Energy Policy 29 (2001) 689}699

Renewable energy investment by the World Bank

Eric Martinot*
Stockholm Environment Institute~Boston, 11 Arlington St., Boston, MA 02116, USA
Received 20 November 2000

Abstract

World Bank Group lending for renewable energy accelerated in the 1990s and resulted in 17 approved projects with $700 million in
Bank loans and $230 million in grants by the Global Environment Facility. The Bank's 1999 energy-sector strategy Fuel for Thought
charted new directions for renewable energy investment. The present paper considers the implementation challenges of Fuel for
Thought strategies and the opportunities for carrying them out. The paper distinguishes between agendas in the energy and
rural-development sectors, and reviews limitations to implementing these agendas. Lessons from projects are just emerging, but
suggest "ve areas of support for renewable energy by the Bank in the future: renewable energy "nancing, electric power policy
frameworks, rural energy enterprises, regulated rural energy concessions, and domestic technology manufacturing. Interviews with
the private sector suggest additional forms of support: assist with business plans, "nance pre-feasibility studies, reduce commercial
risks, support joint ventures, build market volume and stability, and pilot and test innovative business models. The e!ectiveness of the
Bank in following through on its ambitious agenda fundamentally rests on the willingness and commitment of developing countries
to pursue these strategies and the degree to which renewable energy applications are seen to serve countries' development
priorities.  2001 Elsevier Science Ltd. All rights reserved.

Keywords: Electricity; Rural energy development; Multilateral development banks

1. Introduction Three milestones stand out in the evolution of renew-

able energy within the Bank during the 1990s. In 1992,
World Bank Group lending for renewable energy pro- the Bank established the Asia Alternative Energy Pro-
jects in developing countries accelerated during the 1990s gram (ASTAE) to bring renewable energy and energy
as a con#uence of development, environment and social e$ciency into the Bank's power sector lending in Asia.
factors began to convince the Bank and its client coun- Funded primarily from outside the Bank by bilateral
tries that renewable energy projects were viable invest- donors and other partners, ASTAE greatly facilitated
ments. By 1999, assisted by grants from the Global preparation and implementation of a broad portfolio of
Environment Facility (GEF), the Bank had approved 17 renewable energy projects and activities throughout
renewable energy projects with a total cost of $3.2 billion, Asia. ASTAE experts also assisted with renewable energy
which includes Bank loans and credits of $700 million projects in other operating regions of the Bank and were
and GEF co-"nancing of $230 million. An additional 13 able to boost renewable energy activity by enabling Bank
projects had been awarded about $210 million in grants task managers to proceed with greater con"dence in
by the GEF but not yet approved by the Bank (see the developing the new project approaches demanded by
appendix for a description of the Bank's renewable renewable energy and in working with new local country
energy project portfolio). counterparts.
In 1996, the Bank published Rural Energy and Develop-
* The research underlying this paper was conducted when the author ment: Improving Energy Supplies for Two Billion People
was an Associate of the Stockholm Environment Institute*Boston. He (World Bank, 1996a). This small book emphasized the
now works for the Global Environment Facility. connection between energy services and rural poverty
E-mail address: [email protected] (E. Martinot). alleviation. It highlighted the fact that two billion people
 The term `Banka refers to the World Bank Group throughout the
paper. The term `renewable energya refers to solar, wind, biomass,
geothermal and mini-hydro technology applications; large-scale hy-
dropower is excluded.  For more information see www.worldbank.org/astae.

0301-4215/01/$ - see front matter  2001 Elsevier Science Ltd. All rights reserved.
PII: S 0 3 0 1 - 4 2 1 5 ( 0 0 ) 0 0 1 5 1 - 8
690 E. Martinot / Energy Policy 29 (2001) 689}699

still lack access to modern energy services like electricity. tions. This paper is based partly on a series of interviews
This book, and the work that preceded it, helped to conducted during 1998 and 1999 while the author
launch preparation of more than 10 Bank projects worked as a consultant for the Environment Department
for providing energy services to rural populations of the World Bank. Nevertheless, the views expressed are
lacking access to electricity, through the use of solar strictly those of the author and do not necessarily re#ect
photovoltaics and other renewables (see appendix for the views of the World Bank Group or any of its client
a review of the early studies). This book outlined an governments.
8-point action plan for broadening energy access in rural
areas, including a greater emphasis on rural energy
throughout the Bank and a new rural energy initiative for 2. Agendas for renewable energy investment by the Bank
Africa.
In 1999, the World Bank board of directors adopted Renewable energy and environmental advocates have
a sector-strategy paper entitled Fuel for Thought: Envir- often encouraged the Bank to lend more for renewables
onmental Strategy for the Energy Sector. This paper (Kozlo! and Shobowale, 1994; Flavin and Tunali, 1996;
promised to promote energy sector reform that makes Flavin, 1997; Institute for Policy Studies, 1998; PCAST,
renewables more competitive with conventional energy 1999). Typical was the US President's Committee on
sources (often called `leveling the playing "elda). It also Science and Technology (PCAST) 1999 report, which
promised to promote renewable energy projects `as called upon multilateral development banks to begin
mainstream activities where they are cost-e!ective solu- a transition `from conventional technologies to support-
tions to energy and environmental prioritiesa and to ing clean technologiesa (p. 3}36). While this report did
`expand support for the identi"cation and preparation not call for speci"c targets, others have advocated targets
of renewable energy2projectsa (World Bank, 1999a, like 15% of total Bank energy-sector lending devoted to
p. 35). renewable energy. These `supply-sidea views see the
Fuel for Thought also proposed a new `strategic part- Bank in a position to supply greater aggregate amounts
nershipa with the GEF for renewable energy. The pro- of investment capital and other assistance for renewable
posed strategies under this partnership recognized the energy projects in the energy sector.
experimental, iterative and time-intensive nature of e!ec- The `supply-sidea agenda of renewable energy and
tive models and strategies, along with the need for rapid environmental advocates is often shared by technology
response to private sector proposals. Four key features of manufacturers, R&D planners and bilateral assistance
the proposed program were: (a) adaptable program loans agencies pushing technology exports. These players often
to provide long-term policy and investment programs think of renewables by speci"c technology type * wind,
(up to 10 years) in a particular country with #exible solar, biomass, small hydro. Together with environ-
tranches and adjustable designs; (b) policy tools that facil- mental advocates, their agendas may include combating
itate supportive regulatory frameworks for grid-connec- global climate change, expanding market opportunities
ted renewable energy in developing countries; (c) a rapid for technology suppliers in developing countries, and
response envelope to allow quick investment decisions on stimulating technology advances and cost reductions en-
private-sector opportunities meeting pre-agreed criteria; gendered by greater technology deployment.
and (d) country-based intermediaries to identify and ap- A common response to the `supply-sidea view is that
praise projects. As an interim target, the Bank proposed client countries must be willing to borrow for renewables
$150 million annually in GEF resources combined with before the Bank can lend * the `demand sidea of the
$600 million per year from the Bank and other sources. equation. This means countries must understand the
This commitment was far larger than historic lending costs and bene"ts of renewable energy technologies and
patterns. their potential to solve the most pressing development
Although the emerging renewable energy project port- problems before they are willing to borrow, say Bank
folio and the plans in the 1996 paper and in Fuel for managers. `Borrower commitmenta is constrained by
Thought are laudable, the prospects for renewable energy lack of familiarity with renewable energy technologies,
investments by the Bank are still uncertain. What chal- lack of understanding of the costs and bene"ts and
lenges does the Bank face in continuing to "nance renew-
able energy investments in its client countries? What are
the strategic opportunities for carrying out an ambitious  During the 1990s, renewable energy markets in developed countries
renewable energy program, particularly in partnership accelerated rapidly as renewables became competitive with conven-
with the private sector? Interviews conducted by the tional energy supplies in some applications and as favorable policies
author with Bank task managers, private-sector renew- towards renewables were enacted * see for example Grubb (1995),
Shepherd (1998), and IEA (1997, 1999). Developing countries' experi-
able energy "rms, and energy/environment researchers ence has been more moderate but quite varied in the range of applica-
and advocates, combined with emerging project experi- tions and experiences * see for example Goldemberg and Johansson
ences from Bank projects, help to address these ques- (1995), Ramana (1997), and ESMAP (2000a, b).
 E. Martinot / Energy Policy 29 (2001) 689}699 691

international experience, perceptions of increased risk, electri"cation in developing countries over the past few
and entrenched political interests which bias policy- decades, providing grid-extension to the 2 billion people
makers towards conventional energy supply. Borrower currently without access to electricity is going to be very
commitment also can be a!ected by other factors: costly and will take time; (iii) solar home systems repres-
ent least-cost ways of providing rural electri"cation in
E energy prices may be too low for renewable energy to
the short or medium term while also satisfying environ-
compete on economic grounds (perhaps because of
mental concerns; and (iv) solar home systems are `com-
explicit or implicit subsidies for conventional energy);
merciala because their lifecycle costs are comparable to
E Ministries of Finance (the o$cial counterpart for Bank
the costs of avoided kerosene, candles, and batteries, and
lending) may be conservative, may not understand
so can be `delivereda by the private sector.
renewable energy, may be reluctant to provide guaran-
But despite the fact that rural home lighting meets an
tees, and may not trust new "nancing arrangements
important development need, `rural electri"cationa is
like credit lines;
still fundamentally an energy-sector agenda. The ques-
E countries may be reluctant to borrow for renewables
tion `how to most cheaply provide electricity to rural
when they are able to receive bilateral grant money for
households?a is not the question rural development pro-
the same thing;
fessionals ask "rst. Instead, they ask how renewables
E the `shadowa cast on renewable energy by failures of
o!er advantages to people in developing countries in
rural electri"cation projects in the 1970s and 1980s
meeting basic everyday needs. A `developmenta view
may dampen enthusiasm for the current generation of
focuses on the productive applications of renewable
projects;
energy (i.e., water pumping in agriculture, lighting in
E private-sector project developers may be unable to
residential housing and public schools, vaccine refriger-
absorb "nancing and conduct projects if the private
ation, cottage industries, and telecommunications servi-
sector is not strong enough;
ces), particularly those that result in new income
E urban utilities may be in poor "nancial condition and
generation or better ways to provide social services. And
unable to a!ord work in rural areas; and
funds spent on energy services from renewable energy
E renewables may be perceived to have an insu$cient
must be weighed against competing priorities for rural
technological track record.
development "nance * such as clean water, sanitation,
Nevertheless, renewable energy investments may be health care, education, transport, and housing.
consistent with the agendas of developing country gov- As the Bank more strongly began to embrace a mission
ernments. For example, governments may wish to reduce of poverty alleviation in the 1990s, the role of renewables
local air pollution as part of environmental policy, or within the Bank started to become less driven by an
promote domestic renewable-energy-based businesses energy-sector agenda and more driven by a rural-devel-
and industries as part of industrial or technological pol- opment agenda. Still, there are obstacles to both agendas
icy. They may wish to accelerate rural electri"cation with * the subject of the next section.
renewables through the private sector in the absence of
su$cient public resources to do so. They may wish to
attract grant "nancing for certain technologies from bi- 3. Factors limiting renewable energy lending by the Bank
lateral and multilateral donor agencies. And some devel-
oping countries, such as China and India, are beginning Borrower commitment to grid-connected renewable
to embrace ambitious plans and quantitative targets for energy and the productive use of renewable energy in
future renewable energy development, and look to the rural development may overcome many of the internal
Bank and other multilateral agencies for both "nancing barriers to renewable energy within the Bank. Neverthe-
and assistance with policy development (Gupta, 2000; less, these barriers still exist. Many internal barriers re-
SDPC, 2000; SETC, 2000). sult because renewable energy projects are not simple
O!-grid renewable energy applications highlight a dis- investment projects. Rather, projects may develop new
parity between energy-sector agendas and rural-develop- "nancing and institutional mechanisms, support rural
ment agendas. Consider the case of rural electri"cation. energy enterprises or private-power-project developers,
By 1999, the Bank had approved a portfolio of 12 `solar conduct extensive training, promote consumer aware-
home systemsa projects. These projects by and large ness, facilitate codes and standards, and undertake other
support the private sector to provide small amounts of so-called `barrier removala activities that are often part
electricity to individual homes for lighting, TV, radio, of elaborate technical assistance strategies rather than
and other small appliances through the use of solar pure "nancing (see Martinot and McDoom, 2000 for
photovoltaic systems (Martinot et al., 2000a, b). Initially, detailed descriptions of project approaches). In inter-
these projects were driven by a `supply-sidea agenda views with a number of Bank managers by the author
which went something like this: (i) electri"cation is neces- during 1998 and 1999, several types of internal barriers to
sary for rural development; (ii) despite large gains in rural renewable energy were revealed.
692 E. Martinot / Energy Policy 29 (2001) 689}699

3.1. Pressure on project preparation resources 3.5. Limited experience and interest in the social and rural
development sectors
Task managers and investment o$cers are under pres-
sure to deliver projects in the shortest possible timeframe, Renewable energy is still seen as primarily an energy-
and often lack the resources, knowledge, and time to sector activity. The social and rural development sectors
pursue renewable energy projects. Preparation time and still have limited experience with renewable energy, des-
expense for renewable energy projects can be substan- pite being in a good position to implement rural energy
tially more than for conventional projects (one manager projects and utilize community-based institutions. As an
estimated 30}60% more), which makes them unattrac- indication of how far the social sectors have to go, a 1999
tive to a manager with "xed resources and time pressures. Bank review of current thinking on rural infrastructure
Managers under pressure to deliver new investments in issues did not mention solar home systems or renewable
the shortest possible time will tend to avoid renewable energy (Pouliquen, 1999).
energy projects, all else being equal.
3.6. GEF project preparation burden
3.2. Nontraditional project risks
All Bank renewable energy projects in the 1990s were
In additional to traditional project risks (procurement, facilitated by GEF grants (many said none of these pro-
construction, future energy prices, and cost overruns), jects would have happened without the GEF). Yet addi-
renewable energy projects often require new technology tional burden comes from the need to prepare separate
experience, new institutional development, new "nanc- GEF documentation and get GEF Council approval for
ing/contractual mechanisms, and technology acceptance these grants. Even though special GEF project-prepara-
by "nanciers and stakeholders. Building new institutions tion resources are available, the process can be burden-
in projects is often an expensive, di$cult and time-inten- some for Bank managers. They generally do not want the
sive activity. Many project features could be considered added complexity, and, if unfamiliar with GEF proced-
`experimentala because an accepted set of best practices ures, do not want to have to learn another set of rules.
for the kinds of project interventions needed to develop
renewable energy markets simply does not exist. These 3.7. Corporate reorganization
nontraditional project risks require a risk-taking mentality
and incentives that do not penalize managers when parts During the 1990s, renewable energy expertise accumu-
of projects do not turn out as expected. lated within the Bank's central Energy Department. Dur-
ing the period 1998}1999, as part of a corporate-wide
3.3. Lack of appropriate skills reorganization, most of the Bank's renewable energy
experts not in ASTAE were moved from the central
Renewable energy projects can be quite di!erent from Energy Department to Africa, Latin America/Caribbean,
traditional energy-sector investment projects. Task man- and South Asia operating regions. Although these ex-
agers in the energy sector and their normal set of consul- perts became tied together through a Bank-wide `Rural
tants may not possess the necessary skills or knowledge. and Renewable Energy Thematic Groupa, the success of
Relevant lessons and `best practicesa must be identi"ed these new groupings remained to be seen. Perhaps the
and incorporated into project design, for example institu- ASTAE concept can be replicated in each region through
tional development, test activities, and social surveys. this process, but the potential downside is the dilution of
Task managers, as engineers and economists, may not be a previously existing core group of expertise.
trained or skilled in institution building. They may have
to hire consultants outside of the normal skill set or
quali"cations to which they are accustomed. In the Asia 4. Opportunities for the Bank: lessons from
operating units, ASTAE has assisted task managers with emerging experience
best practices and in doing so has played an important
role in facilitating energy projects. Lessons from Bank renewable energy projects are just
beginning to emerge. Of the 17 projects approved since
3.4. Lack of attention in country assistance strategies 1992, only one project had been formally completed by
the end of 1999 (Mauritius Sugar Bio-Energy). Two
Rural and renewable energy are not often explicitly others were essentially completed (India Renewable
called out in the Bank's formal country assistance strat- Resources Development and Phillippines Leyte-Luzon),
egies. Consequently, country directors * who control three had substantial implementation progress (Small
preparation and manpower budgets and assignments and Medium Scale Enterprise Program, Sri Lanka
* have no mandates or strong managerial incentives to Energy Service Delivery, and Tunisia Solar Water
devote their resources to renewable energy projects. Heating), and the remainder were in the early stages of
 E. Martinot / Energy Policy 29 (2001) 689}699 693

implementation. Thus the volume of available experience loans (at typically 2}5 year terms) to purchase the sys-
is still small relative to the potential for future project tems from a separate non-pro"t micro"nance institution
assessments and lessons as the portfolio matures (Mar- with many local branches and strong ties to the commu-
tinot, 1998). nities in which it operates. By 1999, over 1000 systems
Nevertheless, emerging lessons point to at least "ve had been installed and arrangements were in place for
areas where the Bank has been e!ective in lending and installation of an additional 5000 systems with micro-
technical assistance for renewable energy (Martinot and "nance through 2000.
McDoom, 2000; Martinot et al., 2000, 2001). These are
highlighted below. Further analysis and experience has 4.2. Support for electric power policy frameworks
and will likely point to areas beyond these "ve as well.
Frameworks that enable IPPs to invest in renewable
4.1. Support for renewable energy xnancing energy and sell power to a utility network are funda-
mental to development of grid-connected renewable en-
Project "nancing for developers of grid-connected ergy (other policies are also helpful; see Martinot, 2000).
renewable energy projects has been one barrier to renew- Experience in Mauritius and Sri Lanka suggests that the
able energy that Bank projects have helped overcome. Bank is able to facilitate such frameworks.
The India Renewable Resources Development project In Mauritius, the Sugar Bio-Energy project indirectly
occurred in parallel with the explosive market growth of catalyzed dramatic changes in electricity generation from
wind power in India in the mid-1990s, fuelled by favor- bagasse. The project "nanced e$ciency investments in
able investment tax policies and a supportive regulatory sugar mills to provide surplus bagasse for power genera-
framework. By 2000, more than 1200 MW of wind tur- tion. The project also provided technical assistance and
bine capacity had been installed in India and dozens of technology demonstrations to promote private/public
wind turbine manufacturers had emerged. During the sector cooperation in power plant ventures and to opti-
1990s, the project directly "nanced 41 MW of private- mize the use of sugar cane for power generation. Partly as
sector wind turbine installations in India. More impor- a result of the project, electricity generation from bagasse
tantly, the project strengthened the capabilities of the in Mauritius increased from 70 GWh/yr in 1992 to
India Renewable Energy Development Agency (IREDA) 118 GWh/yr by 1996. A project completion report stated
to promote and "nance private-sector investments, and that `extensive dialogue between the public and private
more than 360 MW of wind projects were "nanced sector on design work, the least-cost power development
through IREDA. The project also promoted the accepta- plan, and power purchasing agreements have directly or
bility of wind power among investors and banking insti- indirectly led to the development of other power plantsa.
tutions. As a result, along with favorable market One of the lessons from the Mauritius project is how
conditions, many sources of "nance became available for creating an investment climate for renewable energy
wind power, which was a key project goal. power projects, and creating public}private partnerships,
In Sri Lanka, the Energy Services Delivery project was can lead to supportive regulatory frameworks. In this
providing "nancing to private-sector small-hydropower case, the project led to the establishment of an IPP
developers and was innovating micro"nancing schemes framework and an administrative focal point for pri-
for installations of rural solar home systems. By 2000, the vate/public sector partnerships in IPP development.
project had "nanced, through commercial-bank inter- A project evaluation states that `the project's major
mediaries, more than 21 MW of small hydropower by accomplishment was progress in helping to establish an
independent power producers (IPPs). institutional and regulatory framework for private power
Business "nancing for delivery of rural energy services generation in Mauritius and the provision of technical
and credit to improve the a!ordability of those services studies and trials to support technologies for improved
among rural households have been signi"cant barriers to bagasse production and improved environmental
renewable energy in rural areas. The Sri Lanka project monitoringa. A planned demonstration bagasse plant
was also providing business "nancing for rural solar under the project was never constructed, which suggests
home enterprises through commercial banks and was that technical demonstration can have less of an in#u-
demonstrating the initial viability of a `micro"nance ence on promoting renewable energy than other types of
modela. Under this model, piloted through the project, project interventions.
solar home systems enterprises market, sell, service and In Sri Lanka, the project successfully developed IPP
warrantee their products to rural consumers through frameworks and standardized `non-negotiablea power-
their own local sales/service o$ces. Consumers obtain purchase tari!s and contracts for power from small-
hydro, biomass and wind. The project provided enough
incentive for the national utility to adopt IPP frame-
 The late 1990s saw a marked downturn in wind power development works and agree to private-sector small hydro develop-
in India. See Jagadeesh (2000). ers, which together with the demonstration e!ects of
694 E. Martinot / Energy Policy 29 (2001) 689}699

prior mini-hydro installations and new incentives for Comoros, technical assistance through an ESMAP pro-
project developers (such as import duty waivers and ject assisted with developing the rural market for solar
income tax concessions) spurred private small-hydro de- equipment there (ESMAP, 2000b).
velopers.
4.4. Support for regulated rural energy concessions
4.3. Support for rural energy enterprises
There appears to be a clear role for the Bank to assist
The Bank has had initial success supporting rural local and regional governments to establish and regulate
enterprises in providing energy services based on renew- private-sector rural energy service concessions. The
able energy. In addition to the Sri Lanka case mentioned Argentina Renewable Energy in Rural Markets project
above, business "nancing for rural energy enterprises was was the "rst to pilot such an approach * which was also
being provided under the Small and Medium Scale En- being adopted in several other countries in Latin
terprise (SME) Program to three rural solar home America and Africa (Reiche et al., 2000). In Argentina,
systems businesses in Bangladesh, Vietnam and the 15-year concessions are to be regulated by provincial
Dominican Republic. The Bangladesh project was dem- governments and selected competitively. These conces-
onstrating a business model in which the dealer sions are obligated to provide energy services to rural
(Grameen Shakti, legally a non-pro"t) performs market- households in the concession territory upon demand.
ing, sales, service, credit provision (up to 3-year terms), They must carry out all necessary maintenance, repairs
credit collections, and guarantees. From 1997 to 1999, or replacement of components as needed to ensure the
Grameen Shakti installed 1500 systems using this model, continuity of the electricity service to each and every
and planned to install 2000}2500 systems in 2000. In customer, and must charge standard regulated tari!s
Vietnam, sales by a private dealer were being assisted by for energy services. The Bank- and GEF-supported
a complex credit delivery scheme "nanced partly by the project is helping to resolve numerous issues associated
SME program involving the Vietnam Women's Union with such a scheme, such as setting tari!s, "nding
(VWU), an NGO, and the Vietnam Bank for Agriculture and attracting capable bidders, conducting competitive
and Rural Development (VBARD), a development bidding procedures, and ensuring service quality
"nance institution. standards.
In the Dominican Republic, SME Program business
"nancing helped a private "rm to develop a successful 4.5. Support for domestic technology manufacturing
fee-for-service business model. The "rm charges rural
households an a!ordable monthly fee for electricity from A few projects explicitly target support to improve
a solar home system and expects to be able to serve up to domestic technology manufacturing capabilities. For
50% of the population in the rural communities it serves. example, the Renewable Energy Development project in
As of April 2000, the "rm had installed over 3500 systems China "nances 190 MW of wind farms. Project develop-
and had passed the break-even point where revenues ers will construct the wind farms and sell power to
covered the direct costs of operations. The "rm was utilities through commercial power-purchase agree-
developing a business model that would support 25,000 ments. These investments would practically double the
customers and was seeking additional assistance for such installed wind capacity in China. Procurement will occur
`business-model R&Da, which it found very di$cult to through international competitive bidding to provide the
fund from operating revenue alone. lowest commercial prices possible, and the project
Technical assistance to rural energy enterprises to help strengthens the ability of Chinese wind turbine manufac-
develop sustainable local sales and service infrastructure turers to compete in such procurements. The project
is also part of many rural energy projects. For example, includes a $90 million technology improvement compon-
in Sri Lanka, the project helps train technicians, develop ent to assist domestic wind-turbine and photovoltaic
business plans, and market solar home systems. In the manufacturers to innovate, improve quality, and reduce
costs * with both technical assistance grants and
production investment loans (World Bank, 1999b).
 However, one of the lessons from the Sri Lanka project is that
variable power-purchase tari!s can hinder market development. In this
case, tari!s were tied to short-run avoided utility costs based on the 5. Strategies for involving and supporting the private
international price of oil. In 1997 and 1998 tari!s were set at the sector
equivalent of 5 cents/kWh and mini-hydro development #ourished.
However, because of the downturn in oil prices during 1998}1999, In the past three decades, `donor-drivena renewable
prices were only the equivalent of 3.5 cents/kWh in 1999. As a result, all
development essentially stopped in 1999. And this #uctuation has energy projects, mostly through bilateral assistance, have
seriously hurt the longer-term interest of private mini-hydro developers resulted in signi"cant equipment installations in de-
in Sri Lanka (Bandarenke, 2000). veloping countries. But an `equipment demonstrationa
 E. Martinot / Energy Policy 29 (2001) 689}699 695

mentality has prevailed, in which the main objective has servers in the United States and Europe by the author in
been simply the installation and maintenance of a certain 1998 and 1999 suggested a number of speci"c areas where
number of systems or installed capacity. While there are the Bank and GEF could support the private sector with
exceptions, these projects have generally failed to pro- renewable energy development.
mote commercial sustainability and replication. In fact,
the expectation of continued equipment demonstrations 5.1. Assist xrms with business plans
through bilateral assistance can hinder the formation of
commercial markets. In addition, there are examples Firms want help with implementing their own business
of growing markets for renewable energy applications plans, for which they may have already expended con-
in developing countries that have emerged through siderable resources to develop. These business plans may
private-sector initiative and favorable national policies lack certain elements to become `bankablea, thus com-
with little explicit development assistance, such as the panies want to get grant assistance to support their plans.
solar photovoltaic market in Kenya (Barnett, 1990; For example, grants could be used for developing market
Foley, 1993; Kozlo! and Shobowale, 1994; Goldemberg infrastructure (site surveys, distributors, dealers, service,
and Johansson, 1995; Kammen, 1999; Duke et al., marketing (demonstrations), for subsidizing capital
2000). equipment costs, and for guaranteeing "nancing risk by
In view of the lessons from bilateral assistance pro- local "nancing institutions.
grams, and the potential for private-sector investment to
eclipse development assistance for renewable energy, the 5.2. Support favourable regulatory policies
Bank and GEF have sought to involve and support the
private sector. As illustrated in the previous section, For grid-connected projects, "rms welcome electric
recent Bank renewable energy projects have been de- power regulatory frameworks that support renewable
signed to promote commercial, sustainable markets and energy. Generally, "rms cannot in#uence these frame-
regulatory frameworks for a variety of renewable energy works themselves, so this is a high-value-added activity
applications. These projects are aligned with the GEF for the Bank from their perspective. Some brought up the
Operational Program `Promoting the Adoption of Re- `Electricity Feed Lawa approach that has been used in
newable Energy by Removing Barriers and Reducing Germany and Spain. `If you want market volume, there
Implementation Costsa (GEF, 1997). Within this pro- is no substitute for a feed lawa said one "rm. Firms also
gram, support for the private sector can take several want the Bank to push for elimination of subsidies to
forms: conventional generation to create a more level playing
"eld for renewables.
E Support private "rms by providing "nancing, tech-
nical and business assistance, equipment subsidies,
5.3. Finance pre-feasibility studies for small companies
and marketing support. This support helps "rms ex-
pand their business, innovate and test new business
Many smaller companies work on very short time
models (i.e., demonstrate pro"tability), and lower
horizons and identify speci"c project opportunities on
costs.
a case-by-case basis. Firms need funding for pre-feasibil-
E Build capacities of government agencies and NGOs to
ity studies to support their project development work,
promote renewable energy development, often in con-
often lacking the resources themselves to invest in
junction with studies, resource assessments, and mar-
studies with uncertain outcomes. One company sugges-
ket characterization.
ted providing forgivable loans for pre-feasibility work;
E Create new "nancing vehicles like revolving funds,
the loan would only be repayable if the resulting
credit lines, and contingent business loans that are
project was viable and was "nanced. Smaller project
forgivable under speci"ed conditions.
developers especially look to grants or forgivable loans
E Develop or strengthen regulatory frameworks for
to "nance pre-feasibility studies because of the risks
grid-based independent power producers, transparent
involved.
power-purchase tari!s, and o!-grid utility concessions
that provide energy services to rural areas using re-
5.4. Reduce commercial risks and provide xnancing
newable energy technologies.
guarantees
Beyond these general topics, interviews with 21 pri-
vate-sector renewable energy technology manufacturers, Private project developers were generally not con-
trade associations, project developers, and industry ob- cerned about technical risks because they had con"dence
in their skills, technologies, and marketing ability. Com-
mercial risks are the real problem, they said: contractual
 For treatment of the Bank's policies related to electric power sector mechanisms and enforcement, cash #ow, currency issues,
reform, see World Bank (1993). uneven competition, etc. For example, one "rm suggested
696 E. Martinot / Energy Policy 29 (2001) 689}699

the Bank could provide a `secondary mortgage marketa Solar Development Group (SDG). These projects are
to provide guarantees against project risks. Another said pre-established conduits for rapidly providing business
the Bank could induce governments to guarantee power "nancing to "rms meeting established eligibility criteria
purchase agreements with utilities, so that project devel- or to eligible projects, along with business advisory servi-
opers could more easily obtain commercial "nancing ces. By 1999, however, only the SME program had been
based upon the power purchase agreements. under implementation. PVMTI was just starting in 1999
and SDG and REEF were still awaiting further capital-
5.5. Support joint ventures ization.

Firms were interested in joint ventures. Joint ventures

to produce equipment locally will be critical for some 6. Conclusion
technologies like wind turbines, where the costs of local
production may be signi"cantly lower than the costs of The Bank has put much e!ort and resources into
imports. Smaller companies need help to identify joint renewable energy in the last 10 years but still has com-
venture partners. Local joint venture partners need paratively little on the ground to show for it. The volume
capital and technical assistance and training. of completed investments from all projects is still small
relative to their targeted investments. For example, of the
5.6. Build market volume and stability more than half-million solar home systems which could
be installed as a result of approved projects, only about
One "rm said, `we need larger projects to get volume 8000 had been "nanced and installed by 1999. Of the
up2. The future looks potentially rosy if we can build several hundred megawatts of grid-connected wind
up the volumea. With increased volume, private invest- power, biomass and small-hydro in approved projects in
ments in market infrastructure, service, and marketing the 1990s, less than 100 MW total had been installed.
become commercially viable. Companies want a long- Several factors have limited and continue to limit the
term market in which their investment to enter the mar- ability of the Bank to lend for renewable energy. Still, one
ket is not jeopardized by a market collapse after outside could argue that past e!orts, conducted together with
support ends. They do not want one-time interventions client countries, have raised awareness, understanding,
or short programs (i.e., 1}3 years) that will leave them interest, and commitment among these countries to con-
hanging at the end. They want `market involvement sider renewables more seriously in the future and to enact
riska to be reduced through long-term programs lasting domestic policies and programs more inclusive of renew-
5}10 years. able energy. Through studies and dialogue, the Bank
appears able to develop borrower understanding and
5.7. Pilot and test innovative business models commitment to renewable energy applications. GEF and
bilateral "nancing for studies, project preparation, and
Especially for rural markets, "rms believed that proven project "nancing has also been a key factor facilitating
organizational and businesses structures and models to project decisions on both the Bank and borrower side,
e!ectively deliver energy services using renewables do especially in the face of risk. The Bank should continue to
not yet exist. The Bank could help pilot di!erent business use education, technical assistance and persuasion to
models, even with small-sized grants. `There is a high increase client demand for renewables while at the same
degree of risk involved, as some models won't work; but time removing internal barriers to increase the Bank's
to learn we need to try many di!erent approachesa, one ability and willingness to lend for renewables.
"rm said. Another agreed: `there's a high value-added in But these are di$cult tasks, and no one should expect
terms of developing and improving business models2. the Bank to be able to expend adequate resources and
You want to stimulate markets based on these business time needed to do so * there is clearly a role for others
modelsa. Yet another said `grants should be tools for here. It may be di$cult for outside agents to help the
taking business risksa. Fundamentally, the Bank and Bank to increase internal support for renewables and
GEF should support business innovation because no one overcome internal barriers. But NGOs, bilateral donors
else is in a position to take such risks, "rms seemed to and other international agencies can work with develop-
agree. ing countries and countries in transition to increase
The International Finance Corporation (IFC), the pri-
vate-sector a$liate of the Bank, is implementing four
private-sector projects with GEF support that respond
to many of these concerns: the Small and Medium Scale  IFC has also "nanced, without GEF support, hydro projects in
several countries, PV manufacturing in China, and biomass cogener-
Enterprise Programme (SME), the Renewable Energy ation and geothermal power plants in Guatemala. A variety of addi-
and Energy E$ciency Fund (REEF), the Photovoltaic tional geothermal, wind, biomass, and PV projects were under
Market Transformation Initiative (PVMTI), and the consideration.
 E. Martinot / Energy Policy 29 (2001) 689}699 697

Table 1
World Bank Group renewable energy project portfolio (end of 1999)

Bank GEF Total

Project "nance "nance costs
(year approved by Bank) ($ mil.) ($ mil.) ($ mil.) Technology applications

India: Alternate energy/renewable resources development (1992) 190 26.0 450 PV for home/commercial; wind
& mini-hydro for grid
Mauritius: Sugar bio-energy development (1992) 15 3.3 55 Bagasse power for grid
Costa Rica: Tejona wind power (1993) 0 3.3 31 Wind power for grid
Philippines: Leyte-Luzon geothermal (1994) 227 30.0 1334 Geothermal for grid
Tunisia: Solar water heating (1994) 0 4.0 21 Solar hot water for home, public,
commercial markets
Small and Medium Scale Enterprise program (IFC, 1995) 0 1.6 5 PV home systems for rural o!-grid markets
["gures for renewable energy subloans only]
Lithuania: Klaipeda geothermal demonstration (1996) 6 6.9 18 Geothermal for district-heat production
Indonesia: Renewable energy small power (1997) 66 4.0 141 Mini-hydro, biomass
Indonesia: Solar home systems (1997) 20 24.0 118 PV home systems for rural o!-grid markets
Renewable Energy and Energy E$ciency Fund (IFC, 1997) 0 30.0 130 All applications possible
Sri Lanka: Energy services delivery (1997) 24 5.9 55 PV home systems o!-grid; wind/mini-hydro
for grid
PV Market Transformation Initiative (IFC, 1998) 0 30.0 120 All applications of PV
Argentina: Renewable energy in rural markets (1998) 30 10.0 120 PV and wind for variety of rural o!-grid
applications
Lao PDR: S. provinces renewable energy pilot (1998) 1.5 0.7 2.1 Village power for o!-grid
Cape Verde: Energy & water sector reform and development 18 4.9 65 PV home systems for o!-grid; wind power
(1999) for grid
China: Renewable energy development (1999) 100 35.0 445 PV home systems for rural o!-grid markets
Solar Development Group (IFC, 1999) 6 10.0 50 PV home systems o!-grid; other PV
applications
Projects with GEF grants approved (and year); still to be approved by Bank ("nancing subject to change)
India: Solar thermal electric (GEF, 1996) 0 49.0 254 Solar thermal power plant
Brazil: Biomass power commercial demonstration (GEF, 1997) 53 40.5 122 Biomass integrated gasi"cation/gas turbine
Benin: Decentralized rural energy (GEF, 1998) 2.2 1.1 6 PV home systems for rural o!-grid markets
Togo: Decentralized rural energy (GEF, 1998) 2.2 1.1 6 PV home systems for rural o!-grid markets
Guinea: Rural energy (GEF, 1999) 3 2.0 10 PV home systems and village-power for
o!-grid
Hungary: Szombathely CHP/biomass (GEF, 1999) 10 2.5 28 Biomass power for grid
India: Biomass energy for rural India (GEF, 1999) 0 4.2 9 Biogas for rural use (lighting, cooking,
water)
Mexico: Hybrid solar thermal power plant (GEF, 1999) n/a 49.4 178 Solar thermal power plant
Mexico: Renewable energy for agriculture (GEF, 1999) 14 8.9 31 PV and wind for electricity, water pumps in
agriculture
Morocco: Solar based thermal power plant (GEF, 1999) n/a 43.9 114 Solar thermal power plant
Philippines: CEPALCO distributed generation PV (GEF, 1999) [see left] 4.0 8 Distributed grid-connected PV
[$21 m. "nancing facility for CEPALCO]
Poland Geothermal and Environment (GEF, 1999) 44 5.4 85 Geothermal for heat supply
Thailand: Removing barriers to biomass power/CHP (GEF, 1999) 0 6.8 73 Biomass power for grid

Total project costs may be rounded and may include amounts for other, non-renewable-energy components.

understanding of the contribution renewables can make energy in developing countries while meeting pressing
to development priorities, and to enact policies that sup- development needs. But many of these lessons still re-
port renewables. main to be learned in the future as the project portfolio
The Bank needs to devote considerable attention and matures.
resources to integrating renewable energy into the devel-
opment agenda of its social and rural development units.
It should also consider additional strategies for involving Appendix. World Bank Group renewable energy
and supporting the private sector. The lessons and ex- project portfolio
periences from existing projects have the potential to
contribute substantially to an understanding of the most After a number of geothermal projects in the 1970s and
e!ective approaches to developing markets for renewable 1980s, renewable energy lending by the World Bank
698 E. Martinot / Energy Policy 29 (2001) 689}699

Group began more seriously with the India Renewable Bank 1994, 1996b; ESMAP, 1996; Taylor and Bogach,
Resources Development Project approved in 1992. Since 1998).
then, 17 projects with renewable energy components in As concern about global climate change increased in
16 countries have been approved (see Table 1). All of the 1990s, the Bank conducted a retrospective review of
these projects are partially "nanced with grants from the energy projects * a `backcasting studya * to look at
Global Environment Facility. Table 1 also contains an alternatives to traditional Bank energy projects (World
additional 13 near-term pipeline projects with grants Bank, 1998). The study reviewed project preparation
already approved by the GEF but not yet fully prepared reports for a sample of Bank energy loans between 1990
and approved by the Bank (GEF, 1999; Martinot and and 1996, in part to analyze whether assigning a monet-
McDoom, 2000). In addition, other projects are in earlier ary value to carbon emissions would encourage invest-
stages of preparation (in the `pipelinea). ment in low-carbon alternatives. The study developed
Financing for the 17 projects amounts to $700 million plausible alternative options to reduce carbon emissions
in Bank loans and credits and $230 million in GEF in the sampled projects using wind, mini-hydro and
grants, and leverages total project costs of $3.2 billion. biomass.
Financing for all 30 projects could amount to $1.3 billion
in Bank/GEF loans, credits, and grants, and leverage
total project costs in excess of $4 billion. In most projects,
References
substantial shares of project "nancing come from other
donors, client countries, and/or the private sector. Ahmed, K., 1994. Renewable energy technologies: a review of the status
Projects target o!-grid photovoltaics (PV) for rural and costs of selected technologies. World Bank Technical Paper
markets; wind, biomass, bagasse and mini-hydro power No. 240, Washington, DC, World Bank.
generation for utility markets; village-scale power from Anderson, D., Ahmed, K., 1995. The case for solar energy investments.
PV, wind, mini-hydro and biogas; solar hot water heat- World Bank Technical Paper No. 279. Washington, DC, World
Bank.
ing for home, public and commercial markets; geother- Bandarenke, R.D., 2000. Grid-connected small hydro power in Sri
mal for electricity or heat production; solar thermal Lanka: the experiences of private developers. Paper presented at the
central-station power plants; biomass integrated gasi"ca- International Conference on Accelerating Grid-Based Renewable
tion/gas turbine power plants; and distributed grid- Energy, Washington, DC, March 7}8, Washington, DC, US Energy
connected PV. Association.
Barnett, A., 1990. The di!usion of energy technology in the rural areas
Besides formal projects, the Bank supported several of developing countries: a synthesis of recent experience. World
renewable energy technical papers and studies (Ahmed, Development 18(4), 539}553.
1994; Liebenthal et al., 1994; Anderson and Ahmed, 1995; Cabraal, A., Cosgrove Davies, M., Schae!er, L., 1996. Best practices for
Foley, 1995; Stassen, 1995; Cabraal et al., 1996, 1998; photovoltaic household electri"cation programs: lessons from ex-
Piscitello and Bogach, 1998). The joint UNDP/World periences in selected countries. World Bank Technical Paper No.
324, Washington, DC, World Bank.
Bank Energy Sector Management Assistance Program Cabraal, A., Cosgrove-Davies, M., Schae!er, L., 1998. Accelerating
(ESMAP) program contributed to or sponsored many of sustainable photovoltaic market development. Progress in Photo-
these studies, and also conducted renewable energy voltaics: Research and Applications 6, 297}306.
development activities in several countries (ESMAP, Duke, R., Graham, S., Hankins, M., Jacobson, A., Kammen, D. et al.,
1999, 2000a, b). 2000. Field performance evaluation of amorphous silicon (a-Si)
photovoltaic systems in Kenya: methods and measurements in
Country-speci"c renewable energy studies, technical support of a sustainable commercial solar energy industry. ESMAP
assistance and outreach were also conducted in particu- Technical Paper No. 5, Washington, DC, World Bank.
lar countries. China is perhaps the most prominent Energy Sector Management Assistance Program (ESMAP), 1996. En-
example. Several ESMAP studies on rural energy in ergy for rural development in China: an assessment based on a joint
China in the early 1990s, along with a major green- Chinese/ESMAP study in six counties. ESMAP Report No. 183/96,
Washington, DC, World Bank.
house-gas mitigation study in 1994 funded by the GEF Energy Sector Management Assistance Program (ESMAP), 1999. A
and other sectoral studies eventually led to the current review of the renewable energy activities of the UNDP/World Bank
Renewable Energy Development Project there (World Energy Sector Management Assistance Programme 1993}1998.
ESMAP Report No. 223/99, Washington, DC, World Bank.
Energy Sector Management Assistance Program (ESMAP), 2000a.
Energy services for the world's poor. Energy and Development
 Project preparation reports were formerly called `Sta! Appraisal Report 2000, Washington, DC, World Bank.
Reportsa and now are called `Project Appraisal Documentsa. These Energy Sector Management Assistance Program (ESMAP), 2000b. In
reports dated prior to 1997 are not public, but a policy change made the search of better ways to develop solar markets: the case of Comoros.
"nal versions of reports issued in 1997 or later available to the public ESMAP Report No. 230/00, Washington, DC, World Bank.
from the Bank's public information center (www.worldbank.org/pic). Flavin, C., 1997. Banking against warming. World Watch (Novem-
Descriptions of all projects discussed in this paper can also be obtained ber/December), 25} 35.
from Martinot and McDoom (2000), which can be downloaded from Flavin, C., Tunali, O., 1996. Climate of hope: new strategies for stabiliz-
www.gefweb.org. Project descriptions are available for some projects ing the World's atmosphere. Worldwatch Paper No. 130, Washing-
from www.worldbank.org/astae. ton, DC, Worldwatch Institue.
 E. Martinot / Energy Policy 29 (2001) 689}699 699

Foley, G., 1993. Renewable energy in third world development assist- President's Committee of Advisors on Science and Technology
ance. In: Jackson, T. (Ed.), Renewable Energy: Prospects for Imple- (PCAST), 1999. Powerful Partnerships: The Federal Role in Inter-
mentation. Stockholm Environment Institute, Stockholm, pp. national Cooperation on Energy Innovation. Washington, DC,
193}202. Executive O$ce of the President.
Foley, G., 1995. Photovoltaic applications in rural areas of the develop- Piscitello, S., Bogach, S., 1998. Financial incentives for renewable
ing world. World Bank Technical Paper No. 304, Washington, DC, energy development. Proceedings of an International Workshop,
World Bank. February 17}21, 1997, Amsterdam, Netherlands. World Bank
Global Environment Facility (GEF), 1997. Operational Programs. Discussion Paper No. 391, Washington, DC, World Bank.
Washington, DC. Pouliquen, L.Y., 1999. Rural infrastructure from a World Bank per-
Global Environment Facility (GEF), 1999. Operational Report on spective: a knowledge management framework. World Bank dis-
GEF Programs. December 31, Washington, DC. cussion paper, Washington, DC.
Goldemberg, J., Johansson, T.B., (Eds.), 1995. Energy as an Instrument Ramana, P.V., (Ed.) 1997. Rural and Renewable Energy: Perspectives
for Socio-Economic Development. New York, United Nations from Developing Countries. New Delhi, Tata Energy Research
Development Program. Center.
Grubb, M., 1995. Renewable Energy Strategies for Europe: Vol. I, Reiche, K., Covarrubias, A., Martinot, E., 2000. Expanding electricity
Foundations and Context. Earthscan Publications, London. access to remote areas: o!-grid rural electri"cation in developing
Gupta, A., 2000. Policy approaches: the India experience. Paper pre- countries. In: Isherwood, G. (Ed.), World Power 2000. Isherwood
sented at the International Conference on Accelerating Grid-Based Production Ltd., London, pp. 52}60.
Renewable Energy, Washington, DC, 7}8 March, Washington, DC, Shepherd, D., 1998. Creating a market for renewables: electricity policy
US Energy Association. options for developing countries. Discussion paper, Washington,
Institute for Policy Studies and International Trade Information DC, World Bank Environment Department.
Service, 1998. The World Bank and the G-7: still changing the Stassen, H.E., 1995. Small-scale biomass gasi"ers for heat and power:
earth's climate for business. Downloaded from www.seen.org on a global review. World Bank Technical Paper No. 296, Washington,
10/12/98. DC, World Bank.
International Energy Agency (IEA), 1997. Enhancing the Market De- State Development Planning Commission of China (SDPC), 2000.
ployment of Energy Technologies: A Survey of Eight Technologies. China New & Renewable Energy (1999 White book). Beijing, China
OECD, Paris. Planning Press.
International Energy Agency (IEA), 1999. The Evolving Renewable State Economic and Trade Commission of China (SETC), 2000. Indus-
Energy Market. OECD, Paris. trial development planning for new and renewable energy. Paper
Jagadeesh, A., 2000. Wind energy development in Tamil Nadu and presented at the US}China Renewable Energy Forum, April 19}20,
Andhra Pradesh, India * institutional dynamics and barriers. Rosslyn, VA, National Renewable Energy Laboratory, Golden CO.
Energy Policy 28, 157}168. Taylor, R.P., Bogach, V.S., 1998. China: a strategy for international
Kammen, D., 1999. Promoting appropriate energy technologies in the assistance to accelerate renewable energy development. World
developing world. Environment 41(5), 11}15; 34}41 (June). Bank Discussion Paper No. 388, Washington, DC, World Bank.
Kozlo!, K., Shobowale, O., 1994. Rethinking Development Assistance World Bank, 1993. The World Bank's Role in the Electric Power
for Renewable Energy. Washington, DC, World Resources Institute. Sector. Washington, DC.
Liebenthal, A., Mathur, S., Wade, H., 1994. Solar energy: lessons from World Bank, 1994. China issues and options in greenhouse gas emis-
the Paci"c Island experience. World Bank Technical Paper No. 244, sions control: summary report. Report of a joint study team from
Washington, DC, World Bank. the National Environment Protection Agency of China, The State
Martinot, E., 1998. Monitoring and evaluation of market development Planning Commission of China, UN Development Programme,
in World Bank * GEF climate change projects: Framework and and the World Bank, Washington, DC.
Guidelines. World Bank Environment Department Paper No. 66, World Bank, 1996a. Rural Energy and Development: Improving En-
Washington, DC. ergy Supplies for 2 Billion People. Washington, DC.
Martinot, E., 2000. Power sector reform and environment: a role for the World Bank, 1996b. China renewable energy for electric power. Report
GEF? Paper prepared for the GEF STAP brainstorming on power No. 15592-CHA, Washington, DC.
sector reform, Bangalore, India, 26}28 June, Washington, DC, World Bank, 1998. The e!ect of a shadow price on carbon emissions in
Global Environment Facility. the energy portfolio of the World Bank: a backcasting exercise.
Martinot, E., McDoom, O., 2000. Promoting Energy E$ciency and Environment Department Paper No. 63, Washington, DC.
Renewable Energy: GEF Climate Change Projects and Impacts. World Bank, 1999a. Fuel for thought: environmental strategy for the
Washington, DC, Global Environment Facility. energy sector. Sector strategy paper prepared jointly by the Envi-
Martinot, E., Ramankutty, R., Rittner, F., 2000. The GEF solar PV ronment Department; the Energy, Mining, and Telecommunica-
portfolio: emerging experience and lessons. Washington, DC, Glo- tions Department; and the International Finance Corporation,
bal Environment Facility. Washington, DC. [available at www-esd.worldbank.org].
Martinot, E., Cabraal, A., Mathur, S., 2001. World Bank/GEF World Bank, 1999b. China renewable energy development project.
solar home systems projects: experiences and lessons learned Project Appraisal Document, Report No. 18479-CHA, Washington,
1993}2000. Renewable and Sustainable Energy Reviews 5(1), 39}87. DC.