheless, privatisation of some kind may open opportunities to address some of
the challenges of infrastructure development, as reported by Lee et al (1999) in their comparative
study of self-provision of power in Nigerian, Indonesian and Thai firms. Indeed, such reform
could enable new business models of energy-service delivery at the micro-level, as suggested in
Kirubi et al (2009) and Duke et al (2002), helping to develop supply chains as witnessed in the
Indian SHS project analysed by Rehman et al (2010). Of course, firm level activity in a private
market has been seen to create problems of poor quality products and information asymmetries.
This was seen in the Kenyan PV market, but Duke et al (2002) suggest these problems can be
mitigated through measures such as standards, certifications, voluntary actions of companies,
warranties, testing procedures and disclosure.
The weakest area of the literature examined here is that concerned with the political level. It is
clear that many studies conclude there is a need for political will, and this was also expressed by a
number of interviewees, whether at the national level or the local (interviewees Brew-Hammond,
Yadoo, Lockwood, Barnett). Indeed, Ilskog et al (2005) claimed that institutional commitment
and strong local leadership were important success factors in the cooperative-led rural
electrification project they studied in Tanzania. However, there seems to be little in the literature
selected for this review that demonstrates interventions to overcome more high-level political
difficulties. For example, Hofmeyr (1995) reveals the problems created by Apartheid for farmworker families but gives no examples of how to increase the political power of marginalised
groups. Rather, the paper recommends greater participation and recognition of the status of farm
workers as individuals. Such interventions might overcome the power asymmetries against
marginalised groups but no evidence was captured in this review, suggesting that either such
evidence should be specifically targeted by other reviews or there is a need for further study into
governance issues around energy access.
Cultural/Social interventions
Strategies to overcome cultural and social barriers fall into a number of categories. Marketing,
awareness and education are all linked and often recommended in the literature. Not so
commonly reported are strategies that learn from users and the market, although they do feature
to a certain extent and were mentioned by a number of interviewees. And, linked to this, some
discuss the importance of understanding needs and designing products to suit those needs. Some
of these strategies overlap with other interventions discussed above, which is perhaps testament
to the deep-seated and pervasive nature of the importance of these aspects in the adoption and use
of modern energy services.
Marketing, awareness and education can all be seen as forms of information and, as such, play
overlapping roles. Of course, they are distinct as well. Marketing, for instance, serves to sell a
product and this can lead to inflated claims about performance, as noted in the above discussions.
Measures to address this information failure have already been discussed above and include the
recommendations of Duke et al (2002) in trying to reduce the impacts of poor performing
modules on the Kenyan PV market. Dwelling on the Kenyan market, calls for reducing the
information deficit have been made since at least 1996 in, for example, the paper by Acker and
Kammen (1996), who suggest that the provision of independent information and advice would be
a cost-effective way to reduce the perceived risks to consumers. Similarly, van der Plas and
Hankins (1998) recommend the provision of information on the real costs and benefits of PV,
except that they suggest this be done by importers and retailers. The importance of information
amongst users is underlined in the findings in a study of LED lantern marketing in Malawi by
Adkins et al (2010). Their survey of 54 households that purchased a lantern revealed that 21 had
heard about or seen a lantern through a focus group discussion or sensitisation meeting, 14 at a
vendor’s store, 12 through a friend or acquaintance and seven through other sources (Adkins et al
2010, p1091).
The benefits of learning from the market are expressed in Rehman et al (2010), especially in
developing branding, which is shown to have had a significant effect on sales. Brand promotions
through simple slogans, paintings on walls, banners, newspaper articles and many other
mechanisms were used. Increased sales followed: before branding, sales were INR 0.46 million
and INR 0.59 million in Bikaner and Rishikesh, respectively; after branding, sales were INR 3.61
million and INR 2.43 million, respectively. Dealer response was also large after branding, with
products crossing into seven adjoining districts. Green and Erskine (1999) question whether
marketing and promotional activities were well designed in the SHS project they studied in South
Africa, where the authors suggest that efforts were too focussed on the technical aspect of the
technology and not on encouraging rural inhabitants to buy SHSs.
The notion of designing for user needs is also evident in van der Plas and Hankins (1998). They
recommend private actors to develop small PV systems that have deeper-discharge batteries
specially designed for the energy-use patterns of many Kenyan households who are buying
systems piecemeal. Likewise, Acker and Kammen (1996) suggest that small systems such as
solar lanterns could be a way to deliver small amounts of power in quality systems to users. And
a number of interviewees expressed the importance of aligning with both user needs and cultural
practices (interviewees Collings, Brew-Hammond, Yadoo), although one interviewee also
distinguished between thermal and cooking practices, and electricity use (interviewee BrewHammond). He asserted that cultural factors play a larger role in thermal and cooking practices
than in electricity use.
Finally, a small amount of literature focuses on gender specific interventions. For example,
Nygaard (2010) describes electricity generation programmes under the management of
women’s groups. The study found that the women’s groups faced difficulties in operating and
managing the multi-functional platforms supported by donor funding. However, it is not clear
from Nygaard’s discussion whether the problems were related to gender. The argument, instead,
suggests that the fundamental issue was that community-group management of businesses was
unfamiliar in the particular contexts studied. This management form had been imposed by donors.
Nygaard argues, therefore, that development aid should build on local structures rather than new
institutional forms; something also supported by one of our interviewees (interviewee Lockwood).
As noted at the beginning of this review, modern energy services are not easily defined. It is
apparent from the discussion in this section on cross-cutting issues that they depend on a variety
of different societal needs to be fulfilled (at the domestic, commercial or other productive levels)
which, in turn, depend on the context in which the services are needed (e.g. urban or rural) as
well as the existing technological, institutional and social capabilities to develop interventions.
Furthermore, energy carriers and technological options are diverse, adding complexity to the
analytical and normative nature of understanding barriers to, and interventions for, increased use
of modern energy services.
We also noted that our classification of barriers and interventions is not straightforward. In many
of the discussions it is apparent that barriers interact with each other, and interventions do not
map easily from particular barriers, howsoever they are conceived. This, together with our
observation of the complexity of modern energy services, points to the importance of
understanding barriers and interventions in integrated ways. This is something that was raised by
some papers in the literature in regard to political and institutional strategies, and underlined by
several informants. Therefore, our categorisation of barriers and interventions is likely to have
missed the interlinked nature of the factors limiting the use of modern energy services, and the
integrated and systemic characteristics of interventions to overcome these limiting factors. To
some extent, this reflects the state of the academic literature: there seems to be a relative lack of
high quality research, particularly in an African context, which attempts to pull together such
integrated understanding.
However, it also seems to reflect a potential problem with the concept of a barrier itself. One of
our interviewees noted that the world’s poorest people face difficulty translating their needs into
market demands, precisely because of their poverty (interviewee Barnett). A barrier implies the
restriction of some process – in this case, the translation of human needs into market demands.
But, as a market demand is expressed with money, and the poorest people do not have money, the
barrier concept becomes problematic. Much of the analysis presented in the literature of barriers
and interventions is focussed on market solutions to increasing the use of modern energy services.
Such interventions rest on making new sources of finance available to stimulate market demand –
either by increasing incomes of the poorest people or through some form of financial assistance.
The important consequence of this is that simply removing many of the technical, social and
institutional barriers to energy access discussed in this review will not change the low incomes of
the poorest people. This reinforces the need for systemic or integrated strategies to improve
access to modern energy services by the poor which provide them with long-term, sustainable
opportunities to generate surplus cash income.
Nevertheless, and bearing in mind these caveats and the limitations of a systematic review
discussed in Section 5 below, Table 6 presents a summary of the nature of the evidence which is
present in the 41 papers reviewed and synthesised above about the different barriers, and
interventions that might overcome them. The criteria used to determine the relative strength
and/or consistency of the evidence are as follows:
Consistent: Evidence from the 41 papers reviewed is regularly and consistently reported about
the nature of the barrier or intervention to address it.
Proof of concept: Evidence from the 41 papers reviewed is empirically weak, but the few pieces
of empirical evidence that do exist are strongly suggestive of proof of concept for the barrier or
intervention to overcome it.
Contradictory: Evidence from the 41 papers reviewed is strongly contradictory, with strong
evidence for contradicting arguments about a barrier or intervention.
Mixed: Evidence from the 41 papers reviewed is mixed, with both weaker and stronger evidence
presented, but not enough of each to determine an overall strength assessment.
Weak: Evidence from the 41 papers reviewed is weak with only a few papers providing evidence
about the nature of the barrier or intervention.
In the conclusions section (see Section 6), we further elaborate on the policy implications and
research challenges that arise from our findings. Where relevant, the criteria above are also used
to characterise the relative strength and consistency of the evidence that backs up these
Table 6: Synthesis of Evidence and Strength Criteria
Type of
Upfront cost
Economic Maintenance and
(demand-side) operational costs
Inadequate cost
(supply-side) High cost of
How the barriers are characterised in the reviewed literature
Comments about the strength of the evidence, and cross-cutting
High cost of energy technology hardware (reported for both off grid – renewables Consistent: This barrier is strongly and consistently reported in the
and diesel gen-sets – and grid extension) is a barrier.
reviewed literature.
This type of barrier is characterised by high operational and maintenance costs of
rural electrification and self provision of electricity in industry, but the argument
is contested. Some authors indicate the inability of low income users to pay for
maintenance and operational costs (Haanyika 2008; Kirubi et al 2009) whereas
others suggest there is high willingness to pay for electricity service even in poor
areas (Ilskog et al 2005).
Mixed: The evidence about high operational and maintenance costs of
energy service provision is regularly mentioned in the reviewed literature
and there are some discussions about the efficiency and efficacy of
subsidies. However, this type of barrier seems to be very context specific.
There are several reasons explaining this type of barrier such as low population
densities, distance and dispersed nature of rural populations, low load factors,
and non reflective tariff structures.
Consistent: The literature addressing this barrier suggests integrated
approaches might work better in which productive and income generating
activities are linked to energy service provision.
This barrier is linked to high upfront costs (demand-side) and authors refer to
lack of access to finance, low income levels of customers, inadequate levels of
public capital.
Proof of concept: As per rural electrification (and electrification in
general), there are some successful cases in which public funds and access
to credit have been used. The reviewed literature suggests these initiatives
have to be linked to institutional and policy frameworks.
Quality and
performance of
This type of barrier refers to quality and performance of hardware. It is mainly
Consistent: This barrier is consistently referred in the literature and is
discussed in the case of SHSs (low quality equipment and installations: Acker
linked to availability of information about technology, quality standards,
and Kammen 1996; van der Plas and Hankins 1998) and MFPs (Nygaard 2010)). and warranty and certification/enforcement schemes.
In the case of grid electricity, this refers to poor performance and unreliability of
service (Lee et al 1999).
Low technical
capacity to
adequately maintain
and operate energy
This software barrier is referred to in many different aspects, including: low skills
and knowledge amongst end users and local technicians in the case of SHSs and
other decentralised energy technologies (MFPs, micro-grids); low capacity of
public utilities to operate and maintain power stations and electric networks;
dependence of donors' technical support; and poor managerial skills to provide
adequate after-sales services (Green and Erskine 1999, Haanyika 2008, Lee et al
1999, Foster and Steinbuks 2009, Nygaard 2010).
Consistent: This barrier is consistently referred to in the literature. It is
linked to institutional factors (support from public policies, standards,
sector reform, etc.) but there seems to be little thorough research analysing
the inter-relationships.
Political and Institutional
Institutional weaknesses,
corruption and
vested interests
There is little empirical evidence in the literature, but some interviewees
Weak: This type of barrier is mentioned often in the literature but there is a
highlighted a lack of political will (Brew-Hammond; Yadoo), and others refer to lack of empirical research focusing on politics and power balances and the
political interference (Kirubi et al 2009), suggesting that political barriers are
impact on transitions to modern energy services
important. Also mentioned by some interviewees was the power of voting blocs
which could hamper the implementation of cost effective measures to improve
access to energy services (Lockwood; Barnett).
These barriers refer to difficult to implement sector reforms, inadequate delivery Proof of concept: There are examples in the literature where sector reforms
capacity and lack of accountability of energy policy.
have involved the institutionalisation of access to electricity (mainly rural
electrification) through specific institutions or agencies, devoted funding
mechanisms and enactment of legislation and regulation. However, much
of this is descriptive and there seems to be a lack of research and analysis
on the dynamics of such institutional processes and the impacts of policies.
Political invisibility This barrier refers to the lack of power and mechanisms to bring rural
of rural population populations/farm-workers into energy policy deliberation/implementation
(Hofmeyr, 1995).
Weak: There is evidence of this situation in only one paper reviewed,
although the evidence is convincing in this particular case. If this kind of
situation is more widespread (which is likely) then it is an important area
for more thorough research.
Misalignment of
This barrier arises because energy products (technological designs) do not always
cater to regional and local needs, as interventions are designed and implemented
system design and from centralised institutions (Rehman, et al, 2010).
social practices and
users' needs
Proof of concept: There is some evidence of the mismatch between
technological solutions and the fulfilment of needs and consideration of
local practices. However, the reviewed literature does not put a particular
emphasis on the issues of strategic approaches and systemic understanding
of the framing of access to energy
Desirability of
This refers to barriers which arise from the relative desirability of energy
Contradictory: The evidence is contested in this area. There is a lack of
energy technologies technologies and include the status enhancing role of owning PV technology in more comprehensive analyses aimed at understanding cultural preferences,
rural areas and problems of theft which are linked to this. It also addresses
and linking needs to impacts of access to energy.
whether grid or off-grid technologies are desirable, and in what circumstances.
For example, PV normally provides limited electricity and therefore cannot fulfil
users’ needs or demands.
Type of
How these interventions are characterised in the reviewed literature
Access to finance The literature refers to financial/economic mechanisms to overcome upfront cost
and credit
of (energy conversion) technology/connection costs. There are examples of access
to credit to PV dealers who extended it to customers and credit to access better
quality equipment. There was no academic study found about consumer-financing
experiences in SSA, but evidence of soft credits in India has been reported.
Subsidies and
The literature demonstrates that capital grants/investment subsidies are important
in the case of expensive electrification and other energy services. It is argued that
capital support works better when provided alongside technical assistance and
mechanisms to recover operational costs (Kenya: Kirubi et al 2009; Tanzania:
Ilskog et al 2005). However, it is also shown that subsidies might not yield net
benefits in the case of energy efficiency (Winkler et al 2002). In the case of
increased use of electricity, modelling suggests that subsidies are not decisive as
upfront costs are probably more significant (Dube 2003).
Though broadly promoted, it is suggested that this type of approach might be
ineffective in targeting the poor/poorest rural populations, thereby decoupling the
social role of access to energy and market development/technology diffusion.
Linking energy
services to
options and
Software or
Regulations and
These type of interventions improve the ability to pay; the likelihood of success of
cost-reflective tariffs and cost-recovery. Productive activities improve load factor,
thereby improving financial viability of electrification projects. Evidence is
provided from China (Peng and Pan 2006) and Kenya (Kirubi et al 2009).
Interventions tend to focus on the provision of hardware by means of market
development or government or donor-led interventions. However, quality and
performance of equipment is frequently raised as an important issue to be
addressed for both grid and off-grid electrification.
Comments about the strength of the evidence
Mixed: There seems to be an emphasis on interventions focusing only on
economic incentives/mechanisms, which might result in a lack of
consideration of other factors such as capacity building, adequate
management and maintenance/operation.
Proof of concept: There is isolated but strong evidence on the importance
of well designed subsidies when poor people are targeted to improve
access to modern energy services.
Mixed: There is some evidence about fee-for-service interventions and the
high willingness to pay, but it is also suggested that these interventions do
not reach poor people as intended.
Proof of concept: This sort of intervention is in general supported and
suggested but there is not enough empirical evidence interrogating the
processes underlying effective interventions.
Consistent: The evidence is strong however there seems to be lack of
comprehensive ex-post evaluation of interventions. This type of
intervention is clearly linked to the existence of adequate regulations,
standards, information and training, suggesting there is a need for more
integrated and systemic approaches to interventions.
Interventions of this type involve training and education of technicians which may Mixed: Strong evidence on the importance of training and development of
raise the quality of installations, provide better service to consumers and raise
technical capabilities. However, the provision of training alone does not
awareness and knowledge. This is particularly seen in the case of PV in Kenya
guarantee good service provision or adequate technical support as
with demonstration systems (Acker and Kammen 1996; van der Plas and Hankins capabilities might vanish in the absence of institutional support structures
1998) and in Tanzania (Ilskog et al 2005). However, Mulugetta et al (2000) argue and/or opportunities to continue practising. There is not enough known
that any capabilities acquired vanish quickly after project assistance finishes if
about what kinds of capacity development is needed, and how these
assessments are done.
those capabilities are not fostered with a view to the local context.
These types of interventions refer to consistent application of regulations and
Contradictory: Many authors warn about the need to address technical
standards which might help to ease technical barriers. For many types of energy
interventions alongside other factors such as engagement of a diversity of
technology (grid, PV, mini hydro), some authors suggest that technical standards actors, socio-cultural, and interrelated economic/market conditions.
might need to be lowered to better fit local conditions such as households' quality However, others suggest social and technical/capabilities approaches are
and materials and use of local materials (Haanyika 2008), but others argue that
most likely to generate learning and lead to successful results.
standards and certification of energy equipment must be developed and enforced
(Duke et al 2002).
Strategies and
policy making
Marketing and
participation and
market learning
This type of intervention is mainly centred around recommendations for strategic
policy making, including: policy coordination and integration between energy and
other sectoral/development policies (Kirubi et al 2009), long term policy stability
and integration with national frameworks (Mulugetta et al 2009); and tariff
rationalisation (Mangwengwende 2002). Power sector reform has been widely
introduced across the world and increasingly in SSA. However, much of the
reviewed literature does not refer to governance and impacts of the reforms on the
poorest segments of the population (Karekezi and Kimani 2002; Lee et al 1999).
Interventions of this type refer to politically motivated power sector reforms.
There is evidence of political debate and riots due to some implications of power
sector reforms (i.e. tariffs increased), but no systematic study of the politics of
such reforms was found. However, some studies and experts conclude that
political commitment and leadership at different levels are keys to success in
improving fair access to energy (Ilskog et al 2005; Hofmeyr 1995) but there seems
to be little in the literature that systematically investigates interventions and
political dynamics that help overcome these hindering factors.
This type of intervention refers to the establishment of appropriate institutional
arrangements and structures. There seems to be no consensus on whether the
institutional structure has to follow bottom-up or top-down approaches. Instead,
the analyses and expert opinions suggest that interventions at various levels may
be necessary; the important consideration being whether they are aligned with
each other or not.
In general, reported interventions in this area aim to address an information failure
about technology options, potential and real benefits of energy technologies. This
can be done by different actors. Duke et al (2002) and Acker and Kammen (1996)
suggest reducing the risk to consumers through independent provision of
information about performance, relying on improvement of the market. Others
rely on the role of technology suppliers (van der Plast and Hankins 1998). Adkins
et al (2010) show the variety of means consumers use to access information about
LED lanterns.
This type of intervention focuses on understanding users’ needs/wants and
learning from the market. For example, Rehman et al (2010) show the benefits of
developing branding leading to increased sales. Green and Erskine (1999) and
D'Agostino et al (2011) suggest that effective marketing needs to focus not only
on technical aspects but also on how the information is used and internalised by
customers so they make purchase decisions beyond price factors.
Proof of concept: There is a patchy record on interventions focusing on
strategic policy making. However, it is clear that where strong, focused
and stable policies have been developed, access to modern energy services
has improved and is likely to have better impacts on development
processes. There seems to be a need for further research about different
approaches to policy-making (top-down/centralised vs. bottomup/decentralised) and at different levels (international/donors; national
governments; local authorities/government).
Weak: This seems to be the weakest area in the literature, but one that is
frequently considered critical, and therefore is identified as an important
area for further research. Many of the cases are given as examples and
anecdotal evidence.
Proof of concept: Suggested by many authors, but there is no particular
literature addressing institutional change, the adequacy of institutional
structures and the alignment of institutions to local conditions and
national/donors’ imperatives.
Proof of concept: There is consistency about the importance of marketing
and awareness raising at different levels, and focusing on different actors:
from users, to market actors, to policy-makers. However, there seems to
be no linking between these types of interventions, and more integrative
and systemic approaches that involve coordination at the policy-making
level to support market development and the involvement of users.
Proof of concept: Many authors mention the need to adequately involve
users in the process of developing and diffusing energy technologies,
particularly renewable energy. However, beyond recommendations on the
importance of delivering information and learning from users (and the
market), it is not clear the extent to which existing research has addressed
the role of users in the shaping of technology and the role they play in the
development of policies and projects, or how cultural factors affect the
design and performance of energy technology.
5. Discussion
The conduct of the study was limited in some ways due to the broad scope of the review question,
the breadth to be covered in both the academic and grey literature, and the resource constraints on
the project. The searches and the analysis were therefore limited in several ways, including a
greater emphasis in the academic literature on Sub-Saharan Africa and a targeted focus in the
grey literature on interventions and lessons learned in Sub-Saharan African countries. There is
also private sector evidence and analysis in existence on the topic, but this was inaccessible for
the purposes of this project. It also emerged from the interviews and the peer review that it seems
in this field in particular there is more ‘tacit’ knowledge that rests with practitioners and is highly
location specific, therefore making it difficult to ‘manualise’, or in other words report formally in
the literature which would be captured in this kind of systematic approach.
Future enquiry and analysis might therefore focus more strongly on unearthing this kind of tacit
knowledge, in addition to barriers and interventions in BRICS countries, and on the vast body of
grey literature which exists for these countries. This type of work would be highly
complementary to the present review and together could provide a good evidence base from
which to make future policy decisions.
In addition to the limitations in scope, related methodological limitations exist, including
limitations in the search terms and in the heterogeneity of the studies. The latter is due to the fact
that the majority of studies reviewed were qualitative in nature and heterogeneity of study type
and output prevails. For qualitative studies, it was not easy to determine the factors which
contribute to methodological rigour because of the diversity of disciplinary approaches present in
qualitative research. In quantitative research, problems related to analytical approach and study
design are usually discussed in terms of reliability and validity (both internal and external). These
terms are less appropriate for qualitative research. Instead ‘equivalents’ were used such as
dependability/audit ability in place of reliability, credibility/authenticity in place of internal
validity, and transferability/fittingness in place of external validity (Miles and Huberman 1994;
Lewis and Ritchie 2003). However, inevitably the nature of such assessments may have the
implication of some studies which provided useful insights, but did not meet the full quality
standard, were not included. In addition, due to the experience and expertise of the review team,
there were limitations in the extent to which a more detailed quantitative analysis of the review
papers could be done. Future reviews in this area might conduct a much more targeted analysis.
Other methodological limitations include the limiting of the study search to papers published only
in English, as well as the need to limit the scope of the search terms by introducing the strategy of
including ‘energy’ or ‘electricity’ alongside every set of search strings. While the former may
have the effect of biasing the results towards English speaking countries, the latter should have
prevented the exclusion of some types of energy services which are not explicitly related to
electricity generation, however we recognise that there may, nevertheless have been some
limitations this restriction introduced. In addition, any study which includes an analysis of grey
literature is limited in the extent to which the grey literature can be adequately covered, the
databases vary significantly and cannot be searched in similar ways, and the quality of the reports
is often difficult to judge. The review team has taken steps to address this by using different
combinations of search terms and triangulating our search with stakeholder interviews to ensure
the most relevant grey literature was captured, but there will inevitably be reports which the team
was not able to capture. A future review might focus solely on the grey literature so that all
available resources could be concentrated on this data source.
6. Review Conclusions
With respect to barriers to modern energy services some of the main policy implications are
discussed first.
1. For the sake of clarity, the review has analysed barriers and interventions in four separate
categories (economic/financial, technical, political/institutional and social/cultural). This
separation is reflected in many of the papers reviewed. However, the evidence base also
includes many instances where these barriers are inter-related. For example, the economics of
new energy hardware is often dependent on the political and institutional context in a
developing country. It is therefore important that the analysis of barriers from a policy
perspective should take these inter-relationships into account, and would benefit from taking
an integrated approach to the analysis of barriers and hence, to interventions.
2. The review has analysed a wide range of barriers to modern energy services. It is difficult
from the evidence base reviewed to produce a simple hierarchy of such barriers which
identify their relative importance. However, there is an understandable emphasis on economic
and financial barriers as being the most pervasive and important. One obvious reason for this
is that the poor have very limited incomes. This limited spending power makes sustainable
improvements in energy access particularly difficult to achieve, even if the costs of modern
energy services are significantly reduced.
3. Within the category of economic and financial barriers, the upfront costs of new energy
hardware (whether for electricity generation or energy efficiency on the supply side and
energy efficient devices on the demand side) are a particular problem. We found the evidence
for this to be consistent and strong. Access to finance to help users cover these costs is often
limited, which may in many cases be linked to political/institutional barriers. Other costs are
important too, including operational costs. The evidence reviewed suggests that even if
upfront cost barriers are overcome, ongoing operating costs can be too high for owners in
some cases (though this depends on the technology in question, and what kind of service
model has been used to finance it). We found the evidence on operational and maintenance
cost barriers to be mixed, with their importance heavily dependent on context. In addition to
these demand side barriers, the review found consistent evidence of supply side economic
barriers associated with inadequate cost recovery for energy service providers; and proof of
concept8 evidence on the high up front costs of investing in infrastructure.
4. Within the category of technical barriers, the literature reviewed makes it clear that these are
both hardware and ‘software’ (skills) related. There is consistent evidence from the review
that problems have arisen due to the poor performance of hardware, either due to its low
quality, a lack of information or a lack of certification schemes. As this evidence suggests,
skills and capabilities for installation, operation and maintenance of energy hardware are very
important too – and there is often insufficient attention to these when new hardware is
supplied. Again, there is consistent evidence for the importance of these operation and
maintenance barriers, though there is a lack of research on how institutional factors contribute
to these barriers.
5. Institutional and political barriers are extremely varied within the evidence base reviewed. In
some cases, papers emphasised the constraints imposed by particular institutional
arrangements (e.g. the lack of power sector reform that restricts private provision). Overall,
Implications for Policy/Management
Our definition for ‘proof of concept’ is given in the summary at the end of section 4.3 above, along with the other
assessment terms. However, for convenience, the proof of concept definition is repeated here: “Evidence from the 41 papers
reviewed is empirically weak, but the few pieces of empirical evidence that do exist are strongly suggestive of proof of
concept for the barrier or intervention to overcome it.”
the evidence is proof of concept within this sub-category. Barriers due to a lack of political
will, political interference and the power of political opposition are often mentioned, but we
found that the specific evidence was weak with respect to these barriers. Related to this, there
is weak evidence about the extent to which effective policy making for rural electrification is
hampered by a lack of power amongst rural populations. It is clear from the papers reviewed
that institutional arrangements and the prevailing political climate can have very large
impacts on other barriers – particularly those that are economic and financial.
6. Cultural barriers are also closely related to other barrier categories in the review, though they
bring up distinctive issues. Whilst many papers mention such barriers, the evidence for the
importance of these barriers is proof of concept (in relation to the misalignment of technical
system design and social needs) or contradictory (in relation to the desirability of particular
energy technologies amongst potential users). As noted above, there is also consistent
evidence that there are social barriers associated with a lack of operations and maintenance
capabilities for new technologies.
Turning to interventions to overcome these barriers to modern energy services for the poorest
people, the review comes to the following conclusions.
1. Given that there is significant evidence that different categories of barrier are related, it is
important to approach interventions in an integrated way. This evidence, which comes from
studies of a number of different countries, points to a need for interventions that address
interdependent economic, technical and social factors. Added to this, it is important that any
desire to reach general conclusions about successful intervention strategies does not
downplay the importance of local context. The diversity of evidence in this review illustrates
that it is unlikely that a ‘one size fits all’ approach to overcoming barriers to modern energy
services will be effective.
2. A range of economic strategies and interventions have been successful, many of which
included funding for the installation of new equipment. Some of the studies reviewed said
that these are more likely to work if users are willing and able to cover the full operational
costs of new equipment (but not necessarily the upfront costs). Overall, the evidence is proof
of concept for the effectiveness of subsidies and grants (i.e that there are isolated examples of
strong evidence of this), and mixed for the effectiveness of interventions that strengthen
access to finance and credit. Alternative fee-for-service approaches to overcome economic
barriers are discussed in the literature, but the evidence for their effectiveness is also mixed.
The evidence suggests that this approach is not effective in reaching the poor. Some studies
also emphasised the importance of supporting income generating activities through energy
access, though the specific evidence for the effectiveness of this is proof of concept.
3. With respect to technical interventions, there is consistent and strong evidence that the
provision of hardware needs to be accompanied by interventions to promote sustainable
performance over time. These include regulations, standards information and training. The
evidence on the latter is mixed. Whilst there is strong evidence that training to develop the
capabilities of technology users is important, it is also clear from the evidence reviewed that
training alone does not guarantee good performance. There is a lack of evidence on what
kinds of user capacity development are needed to improve outcomes. There is contradictory
evidence on the need for the enforcement of regulations and standards to overcome technical
barriers to the adoption of new technologies.
4. This leads on to interventions relating to institutions and the policy/political environment in
which these interventions take place. The evidence that strategic policy making and policy
co-ordination (e.g. to link energy access policies to wider development strategies) is effective
is proof of concept. However, the evidence stresses that policy stability has an important role
to play in improving access to modern energy services. There is a need for further research to
understand the relative effectiveness of different approaches to policy making. Whilst power
sector reforms are often an important focus for debate and are controversial, the evidence in
the papers we reviewed is weak and anecdotal on how the politics of these reforms influences
their effectiveness.
5. With respect to electrification programmes in sub-Saharan Africa, the BRICs literature
reviewed suggests that there may be lessons to learn from the experience of China, but also
from Brazil and India in relation to smaller scale electrification programmes, particularly
those involving renewable energy technologies. One of the papers reviewed showed how a
step by step approach was taken by the Chinese government over a long period of time to
improve electrification rates. Of course, care is needed in drawing such lessons – not least the
differences in economic, political and institutional context in China when compared to many
lower income countries.
6. The evidence that interventions to address cultural and social barriers are effective is proof of
concept. In other words, there is some isolated evidence that these interventions are effective,
but the evidence base in weak in general. There is evidence that interventions to market and
raise awareness of new technologies can be effective in some cases, but the analysis of these
interventions does not seem to consider them together with other, complementary
interventions that might be required. Many studies also mention the need to involve users in
developing and diffusing energy technologies, but there is a lack of research on how users
affect technologies, policies and projects.
Implications for Research
This section summarises some implications for further research, with a particular emphasis on
some of the gaps in the evidence base that have been identified. These implications include:
In general, the evidence base found to answer our question was patchy in quality, consistency
and strength. The criteria used to assess this evidence base are given in Appendix C, and
discussed in detail in section 3.4. High quality studies incorporate credibility and defensibility
of design, a transparently reported methodology and coherence of analysis. It would be
especially helpful if studies – whether in the academic or practitioner literatures – were to
adopt designs and analytical approaches that meet these criteria. Having said this, there are
some very good studies available which provide rigorous analysis of both barriers and
interventions. But, given the volume of literature in this field, the number of high quality
studies that are sufficiently rigorous to meet the criteria we applied in this review is limited.
This is particularly the case with respect to high quality studies that take an integrated
approach to the analysis of barriers and interventions. As noted above, many of the studies
reviewed illustrated the inter-relationships between the four named categories of barriers and
interventions, but there is a tendency to analyse these categories separately.
There are several specific areas where further research is required to strengthen the evidence
base, including:
a) Consumer financing (e.g. of solar home systems), for example via credit schemes;
b) The strength of information barriers to the adoption of new energy technologies;
c) The importance of political barriers, for example due to vested interests or corruption, and
how these impact on communities with little political power;
d) The interaction between technical and cultural barriers to the adoption of new technologies;
e) What specific capabilities are required to support the adoption and diffusion of new
technologies to improve energy access. These specific capabilities could include
capabilities for operations and maintenance, manufacturing, design, policy development
and implementation, and running businesses.
The grey literature focusing on energy access is particularly extensive, and this is likely to
continue to grow – particularly given that 2012 is the UN year of sustainable energy for all.
Whilst it is fair to assume that the academic literature has been subjected to higher standards
and independent scrutiny, some grey literature has also undergone rigorous peer review. It
adds significantly to the available knowledge base, but resource limitations mean that it could
not be comprehensively reviewed for this study. A further systematic review of this literature
might be warranted given its scope and potential value.
The evidence base found is not evenly distributed from a geographical point of view. Some
sub-Saharan African countries are a particular focus for academic and grey literature (e.g.
Kenya), and therefore there may be opportunities to focus more in future on those countries
that are relatively less well studied.
Most of the literature and research on energy access that was reviewed focuses on electricity.
Whilst this literature is highly relevant and important (not least because providing affordable
energy access to rural areas is particularly challenging), this coverage is not comprehensive.
Significant gaps in the literature are apparent with respect to other energy services such as
heating, communications, and productive uses of electricity.
There is surprisingly little literature that seeks to learn lessons from the BRICs countries for
lower income developing countries. Whilst care is needed in drawing any lessons of this kind
(as noted above), there is potential for more research in this area. One possible example (also
noted above) is the long term Chinese electrification programme – and how it has been
relatively successful in achieving its aims.
Finally, some of the papers reviewed emphasised the need for much better monitoring and
evaluation of interventions that have been made – particularly with respect to their
effectiveness. These evaluations are not only required to fulfil donor requirements for
reporting and budgetary control. Genuinely independent evaluations could also strengthen the
evidence base in general – and help to underpin more effective interventions in future.
7. Competing interests
One team member, Jim Watson, at the Sussex Energy Group has published widely on the more
general topic of energy and development. Relevant publications are listed below in Appendix D.
Rob Byrne, another team member, managed a small solar energy project in Tanzania between
October 1997 and October 2000 and was a founding member of the Tanzania Solar Energy
Association (TASEA) and served on its interim executive committee for several months before
completing a contract in Tanzania (2000). TASEA has now become the Tanzania Renewable
Energy Association. It is a membership organisation, involving public and private sector actors,
as well as individuals interested in promoting renewable energies for development in Tanzania. In
order to mitigate against the risk of biasing our review, we have built in a three-person review
process, with additional research oversight by a fourth team member, to ensure objectivity. Other
authors declare that they have no competing interests. A full list of relevant, and potentially
conflicting publications by study authors, is given in Appendix D.
8. Acknowledgments
This study is funded by the UK Department for International Development (DFID). We thank Mr.
David Woolnough, the project lead at DFID, for engaging with us throughout the study, as well
as his colleague, Leanne Jones; and Barbara Livoreil and Andrew Pullin at the Centre for
Environmental Evidence for their advice on undertaking Systematic Reviews. We also thank
Roberta Shanman at the RAND Library for her help in undertaking the study search. Thanks are
also due to Ms Stephanie Diepeveen for the advice she provided in the early stage of the protocol
development; Victoria Wakefield-Jarrett who has helped us in collecting hundreds of articles
from online library; and Professor Joanna Chataway for her internal quality assurance of the draft
report before it was submitted for peer review. We would also like to thank the following
individuals with whom we conducted key informant interviews: Andrew Barnett (The Policy
Practice), Morgan Bazilian (UNIDO), Mike Bess, Prof Abeeku Brew-Hammond (KNUST,
Ghana) Simon Collings (GVEP International), Steven Hunt (DFID), Matthew Lockwood (IDS)
and Annabel Yadoo (University of Cambridge and Renewable World).
All views expressed in this document are those of the authors and do not necessarily represent the
views of the Department for International Development (DFID), SPRU, or the RAND
9. References
Acker RH, Kammen DM: The quiet (energy) revolution: analysing the dissemination of
photovoltaic power systems in Kenya. Energy Policy 1996, 24:81-111.
Adkins E, Eapen S, Kaluwile F, Nair G, Modi V: Off-grid energy services for the poor:
Introducing LED lighting in the Millennium Villages Project in Malawi. Energy Policy 2010,
African Development Bank Group: Tanzania Rural Electrification Project: Project Performance
Evaluation Report (PPER). African Development Bank Group; 1996.
Barnett A: Energy and the Reduction of Poverty: A Narrative by Andrew Barnett, The Policy
Practice Limited, Brighton, January 2010
Bazilian M, Sagar A, Detchon R, Yumkella, K: More heat and light. Energy Policy, 2010, 38: 54095412.
Brew-Hammond A, Energy access in Africa: Challenges ahead. Energy Policy 2010, 38: 22912301.
Campbell B, Vermeulen S, Mangono J, Mabugu R: The energy transition in action: urban
domestic fuel choices in a changing Zimbabwe. Energy Policy 2003, 31:553-562.
Cherni J, Kentish J: Renewable energy policy and electricity market reforms in China. Energy
Policy, 2007 35: 3616-3629.
Clough L: Marketing Challenges and Strategies for Micro and Small Energy Enterprises in East
Africa. GVEP International; 2011.
Collings S: Phone Charging Micro-businesses in Tanzania and Uganda. GVEP International;
D'Agostino AL, Sovacool BK, Bambawale MJ: And then what happened? A retrospective
appraisal of China's Renewable Energy Development Project (REDP). Renewable Energy, 2011.
36(11): 3154-3165.
DFID, AusAID and 3ie: Systematic Reviews in International Development – Call for Proposals.
UK Department for International Development, Australian Agency for International Development and
DFID: Energy for the Poor: Underpinning the Millennium Development Goals. UK Government
Department for International Development, London, UK; 2002.
Diniz ASAC, Machado Neto LVB, Camara CF, Morais P, Cabral CVT, Oliveira Filho D, Ravinetti
RF, Franc ED, Cassini DA, Souza MEM, Santos, JH, Amorim M, Review of the photovoltaic
energy program in the state of Minas Gerais, Brazil. Renewable & Sustainable Energy Reviews,
2011. 15(6): 2696-2706.
Dube I: Impact of energy subsidies on energy consumption and supply in Zimbabwe. Do the
urban poor really benefit? Energy Policy 2003, 31:1635-1645.
Duke RD, Jacobson A, Kammen DM: Photovoltaic module quality in the Kenyan solar home
systems market. Energy Policy 2002, 30:477-499.
Foster V, Steinbuks J: Paying the Price for Unreliable Power Supplies: In-House Generation of
Electricity by Firms in Africa. Policy Research Working Paper 4913. World Bank: African
Sustainable Development Front Office; 2009.[Online]. Available at: http://wwwwds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2009/04/30/000158349_20090430
Gaunt CT: Meeting electrification's social objectives in South Africa, and implications for
developing countries. Energy Policy, 2005. 33(10): 1309-1317.
Gough D: Weight of evidence: a framework for the appraisal of the quality and relevance of
evidence, in Furlong J, and Oancea A, (eds.) Applied and Practice-based Research, Special Edition of
Research Papers in Education, 2007 22(2): 213-228.
Green JM, Erskine SH: Solar (photovoltaic) systems, energy use and business activities in
Maphephethe, KwaZulu‐‐Natal. Development Southern Africa 1999, 16:221-237.
Green JM, Wilson M, Cawood W: Maphephethe rural electrification (photovoltaic) programme:
the constraints on the adoption of solar home systems. Development Southern Africa 2001, 18:1930.
Guimarães L, Caracaleanu C, Sy B, N'Dongo A, Sankaré O: Energy affordability in the Sahelian
region. Applied energy 2003, 76:9-13.
Gumbo R, Katsvairo L, Asai K: State of photovoltaic home systems in Zimbabwe. IEEE; 2003:
2640-2643 Vol. 2643
Gustavsson M, Ellegård A: The impact of solar home systems on rural livelihoods. Experiences
from the Nyimba Energy Service Company in Zambia. Renewable energy 2004, 29:1059-1072.
Haanyika CM: Rural electrification in Zambia: A policy and institutional analysis. Energy Policy
2008, 36:1044-1058.
Hofmeyr IM: Towards adequate and equitable energy provision for farmworker families.
Development Southern Africa 1995, 12:273-279.
Ilskog E, Kjellström B, Gullberg M, Katyega M, Chambala W: Electrification co-operatives bring
new light to rural Tanzania. Energy Policy 2005, 33:1299-1307.
Karekezi S, Kimani J: Status of power sector reform in Africa: Impact on the poor. Energy
Policy 2002, 30:923-945.
Karekezi S, Kithyoma W: Renewable energy strategies for rural Africa: is a PV-led renewable
energy strategy the right approach for providing modern energy to the rural poor of subSaharan Africa? Energy Policy 2002, 30:1071-1086.
Karekezi S, Sihag A: Energy Access theme results Synthesis/Compilation Report. Global Network
on Energy for Sustainable Development, Roskilde, Denmark; 2004.
Ketlogetswe C, Mothudi T: Solar home systems in Botswana--Opportunities and constraints.
Renewable and Sustainable Energy Reviews 2009, 13:1675-1678.
Kirubi C, Jacobson A, Kammen DM, Mills A: Community-based electric micro-grids can
contribute to rural development: evidence from Kenya. World Development 2009, 37:1208-1221.
Lee KS, Anas A, Oh G-T: Costs of Infrastructure Deficiencies for Manufacturing in Nigerian,
Indonesian and Thai Cities. Urban Studies, 1999. 36(12): 2135-49.
Lewis J, Ritchie J: Generalising from Qualitative Research, in Lewis J, and Ritchie J, (eds.)
Qualitative Research Practice: A Guide for Social Science Students and Researchers, Sage
Publications, London, Thousand Oaks and New Delhi; 2003.
Mapako M: Provision of Long-term Maintenance Support for Solar Photovoltaic Systems Lessons from a Zimbabwean NGO. Journal of Energy in Southern Africa 2005, 16:21-26.
Miles M, Huberman, A: Qualitative Data Analysis: An Expanded Sourcebook, 2nd ed., Sage
Publications, London, Thousand Oaks and New Delhi; 1994.
Modi V, McDade S, Lallement D, Saghir, J: Energy Services for the Millennium Development
Goals. Energy Sector Management Assistance Programme, United Nations Development Programme,
UN Millennium Project and World Bank, New York; 2005.
Mulder P, Tembe J: Rural electrification in an imperfect world: A case study from Mozambique.
Energy Policy 2008, 36:2785-2794.
Mulugetta Y, Nhete T, Jackson T: Photovoltaics in Zimbabwe: lessons from the GEF Solar
project. Energy Policy 2000, 28:1069-1080.
Murphy JT: Making the energy transition in rural East Africa: Is leapfrogging an alternative?
Technological Forecasting and Social Change 2001, 68:173-193.
Nygaard, I., Institutional options for rural energy access: Exploring the concept of the
multifunctional platform in West Africa. Energy Policy, 2010. 38(2): 1192-1201.
OECD-IEA: Energy Poverty: How to make modern energy access universal? Special early
excerpt of the World Energy Outlook 2010, IEA-UNDP-UNIDO, International Energy Agency, Paris;
Peng WY, Pan JH: Rural electrification in China: History and institution. China & World
Economy, 2006. 14(1): 71-84.
Pereira, M.G., M.A.V. Freitas, and N.F. da Silva, Rural electrification and energy poverty:
Empirical evidences from Brazil. Renewable & Sustainable Energy Reviews, 2010. 14(4):12291240.
Practical Action: Poor People’s Energy Outlook 2010. Rugby, Warwickshire: Practical Action; 2010
Rehman, I.H., Kar A, Raven R, Singh D, Tiwari J, Jha R, Sinha PK, Mirza A, Rural energy
transitions in developing countries: a case of the Uttam Urja initiative in India. Environmental
Science & Policy, 2010. 13(4): 303-311.
Schiffmand, L.G., Kanuck, L., Consumer behaviour, 1987. 3rd edn. Englewood
Cliffs, NJ: Prentice-Hall.
Schut M, Slingerland M, Locke A: Biofuel developments in Mozambique. Update and analysis of
policy, potential and reality. Energy Policy 2010, 38:5151-5165.
Steel K, Dynamics of Growth and Investment in the Kenyan Electric Power Sector. Power
Engineering Society General Meeting, 2007. IEEE , pp.1-5, 24-28 June 2007
Thiam DR: An energy pricing scheme for the diffusion of decentralized renewable technology
investment in developing countries. Energy Policy 2011, 39(7) 4284-4297.
UNDP: Energizing the Millennium Development Goals. United Nations Development Programme,
New York; 2005.
UNDP-WHO: The Energy Access Situation in Developing Countries: A Review Focusing on the
Least Developed Countries and Sub-Saharan Africa. United Nations Development Programme and
World Health Organization, New York; 2009.
UN-Energy: The Energy Challenge for Achieving the Millennium Development Goals. United
Nations; 2005.
Urban F: Climate-Change Mitigation Revisited: Low-Carbon Energy Transitions for China and
India. Development Policy Review 27(6): 693-715; 2009.
van der Plas RJ, Hankins M: Solar electricity in Africa:: a reality. Energy Policy 1998, 26:295-305.
Walekhwa PN, Mugisha J, Drake L: Biogas energy from family-sized digesters in Uganda:
Critical factors and policy implications. Energy Policy 2009, 37:2754-2762.
WDI: World Bank World Development Indicators January 2009, listing of countries by income
group. ESDS International, (Mimas) University of Manchester; 2009.
Winkler H, Spalding-Fecher R, Tyani L, Matibe K: Cost-benefit analysis of energy efficiency in
urban low-cost housing. Development Southern Africa 2002, 19:593-614.
Wolde-Ghiorgis W: Renewable energy for rural development in Ethiopia: the case for new
energy policies and institutional reform. Energy Policy 2002, 30:1095-1105.
The World Bank Independent Evaluation Group: The Welfare Impact of Rural Electrification: A
Reassessment of the Costs and Benefits. The Welfare Impact of Rural Electrification: A
Reassessment of the Costs and Benefits, World Bank; 2008
Appendix A: Search terms for academic literature
List 1a: Modern energy services
• “modern energy service*”
• electric* OR electrif*
• “mobile *phone charg*” OR “cell *phone charg*”
• batt* OR “batt* charg*”
• refrigerat*
• heat*
• freez* OR cool
• light* OR illuminat*
• communication OR radio OR television OR TV or “information and communication
technolog*” OR ICT
• “thermal comfort”
List 1b: Modern energy technologies (required to deliver and/or use the services)
• “modern energy technolog*”
• “Platform technolog*”
• “high technolog*”
• “decentrali$ed grid” OR “modern decentrali$ed grid” OR “mini$grid”
• “energy system”
• micro$hydro*
• pico$hydro*
• bio$fuel* OR bio$gas OR bio$ethanol OR waste
• photo$voltaic* OR PV OR “solar home system*” OR SHS
• (“traditional fuel*” OR “traditional bio$mass” OR bio$energy) AND (substitute* OR
• “solar power” OR “solar energ*” OR “solar panel” OR “solar water heater” OR SWH
• “wind power” OR “wind energ*”
• “wind turbine” OR “wind generat*”
• “renewable energ*”
• energy$efficien*
• “sustainabl* energy system”
• “gas turbine*”
• “nuclear power”
• “fuel cell*”
List 2: Barriers
• barrier
• difficult*
• challenge
• obstacle
• obstruction
• technical
• access
• poverty
• financ*
• politic*
• polic*
• econom* OR cost
law OR legal
“female headed household”
road$block OR block*
List 3: Interventions
• intervention*
• interference
• investment*
• loan* OR “concessionary loan*” OR concession* OR “soft loan*”
• “market development”
• “public$private partnership*” OR PPP
• subsid*
• micro$credit OR micro$finance
• “capacity building”
• “development programme” OR “development program”
• adopti*
• diffusi*
• enabl*
• facilitat*9
• help
• enhanc*
• roadmap
List 4: Effectiveness measures
• effectiv*
• useful OR use
• success*
• cost-effectiveness
• “economic growth” OR “productive use”
• Health OR “eye disease” OR “eye infection” OR “respiratory disease”
• uptake
• roll-out
We note the terms “enabl*”,“facilitat*”, “help”, “enhance*” and “roadmap” under the list of interventions (List 3) can also
be interpreted as antonyms of barriers, so it would be equally valid to list them under the list of barriers (List 2). In practice,
it makes no difference which list they are on as long as they are included at least once.
scale-up OR “scaling up”
improvement AND (“rate* of access” OR “level* of access”)
List 5: Geographic search terms
• sub-Saharan Africa OR Africa OR SSA
• “developing countr*”
• “southern countr*”
• “global south”
• “low income countr*”
• “least industriali$ed countr*” OR LDC
• “poor countr*”
• “developing region”
• “developing econom*”
• “underdeveloped countr*”
• “third world”
• Benin
• “Burkina Faso”
• Burundi
• “Central African Republic”
• Chad
• Comoros
• “Democratic Republic of the Congo” OR DRC
• Eritrea
• Ethiopia
• “The Gambia”
• Ghana
• Guinea
• Guinea-Bissau
• Kenya
• Liberia
• Madagascar
• Malawi
• Mali
• Mauritania
• Mozambique
• Niger
• Rwanda
• Senegal
• “Sierra Leone”
• Somalia
• Tanzania
• Togo
• Uganda
• Zambia
• Zimbabwe
• Afghanistan
• Bangladesh
• Cambodia
“Democratic Republic of Korea” OR “North Korea”
“Kyrgyz Republic”
“Lao PDR” OR “Lao People’s Democratic Republic” OR Lao
Appendix B: Grey literature search terms
A * in the lists below indicates that no specific search terms were used, but rather all titles under
energy-related categories which were already built into the database were collected. This strategy
was used to minimise the risk that our search terms did not unnecessarily limit the search as one
is not always able to determine the robustness of different databases and associated search
World Bank
OECD Papers
Search terms: *
(Search terms applied to full text)
Theme: Energy in PAPERS
Language(s): English
OECD Stats
Search terms: energy OR fuel OR fuels OR electricity OR electric OR power resources
(Search terms applied to titles only)
Document type: Publications and Research
Date: 1995/01/01 to 2011/11/01
Topics: Energy
Region: Africa, East Asia and Pacific
Language(s): English
Search terms: *
(Search terms applied to full text)
Theme: Energy in STATISTICS
Language(s): English
List of Systematic Reviews: Agriculture & Rural Development
Global Village Energy Partnership (GVEP)
Studies and reports: all
o Books and book chapters;
o Pre - Print Articles Published in Energy Policy Special Issue of 2002 on Africa
Published by Elsevier Science Limited, United Kingdom
o Pre- Print Articles Published in Other Leading Energy Journals
o Pre -Print Articles Published in Conference Proceedings
o Presentations during Conference Proceedings
o Occasional Papers
Renewable energy and energy efficiency partnership (REEEP)
Projects: completed projects
World Health Organisation (WHO) website
Search terms: energy OR fuel OR fuels OR green OR power OR electricity
(Search terms applied to titles only)
World Health Organisation (WHO) Library
Search terms: energy OR fuel OR fuels OR green OR power OR electricity
(Search terms applied to titles only)
United Nations
Subjects: energy
African Development Bank
Project/Programme Completion Reports
o Countries: all countries
o Topic and sector: energy and power
International Energy Agency
Publications and Papers: Non-OECD Countries
Appendix C: Detailed quality review criteria list
· Findings/conclusions are supported
by data and study evidence
Detailed qualitative
· Findings/conclusions have a
coherent logic
Credibility of findings
· There is evidence of
validation/triangulation of findings to
support or refine findings
[Yes/No/Not applicable]
· Quotations are numbered or
otherwise identified to demonstrate
they are not from one/two people
Breadth and depth of study
findings (scope for wider
· Study aims and design is set in the
context of existing
knowledge/understanding (e.g.
literature review to summarise
knowledge to date)
[Yes/No/Not applicable]
· Findings are
presented/conceptualised in a way that
offers new insights/alternative ways of
· Presents the potential for wider
inferences from the study:
o Describes how findings
are relevant to the wider population
from which the sample is drawn
o Describes the contexts in
which the study was conducted to
allow applicability to other settings to
be assessed
o Discusses the limitations
on drawing wider inferences
Extent to which the study
addresses original aims and
· Clear statement of study aims and
objectives, and clearly state reasons
for any changes in objectives
[Yes/No/Not applicable]
· Findings are clearly linked to the
purposes of the study
· Discusses limitations of study in
meeting aims (e.g. gaps in coverage,
restricted access to setting or
participants, unresolved areas of
questioning, time constraints, etc)
design · Discusses how overall research
[Yes/No/Not applicable]
strategy was designed to meet the
aims of the study
· Discusses the rationale for study
design; including convincing
arguments for different features of
research design (e.g. multiple
methods, time frames, reasons for
components of research, etc)
· Use of different features of
design/data sources evident in findings
· Discusses limitations of research
design and their implications for the
study evidence
· Describes any changes made to
design, their justification, and
implications for the study
Defensibility of sample design
and coverage
· Target sample is appropriate to
accessing the type of knowledge
[Yes/No/Not applicable]
· Discusses how sample/selections
allowed required comparisons to be
· Sample profile is detailed
o Describes location/areas
and how/why chosen
o Describes population of
interest and sample’s relationship to it
o Discusses missing
coverage and implications for
o Documents reasons for
non-participation/exclusion among the
o Discusses access and
methods of approach and how these
might have affected participation and
o Discusses why some
people may have chosen not to
· Data collection tools were piloted
· Data collection was comprehensive,
flexible and sensitive enough to
provide a complete and/or rich
o E.g. time with participants,
1+ methods of data collection,
following up?
o Discusses how fieldwork
methods or settings may have
influenced data collected
o Discusses the saturation of
· Clear discussion of data collection
o Who conducted data
o Procedures and documents
used for collection and recording;
conventions for field notes
o Checks on origins, status
and authorship
o Audio or video recording
of interviews, discussions and/or
conversations (or justification for why
· Steps taken to ensure participants
were able and willing to contribute
· Discussion of strengths and
weaknesses of data sources and
· Data analysis methods were
o Describes the form of
original data
o Clear rationale for choice
of management methods, tools and
o Evidence of how analytic
categories, classes, labels, etc have
been generated and used
o Discussion, with
examples, of how any constructed
analytic concepts/typologies have
been devised and applied
Quality of the approach and
formulation of the analysis
[Yes/No/Not applicable]
· Balanced analysis in the extent to
which it is guided by preconceptions
or by the data
o Diversity of perspectives
o Discussion of bias in
forming the research question (e.g.
funding agencies)
o Examines the
researcher’s own role in influencing
data collection
· Detailed and clear portrayal of data
sources and the context for data
sources (e.g. background, personal
context, origins)
· Clear conceptual links between
analytic commentary and presentation
of original data
o Explores alternative
o Displays negative cases and
how they lie outside main proposition;
or how propositions have been revised
to include them
· Demonstrates link to aims of the
study/research questions
Clarity and coherence of
[Yes/No/Not applicable]
· Provides a narrative or clearly
constructed thematic account; and has
a useful structure and signposting to
guide the reader
· Provides accessible information for
intended target audience(s)
· Key messages are highlighted or
· Discussion/evidence of the main
assumptions, hypotheses and
theoretical ideas on which the study
was based and how these affected the
form, coverage or output of the study
Clarity of assumptions, that
have shaped the form and output
of the study
[Yes/No/Not applicable]
· Evidence of openness to new and
alternative ways of viewing the
· Discussion of how error or bias may
have arisen in design, data collection
and analysis and how it was
addressed, if at all
· Reflections on the impact of the
researcher on the research process
Further comments on quality
Elaborate as necessary
Overall rating of quality
Appendix D: List of publications by authors with potential conflicts of
Byrne, R. (2011) Learning Drivers: rural electrification regime building in Kenya and Tanzania,
DPhil thesis, SPRU, University of Sussex, Brighton.
Byrne, R. (2011) “Micro-energy systems in low-income countries: learning to articulate the solar
home system niche in Tanzania”, in M. Schäfer, N. Kebir and D. Philipp (eds.) Proceedings of
the International Conference: Micro Perspectives for Decentralized Energy Supply, Technische
Universität Berlin, 7th to 8th April 2011:207-219.
Byrne, R., A. Smith, J. Watson and D. Ockwell (2011) “Energy pathways in low-carbon
development: from technology transfer to socio-technical transformation”, STEPS Working
Paper 46, STEPS Centre, Brighton.
Byrne, R., A. Smith, J. Watson and D. Ockwell (2012) “Energy Pathways in Low-Carbon
Development: The Need to Go beyond Technology Transfer”, in D. Ockwell and A. Mallett (eds.)
Low Carbon Technology Transfer: From Rhetoric to Reality, Routledge, London and New York.
Mallett, A., D. Ockwell, J. Watson et al. (2009) UK-India collaborative study on the transfer of
low carbon technology: Phase II Final Report, Brighton and New Delhi, SPRU and TERI.
Ockwell, D., R. Haum, A. Mallett and J. Watson (2010) “Intellectual property rights and low
carbon technology transfer: Conflicting discourses of diffusion and development” Global
Environmental Change 20(4): 729-738.
Ockwell, D., A. Ely, A. Mallett, O. Johnson and J. Watson (2009) Low carbon development: The
role of local innovative capabilities, Brighton: Sussex Energy Group and ESRC STEPS Centre,
University of Sussex.
Ockwell, D., J. Watson et al (2007) UK-India Collaboration to Identify the Barriers to the
Transfer of Low Carbon Energy Technology, Final Phase 1 Report to Defra and the Indian
Ministry of Environment and Forestry.
Ockwell, D., J. Watson, A. Verbeken, A. Mallett and G. MacKerron (2009) “A blueprint for post2012 technology transfer to developing countries”, Sussex Energy Group Policy Briefing No. 5,
Ockwell, D., J. Watson, G. MacKerron, P. Pal and F. Yamin (2008) “Key policy considerations
for facilitating low carbon technology transfer to developing countries”, Energy Policy 36(11):
Wang, T. and J. Watson (2009) China's Energy Transition: Pathways for Low Carbon
Development, Brighton: Tyndall Centre and Sussex Energy Group, University of Sussex, April.
Wang, T. and J. Watson (2009) “Scenario analysis of China’s emissions pathways in the 21st
century for low carbon transition”, Energy Policy 38(10): 3537-3546.
Watson, J., Byrne, R. et al (2011) UK-China Collaborative Study on Low Carbon Technology
Transfer. Final report to the Department of Energy and Climate Change. Brighton: SPRU,
University of Sussex.
Watson, J. and R. Byrne (2012) “Low-carbon Innovation in China: The Role of International
Technology Transfer”, in D. Ockwell and A. Mallett (eds.) Low Carbon Technology Transfer:
From Rhetoric to Reality, Routledge, London and New York.
Watson, J. and R. Sauter (2011) “Sustainable innovation through leapfrogging: A review of the
evidence” International Journal of Technology and Globalisation 5: 170-189
Watson, J. and O. Johnson (2010) Renewable Energy Technologies for Rural Development,
UNCTAD current studies on science, technology and innovation, Geneva, UNCTAD.
Watson, J. and R. Sauter (2008) Technology Leapfrogging: A Review of the Evidence, Report for
the Department of International Development, Brighton: SPRU, University of Sussex.
Appendix E: Pilot search results
Web of Science pilot search
We undertook a pilot search in the Web of Science using all of the modern energy services terms
in List 1a combined with the barriers terms (List 2). Although this is only a partial search (no
technologies terms from List 1b or intervention terms from List 3), the total number of hits was
over 22,000, with many irrelevant hits. It is particularly problematic with databases that are not
specifically focused on energy, as there were terms that could have other meanings (e.g. “light”,
“illuminate, or “waste”). The review team therefore refined the search strategy to include the
terms “energy or electricity or power or fuel” in every search in order to improve the relevance of
our search.
World Bank Pilot Search
Search terms: energy OR fuel OR fuels OR electricity OR electric OR power resources
(Search terms applied to titles only)
Document type: all
Date: 1995/01/01 to 2011/11/01
Topics: Energy
Region: Africa, East Asia and Pacific
Language(s): English
This pilot search resulted over 4000 hits, many of which were press releases. Therefore, we refine
our search by limiting the keywords to titles only and restricting the document type to
“Publications and Research”. This greatly reduced the number of relevant hits. We had 205 hits in
our final search.
Appendix F: List or file of articles excluded at full text stage
Peer-reviewed academic papers (75 articles)
Ablethomas, U., et al., Dissemination of Photovoltaics in the Gambia. Renewable Energy, 1995.
6(5-6): p. 507-513.
Anas, A., et al., Why Manufacturing Firms Produce Some Electricity Internally. 1999.
Angel-Urdinola, D. and Q. Wodon, Do Utility Subsidies Reach the Poor? Framework and
Evidence for Cape Verde, Sao Tome, and Rwanda. Economics Bulletin, 2007. 9(4): p. 1-7.
Asamoah, J., The uptake of energy conscious housing design in South Africa as a mitigation of
emission of greenhouse gases, in Greenhouse Gas Control Technologies. 1999. p. 653-657.
Bawakyillenuo, S.E.A.b.h.c., Policy and Institutional Failures: Photovoltaic Solar Household
System (Pv/Shs) Dissemination in Ghana. Energy & Environment.
Berg, S.V., M.G. Pollitt, and M.e. Tsuji, Private initiatives in infrastructure: Priorities, incentives
and performance. 2002. 234: p. 234.
Brew-Hammond, A., Energy Access in Africa: Challenges Ahead. Energy Policy, 2010. 38(5): p.
Brew-Hammond, A.C.-R.A., Reducing rural poverty through increased access to energy services :
a review of the multifunctional platform project in Mali. 2004, Bamako, Mali: United Nations
Development Programme, UNDP Mali Office. 80 p.
Cawood, W., An affordable formula for financing solar electrification. Siemens Review, 1996.
63(3-4): p. 23-24.
Coudouel, A., A.A. Dani, and S.e. Paternostro, Poverty and Social Impact Analysis of Reforms:
Lessons and Examples from Implementation. 2006. 520: p. 520.
Covarrubias, A.J., Lending for electric power in sub-Saharan Africa. 1996. 93: p. 93.
Davidson, O.R. and J.E.A.j.t.r.d. Turkson, Overcoming financial barriers to wider use of
renewable energy technology in Africa. International Journal of Global Energy Issues.
Davis, M., Rural household energy consumption - The effects of access to electricity - evidence
from South Africa. Energy Policy, 1998. 26(3): p. 207-217.
Deichmann, U., et al., The economics of renewable energy expansion in rural Sub-Saharan Africa.
Diarra, D.C. and F.O. Akuffo, Solar photovoltaic in Mali: potential and constraints. Energy
Conversion and Management, 2002. 43(2): p. 151-163.
Eberhard, A.E.A.e.g.u.a.z. and K.N. Gratwick, IPPs in Sub-Saharan Africa: Determinants of
success. Energy Policy.
Egels, N., CSR in Electrification of Rural Africa. Journal of Corporate Citizenship, (18): p. 75-85.
Ellegard, A., et al., Rural people pay for solar: experiences from the Zambia PV-ESCO project.
Renewable Energy, 2004. 29(8): p. 1251-1263.
Gillwald, A., The Poverty of ICT Policy, Research, and Practice in Africa. Information
Technologies & International Development: p. 79-88.
Gudmundsdottir, G.B., When does ICT support education in South Africa? The importance of
teachers' capabilities and the relevance of language. Information Technology for Development.
16(3): p. 174-190.
Guèye, C., New Information & Communication Technology Use by Muslim Mourides in Senegal.
Review of African Political Economy. 30(98): p. 609-625.
Gullberg, M., et al., Village electrification technologies - an evaluation of photovoltaic cells and
compact fluorescent lamps and their applicability in rural villages based on a Tanzanian case
study. Energy Policy, 2005. 33(10): p. 1287-1298.
Habtetsion, S. and Z. Tsighe, The energy sector in Eritrea-institutional and policy options for
improving rural energy services. Energy Policy, 2002. 30(11-12): p. 1107-1118.
Hafkin, N.J. and S. Huyer, Women and Gender in ICT Statistics and Indicators for Development.
Information Technologies & International Development. 4(2): p. 25-41.
Harries, M., Disseminating wind pumps in rural Kenya-meeting rural water needs using locally
manufactured wind pumps. Energy Policy, 2002. 30(11-12): p. 1087-1094.
Hodge, J., Liberalising communication services in South Africa. Development Southern Africa.
17(3): p. 373-387.
Hosman, L., Policies, Partnerships, and Pragmatism: Lessons from an ICT-in-Education Project
in Rural Uganda. Information Technologies & International Development. 6(1): p. 48-64.
Jumbe, C.B.L., F.B.M. Msiska, and M. Madjera, Biofuels development in Sub-Saharan Africa:
Are the policies conducive? Energy Policy, 2009. 37(11): p. 4980-4986.
Karekezi, S., Poverty and energy in Africa - A brief review. Energy Policy, 2002. 30(11-12): p.
Kassenga, G.R., Promotion of renewable energy technologies in Tanzania. Resources
Conservation and Recycling, 1997. 19(4): p. 257-263.
Kassenga, G.R., The Status and Constraints of Solar Photovoltaic Energy Development in
Tanzania. Energy Sources Part B-Economics Planning and Policy, 2008. 3(4): p. 420-432.
Kebede, B., A. Bekele, and E. Kedir, Can the urban poor afford modern energy? The case of
Ethiopia. Energy Policy, 2002. 30(11-12): p. 1029-1045.
Kenny, A.E.A.a.a.c.z., Prospects for nuclear power in Sub-Saharan Africa in the 21st century.
International Journal of Global Energy Issues.
Khennas, S., Urban waste management for small scale energy production. 2003, [Sri Lanka: s.n.].
81 p.
Kivunike, F.N., et al., Perceptions of the role of ICT on quality of life in rural communities in
Uganda. Information Technology for Development. 17(1): p. 61-80.
Knoef, H.A.M., The UNDP/World Bank monitoring program on small scale biomass gasifiers
(BTG's experience on tar measurements). Biomass & Bioenergy, 2000. 18(1): p. 39-54.
Le Roux, D., The proposed construction of the nuclear pebble-bed reactor at Koeberg - an
assessment of nuclear energy as a sustainable energy resource in relation to other alternative
natural resources. 2005. p. 113 leaves ; 30 cm. Dissertation: Thesis (LL.M. (Faculty of Law))-University of the Western Cape, 2005.
Lenehan, A.M., Photovoltaic water pumping and its potential for application in community water
supply in South Africa. Water Sa, 1996. 22(3): p. 257-262.
Mangwengwende, S.E., Tariffs and subsidies in Zimbabwe's reforming electricity industry:
steering a utility through turbulent times. Energy Policy, 2002. 30(11-12): p. 947-958.
Matthee, K.W., et al., Bringing Internet connectivity to rural Zambia using a collaborative
approach, in 2007 International Conference on Information and Communication Technologies
and Development. 2007. p. 46-57.
Meso, P., P. Musa, and V. Mbarika, Towards a model of consumer use of mobile information and
communication technology in LDCs: the case of sub-Saharan Africa. Information Systems
Journal. 15(2): p. 119-146.
Minges, M., Mobile cellular communications
Telecommunications Policy. 23(7/8): p. 585.
Mulugetta, Y., Human capacity and institutional development towards a sustainable energy future
in Ethiopia. Renewable & Sustainable Energy Reviews, 2008. 12(5): p. 1435-1450.
Nanka-Bruce, O., The Socioeconomic Drivers of Rural Electrification in Sub-Saharan Africa.
Journal of Energy and Development, 2008. 33(2): p. 215-47.
Odendaal, N., Splintering Urbanism or Split Agendas? Examining the Spatial Distribution of
Technology Access in Relation to ICT Policy in Durban, South Africa. Urban Studies (Sage
Publications, Ltd.). 48(11): p. 2375-2397.
Olukoju, A., 'Never expect power always': Electricity consumers' response to monopoly,
corruption and inefficient services in Nigeria. African Affairs, 2004. 103(410): p. 51-71.
Organisation for Economic, C.-o., Development, and B. World, Liberalisation and Universal
Access to Basic Services: Telecommunications, Water and Sanitation, Financial Services, and
Electricity. 2006. 272: p. 272.
Otiti, T.E.A.t.p.m.a.u. and W.O. Soboyejo, Limited Contribution of Photovoltaic Energy
Technology to Economic Development of Sub-Saharan Africa. Perspectives on Global
Development & Technology.
Parawira, W.E.A.a.y.c.u., Biogas technology in sub-Saharan Africa: status, prospects and
constraints. Reviews in Environmental Science & Biotechnology.
Peters, J., C. Vance, and M. Harsdorff, Grid Extension in Rural Benin: Micro-manufacturers and
the Electrification Trap. World Development, 2011. 39(5): p. 773-83.
Pollitt, M.G., Ownership and performance in electric utilities: The international evidence on
privatization and efficiency. 1995. 240: p. 240.
Ramana, P.V.e., Rural and renewable energy: Perspectives from developing countries. 1997. 317:
p. 317.
Rhodes, J., A Strategic Framework for Rural Micro-Enterprise Development: The Integration of
Information Communication Technology (ICT), E-Commerce, Marketing, and Actor-Network
Theory. Perspectives on Global Development & Technology. 8(1): p. 48-69.
Sebitosi, A.B. and P. Pillay, Energisation of rural Sub-Saharan Africa: Grappling with poor
sustainability, in 2006 Power Engineering Society General Meeting, Vols 1-9. 2006. p. 31303134.
Sebitosi, A.B. and R. Okou, Re-thinking the Power Transmission Model for Sub-Saharan Africa.
Energy Policy, 2010. 38(3): p. 1448-54.
Sheya, M.S. and S.J.S. Mushi, The state of renewable energy harnessing in Tanzania. Applied
Energy, 2000. 65(1-4): p. 257-271.
Sosovele, H., Policy Challenges Related to Biofuel Development in Tanzania. Africa Spectrum,
2010. 45(1): p. 117-129.
Spalding-Fecher, R., et al., The economics of energy efficiency for the poor - a South African
case study. Energy, 2002. 27(12): p. 1099-1117.
Stage, J. and F. Fleermuys, Energy Use in the Namibian Economy from 1995 to 1998.
Development Southern Africa, 2001. 18(4): p. 423-41.
Stassen, H.E., Small-scale biomass gasifiers for heat and power: A global review. 1995. 61: p. 61.
Tall, S.M., Senegalese &Eacute:migr&eacute:s: New Information
Technologies. Review of African Political Economy. 31(99): p. 31-48.
Tatietse, T.T., et al., A new and rational approach to modelling and designing potable water and
electricity supply systems in African cities. Building and Environment, 2000. 35(7): p. 645-654.
Teferra, M., Power sector reforms in Ethiopia: options for promoting local investments in rural
electrification. Energy Policy, 2002. 30(11-12): p. 967-975.
Tewari, D.D. and T. Shah, An assessment of South African prepaid electricity experiment,
lessons learned, and their policy implications for developing countries. Energy Policy, 2003.
31(9): p. 911-927.
Turkson, J. and N. Wohlgemuth, Power sector reform and distributed generation in sub-Saharan
Africa. Energy Policy, 2001. 29(2): p. 135-145.
Turkson, J., Attracting capital for electric power development in sub-Saharan Africa: A
multipronged approach, in Energy and Economic Growth: Is Sustainable Growth Possible?, Vols
1-3. 1997. p. 583-591.
Turyareeba, P.J., Renewable energy: its contribution to improved standards of living and
modernisation of agriculture in Uganda. Renewable Energy, 2001. 24(3-4): p. 453-457.
Uprety, K., Transboundary Energy Security: Emerging Legal and Institutional Framework for
Electricity Trading in Southern Africa. Journal of Energy & Natural Resources Law.
van Eijck, J. and H. Romijn, Prospects for Jatropha Biofuels in Tanzania: An Analysis with
Strategic Niche Management. Sectoral Systems of Innovation and Production in Developing
Countries: Actors, Structure and Evolution, 2009: p. 335-66.
van Horen, C. and G. Simmonds, Energy efficiency and social equity in South Africa: seeking
convergence. Energy Policy, 1998. 26(11): p. 893-903.
vandenBroek, R. and L. Lemmens, Rural electrification in Tanzania - Constructive use of project
appraisal. Energy Policy, 1997. 25(1): p. 43-54.
Viebahn, P.E.A.p.v.w.o., Y.E.A.y.l.c.e. Lechon, and F.E.A.f.t.d.d. Trieb, The potential role of
concentrated solar power (CSP) in Africa and Europe-A dynamic assessment of technology
development, cost development and life cycle inventories until 2050. Energy Policy.
Weggelaar, S., New strategies to enhance the thermal and energy efficiency of housing in South
Africa. 2002. p. xvii, 249 leaves.
Youm, I., et al., Renewable Energy Activities in Senegal: A Review. Renewable & Sustainable
Energy Reviews, 2000. 4(1): p. 75-89.
Grey literature (9 articles)
Kalumiana, O. S. Rural Energy Access: Promoting Solar Home Systems in Rural Areas in
Zambia - A Case Study. AFREPREN/FWD.
Mbaiwa, J. E. An Assessment of Policies and Programme designed for the Production and
Distribution of Modern Renewable Energy to Rural Areas in Botswana. University of Botswana,
Harry Oppenheimer Okavango Research Centre.
GVEP International 2010. Kenya Briquette Industry Study. London: GVEP International.
World Bank 2010. IBRD Partial Credit Guarantee (PCG) to advance Botswana Power Sector
Development. World Bank.
Nguema-Ollo, J. B. 2006. Republic of Benin, Project for the Electrification of 17 Rural Centres:
Project Completion Report. African Development Fund.
Fikru, B., Assefaw, M., Amu, O. & Shannon, E. 2005. Mozambique, Electricity II Project:
Project Completion Report. African Development Fund.
African Development Bank 2011. Summary of the Environmental and Social Management Plan:
Electricity Transmission and Distribution Network Development Project. African Development
African Development Bank Group 2011. Energy Sector Policy of the African Development Bank
Group (Draft). African Development Bank Group.
United Nations 2008. Innovation for Sustainable Development: Local Case Studies from Africa.
United Nations.
Appendix G: Counts of grey literature on barriers
All Africa
Burkina Faso
Cape Verde
DR Congo
Eastern Africa
Middle East and North Africa Region
Sierra Leone
South Africa
Southern Africa
Sub-Saharan Africa in general
Type of energy service or technology analysed
Electricity or electrification
Battery or battery charging
Heat or thermal comfort
Lighting or illumination
Mobile phone charging
Biofuel, biogas, bioethanol or waste
Solar or photovoltaic
Wind power
Gas turbine
Nuclear power
Fuel cell
Improving energy efficiency
Grey literature
Grey literature
Grey literature
Sector reform
Market-based intervention
Capacity (Skill)
Grey literature
Grey literature
Appendix H: Counts of grey literature on interventions
Cape Verde
Sub-Saharan Africa in general
Type of energy service or technology analysed
Electricity or electrification
Battery or battery charging
Heat or thermal comfort
Lighting or illumination
Mobile phone charging
Biofuel, biogas, bioethanol or waste
Solar or photovoltaic
Wind power
Gas turbine
Nuclear power
Fuel cell
Improving energy efficiency
Sector reform
Market-based intervention
Grey literature
Grey literature
Grey literature
Grey literature
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF