EVALUATION BRIEF NUMBER 12
Introduction
General Budget Support (GBS) has stirred interest among many aid donors.
Rather than running their own projects, donors provide GBS funds to a 
government, which spends the money on its own development projects. GBS
funds are not earmarked for specific uses, but support the government’s overall develop￾ment effort. Another approach is Sector Program Assistance (SPA), where policy-based
conditionality promotes institutional reforms. A third is the Sector-Wide Approach
(SWAp), where all donor assistance is coordinated in support of a common set of
reforms and a sector expenditure program. 
This evaluation uses Malawi as a case study to identify
the country conditions needed to make these aid
approaches successful—and pitfalls and problems to
avoid. The study, part of a multicountry field evalua￾tion effort, was undertaken by the Evaluation Studies
division of USAID’s Bureau for Policy and Program
Coordination (PPC). In addition to Malawi, PPC
completed studies on Mozambique,1 Nicaragua, and
East Timor. The Bureau for Africa has completed a
GBS study of Tanzania.2
GBS donors argue that if a developing country has a
logical poverty reduction plan, a sound budget 
system, and the ability to implement development
projects, it makes sense for donors to provide GBS.
Those donors see GBS as an effective way to build
government commitment and ownership. They 
note that if development is to be successful and self￾sustaining, the developing country must “own” the
process. GBS donors are enthusiastic about their new
aid approach and feel it should be used in most
developing countries. Other donors note that GBS
might work well in some countries but not in others:
country conditions matter.
Malawi had successful sector policy reform programs
in the mid-1990s. But in recent years there have been
■ USAID sector programs lasted
five to nine years. Conditions
changed, and the original plans
became outdated. Much better
would be a short timeframe of
two to three years. 
■ Policy reform programs can
change policies and create an
enabling environment. But 
to implement new policies, 
there must be accompanying 
institutional changes. To varying
degrees, this did not occur in
Malawi, a disconnect that proved
to be a problem for USAID’s three
sector policy reform programs.
■ A common vision and commit￾ment is needed by government
leadership and USAID manage￾ment. Changing priorities and
personnel changes for both the
government and USAID delayed
reform efforts.
KEY IDEAS
NOVEMBER 2004 BUREAU FOR POLICY AND PROGRAM COORDINATION PN-ADA-356
What Conditions Favor the Success of General 
Budget Support and Sector Program Assistance?
Malawi Country Case Study
1 PN-ACU-999 and PN-ACW-878, available online at www.dec.org.
2 PN-ADA-029, available online at www.dec.org.
serious problems with GBS and, to a lesser extent,
SPA. Malawi’s economic numbers tell a sad story.
Annual per capita GDP growth rates were a dismal
–0.2 percent from 1982 to 1992 and only slightly
better, at +1.4 percent, from 1992 to 2002. Per
capita gross national income was $220 in 1998 but
had fallen to $160 in 2002. Sixty-five percent of the
population lives below the national poverty line,
and life expectancy at birth is only 38 years.
Donors want to help Malawi, but they are reluctant
to move from donor-managed projects to GBS,
SPA, or SWAps. While Malawi’s development 
policies may be sound, all too often they are not
reflected in budget expenditures. The government
lacks financial and monetary discipline and, as a
result, donors lack confidence in government budg￾et management. Finally, government corruption is a
growing problem. Macroeconomic and budget con￾ditions will have to improve before these assistance
approaches will be widely accepted by donors. 
In early 2004, Malawi failed six Millennium
Challenge Account (MCA) indicators, the most
troubling being those dealing with corruption and
fiscal policy. With the election of a reform-minded
government in mid-2004, the indicators have
improved, and Malawi is now an MCA threshold
country. 
Assistance Approaches
While donor-managed projects are the
usual way to deliver assistance, there are
other approaches. In Malawi donors
have used SPA, SWAps, and GBS. It is important 
to understand the different approaches and how
they can contribute to development. 
Sector Program Assistance: SPA refers to a set of
activities unified around a common output that
includes significant investment and policy issues.
Donor assistance usually supports policy and 
institutional reforms in broad sectors such as 
agriculture, education, or health. USAID’s 
experience in Malawi with SPA includes 
■ NATURE (Natural Resources and the
Environment)
■ GABLE (Girls’ Attainment of Basic Literacy
and Education)
■ ASAP (Agriculture Sector Assistance Program)
All of these sector programs were conditionality￾based. Disbursements were made following govern￾ment implementation of policy and management
reforms. USAID funds were not earmarked for any
specific task. USAID’s key interest was that the
government adopt critical sector reforms to
improve program efficiency and beneficiary impact. 
Some donors have sought to move away from such
conditionality-based approaches in favor of a model
that emphasizes partnership with government and
other donors. The government and donors agree in
advance on the direction of change, which limits
the need for stringent conditionality. Ideally, the
reforms are in place and attention can shift to the
manner in which funds are spent and whether the
funds are supporting the objectives of the agreed
strategies. It is within this context that SWAps and
GBS have emerged. 
Sector-Wide Approach: Individual donors usually
negotiate their own SPA agreements and set their
own conditionality. In contrast, in a SWAp, all
donor assistance is coordinated in support of a
common set of reforms and a sector expenditure
program. In actual practice, SWAps can be 
many things. SWAp financing may include donor￾managed project aid, technical assistance, or cash
Joseph Lieberson, Team Leader
PPC, Evaluation Studies Division
Diane Ray
Academy for Educational Development
Brian Frantz
Bureau for Africa
PREPARED BY
2 Evaluation Brief No. 12
transfers. Donors may pool their funds in a 
“basket” outside the government’s budget, or 
they may provide cash transfers to the government
that are disbursed through the government’s own
budget process. The defining characteristic of a
SWAp is that donor funding supports a set of 
sector policy reforms and a single expenditure 
program. The government provides the leadership,
and donors coordinate a common approach to
support the government’s effort. The only func￾tioning SWAp in Malawi is the HIV/AIDS 
“mini-SWAp.” 
A SWAp is an excellent way to focus donor efforts
on sector policies and institutional issues, along
with needed technical assistance, training, and 
supporting projects. SWAps encourage government
ownership and direction of the development
process and support government budget discipline
and accountability. SWAps have drawbacks, howev￾er. For example, there could be a well-functioning
SWAp in the health sector. However, if the govern￾ment does not pay workers on time, or trade con￾trols prevent the import of essential medicines, it
could be difficult to achieve good results. GBS
donors argue that addressing more fundamental,
government-wide issues is only possible with a GBS
arrangement. 
General Budget Support is used by donors when
they are satisfied with the government’s overall
development policy environment. With agreement
on policy and the budget as a whole, there is no
need to earmark aid flows to specific projects.
There is also no need for conditionality-based dis￾bursements. Donor funds are disbursed to the host
government, which uses its own allocation, pro￾curement, and accounting systems to implement its
development programs. Since support is for the
budget as a whole, donors are interested in overall
developmental results. 
The benefits claimed for GBS are improved recip￾ient-government efficiency, effectiveness, and
ownership of development activities, along with
reduced transaction costs for the government and
donors. From the donors’ perspective, GBS
improves donor coordination and policy dialogue
while firmly fixing responsibility on the govern￾ment to achieve results. GBS works best in coun￾tries that have stable, progressive, and accountable
governments with good management skills; a
recent history of good macroeconomic policies
and strong economic growth; and good working
relations with donors. Development is not helped
by providing GBS to a government that lacks
skilled staff to run effective projects, has ineffec￾tive institutions and inappropriate economic poli￾cies, lacks an adequate financial management sys￾tem, or suffers from high levels of corruption. In
such situations, donor-managed projects are more
appropriate. 
The Development Problem
Determines the Approach 
I
f policy reforms are needed, the donor may con￾centrate on the reforms. The donor and the
government agree on the need for reforms, and
donor funds are provided to the government once
the reforms are in place. The donor is concerned
about the policies, not the specifics of where the
money is spent. This is what USAID did with its
SPA support to NATURE and ASAP.
If good policies are in place, but implementation lags,
the donor can provide funds to support the govern￾ment’s own implementation of the reforms. The
concern is ensuring the right amount of money
goes to the right programs: expenditure tracking is
important. That is what GBS donors want to do to
support implementation of Malawi’s Poverty
Reduction Strategy Paper (PRSP). 
If both policies and implementation are the problem,
the donor may target project funds to specific sec￾tor bottlenecks and problems, and disburse a por￾tion of the funds when policy reforms are achieved.
Both the policies and program expenditures are
important. That is what donors are doing with the
health sector SWAp; and that was USAID’s
approach with GABLE.
What Conditions Favor the Success of General Budget Support and Sector Program Assistance? 3
USAID’s Long History of
SPA in Malawi 
USAID’S Sector Program Assistance 
helped the government replace controls
with market-based incentives and opened
access for disadvantaged groups. Improvements in
investment efficiency, markets, and a reorienting of
government development priorities proved more
difficult and had mixed results. 
USAID funds were disbursed to Malawi’s treasury
and were not earmarked for specific government
expenditures. Unlike GBS, however, the assistance
came with conditionality and, as is often the case
with conditionality-based disbursements, serious
delays occurred when the government failed to
meet its policy commitments. The policy reforms
were sector-specific rather than the comprehensive
reforms of a PRSP or similar national development
strategy. It was also a bilateral USAID-government
of Malawi program that did not have other donor
participation. 
The programs eventually met their targets, but
there were delays. Generally they met the initial
one- and two-year benchmarks, but it was more
difficult to meet targeted reforms in later years. The
earlier targets were well defined and achievable.
However, conditions changed, and outyear targets
often were overtaken by events. While flexibility is
essential, a long-term commitment with undefined
intermediate steps does not contribute to achieving
program expectations. Such flexibility also requires
time-intensive redesign efforts that take scarce
USAID and government staff time away from
implementation.
From 2000 to early 2004, inappropriate govern￾ment fiscal and monetary policies, along with 
problems in budget management, further dimmed
the prospects for additional policy reforms or insti￾tutional changes. Given those problems, USAID
stopped using SPA in Malawi.
Agriculture Sector Assistance
Program
Starting in 1991, USAID worked with the govern￾ment to take state controls and restrictions off 
the backs of small farmers. Malawian agriculture
included highly productive large estates producing
cash crops, with the bulk of the population living
on small farms growing subsistence food crops.
USAID-supported analysis demonstrated that
smallholders growing maize would always live in
poverty. They needed higher-value cash crops. In
order to “help” small farmers, the government had
a number of commodity and trading restrictions,
along with price controls and ineffective govern￾ment parastatals that controlled both input and
output markets. ASAP supported reforms and 
provided supporting analytical studies and techni￾cal assistance. As a result of the USAID-supported
program, restrictions on private imports and
domestic trading of fertilizer and seed, as well as
price controls on fertilizer and seed, were lifted.
The control of parastatals over the production and
sale of cash crops was eliminated, and smallholder
farmers were allowed to grow highly profitable
crops previously reserved for estates. In addition,
the government was able to reduce budget outlays
by eliminating money-losing parastatals, as well as
inappropriate subsidies.
Girls’ Attainment in Basic Literacy
and Education
In the early 1990s, fewer than 50 percent of 
children aged 6–13 were enrolled in school, and
there was a large enrollment gender gap. In 1994,
USAID’s GABLE program started working with
the government on a number of policy, budget, and
organizational problems. In the first few years,
GABLE achieved remarkable results:
■ increased 60 percent to over 85 percent enroll￾ment of children aged 6–13
■ girls’ enrollment increased at all grade levels and
almost fully closed the gender gap
■ government commitment to education
increased: the education budget increased every
4 Evaluation Brief No. 12
year from 1994 to 2000; allocations were
increased for more teachers, teacher training, and
learning materials; and new-school construction
increased
■ through social mobilization, village attitudes
toward school changed, and girl’s education 
and a stronger partnership developed between
communities and their schools
■ a 30-year-old parastatal monopoly for the 
supply of learning materials ended within 
the creation of a privatized and competitive
learning materials system
GABLE was successful in the 1990s, even though
its strategy of pressing for a large number of
changes, rather than concentrating on only a few,
may have distracted government efforts. However,
by 2000 it was clear that government commitment
was weakening. Education’s share of the total budg￾et was only 13 percent in 1991 and had peaked at
27 percent in 2000. In 2004 it had slid backward
to only 17 percent. 
While getting access for girls and the poor was the
success of the 1990s, the current need is to improve
quality through improved teacher training, curricu￾lum development, and materials; dealing with
HIV/AIDS-infected teachers; improving the govern￾ment’s accounting for financial flows to the district
and school levels; and improving community involve￾ment. USAID is working with donors and the gov￾ernment to develop an education SWAp and to help
the Ministry of Education develop its sector strategy
and plans. 
The experience in education policy reform demon￾strates that success with policy reform is not a one￾time effort. As new problems develop, there will
always be a second (and often third) generation of
reforms needed. 
Natural Resources Management
and Environmental Support
Program
In the mid-1990s Malawi’s environmental base
faced increasing pressure from a growing popula￾tion that lacked economic opportunities. The
water, soil, fisheries, forests, and national park 
system were under growing pressure. National 
institutions needed to be strengthened and sustain￾able financing sources identified. 
While all agreed that action was needed, appropriate
solutions were blocked. This was because the gov￾ernment lacked an appropriate policy and legislative
framework and had weak enforcement capacity and
an inadequate budget. To help create the enabling
environment, in 1995 USAID launched NATURE.
It was designed to help the government create, in
just two years, a policy and legislative environment
that promoted conservation in cooperation with
local communities. Then, over the next three years,
USAID would concentrate on training, technical
assistance, and other efforts to help with policy
implementation. 
Reforms were designed to unify policies and 
legislation affecting forestry, fisheries, water and
irrigation, soil conservation, land administration,
and national parks and forests. This meant working
with nine separate government agencies. USAID
did not fully anticipate the bureaucratic and legisla￾tive actions that were required. What was to have
been a five-year program stretched out to nearly
nine years. USAID made the final disbursement 
of funds in September 2003, and the final 
conditions were satisfied in early 2004. USAID’s 
COMPASS-II project will work with the govern￾ment to implement the new policies. 
NATURE was a policy reform program. Once 
policy conditionality was met, USAID disbursed
funds. There was no legal requirement for the 
government to provide the funding to implement
the policy reforms. But it was important to make
sure that policies were actually implemented.
USAID used the time leading up to the final 
disbursement of NATURE funds as an opportunity
to discuss development issues with the government
and to press for action on implementation problems. 
As a result of discussions, the government agreed
on a number of measures to implement the
What Conditions Favor the Success of General Budget Support and Sector Program Assistance? 5
NATURE policy reforms. For example, it put in
place a policy to decentralize its programs to the
district level and to fund the decentralized pro￾grams. The government also created the Malawi
Environmental Endowment Trust (MEET) and, at
USAID’s urging, provided the needed funds. 
Inappropriate Budget
Management and
Corruption
Malawi’s currency, the kwacha, has steadily
lost value against the dollar due to slow
export growth and inflationary pres￾sures. In 1994, the exchange rate was 8.7 to the
dollar, in 2001 it was 72, and in March 2004 it
was 114. Consumer prices rose 27 percent in 2001
and 15 percent in 2002. The real interest rate was
24 percent in 2001 and near 20 percent in 2002.
The Government of Malawi has recently been in a
cycle of revenue shortfalls, overspending and then
filling the gap with domestic borrowing. As a
result, Malawi repeatedly missed IMF fiscal and
monetary targets. From 2001 through most of
2004, Malawi met its IMF targets only once. The
government’s difficulty in staying on track with its
IMF program caused GBS donors to hold back on
aid disbursements—reducing the funds available
for development programs. As a result, the govern￾ment sharply increased its domestic borrowing,
driving up interest rates, reducing capital available
for private sector investment, and making it harder
to meet IMF targets. 
There are serious weaknesses throughout the gov￾ernment’s financial management system. Budget
execution is weak; there has been a lack of com￾pleteness, accuracy, and reliability of accounting
and financial reporting; internal control proce￾dures are ineffective; and external oversight mech￾anisms either have capacity constraints or lack
teeth. Notwithstanding efforts to identify and fully
finance pro-poor expenditures, there is a poor
linkage between stated policies and the budget
process. This makes it extremely difficult to meas￾ure results. Significant expenditures, including
those financing parastatal activities, are not includ￾ed in the budget. In general, there is little enforce￾ment of hard budget constraints, as overspending
is often “regularized” into a final budget late in the
fiscal year. 
As in many parts of sub-Saharan Africa, corruption
is a big problem in Malawi. Transparency
International’s Corruption Perceptions Index
reflects perceived levels of corruption. For 2003,
Malawi scored a low 2.8. on a scale of 1 to 10
(with 10 being a system free of corruption). In
2000, Malawi scored 4.1. 
There is anecdotal evidence of an increase in cor￾ruption during the past decade since the end of
autocratic rule and the move to democracy. Petty
corruption is reported at the local level, but it is
nowhere as severe a problem as high-level corrup￾tion. A report by the Accountant General in 2000
provided evidence of fraudulent contracting in the
Ministry of Education and implicated ministers,
members of Parliament, and senior civil servants.
Prosecution was halted due to political interfer￾ence. In September 2004, a senior government
official was ready to speak about high-level corrup￾tion in court, but “disappeared” before he could
testify. He was later found dead. Some feel that, in
relative terms, the problem of corruption takes sec￾ond place to the mismanagement of Malawi’s
macroeconomic environment. Others say the two
go hand-in-hand.
6 Evaluation Brief No. 12
A. If the problem is policies, concentrate on getting policy
reforms adopted. Since the reforms will solve the prob￾lems, it is not as important how aid funds are spent.
B. If the problem is implementation, provide aid to help
fund the government's budget, but ensure that the
funding implements the policy goals and that the
funds can be tracked to their final use.
C. If the problem is policies and implementation, mix policy
conditionality with funding, making sure that the
funds are spent to support the policies.
The “ABCs” of the Three Donor Approaches
Lessons Learned
This report summarizes the findings from the
analysis in the working paper General Budget
Support and Sector Assistance: Malawi
Country Case Study.3 The analysis in that paper devel￾oped nine lessons related to macroeconomic and sec￾tor policy reforms, budget discipline and sound
financial systems, and the need for the government
to manage and own the development process.
Macroeconomic and Sector Policy
Reforms
Policy reforms are of little use if there is no
will to execute them.
Malawi has sound strategies to develop the country
and qualified personnel to carry out the plans.
Donors are ready to help finance the government’s
efforts to reduce poverty. To get the country’s
macroeconomic situation back on track, the gov￾ernment needs to follow its plans and not constant￾ly change priorities. To reduce poverty, Malawi
needs to take the execution of its PRSP seriously
and closely monitor its progress. 
Cash transfers such as GBS will not have
their expected impact if the government
lacks budget discipline and fails to reform its
macroeconomic policies.
GBS donors are interested in ensuring that the gov￾ernment’s pro-poor expenditures increase and that
their assistance supports improved service delivery.
But if GBS is to be successful, macroeconomic con￾ditions must be sound. In Malawi, monetary and
fiscal policies were inappropriate for most of the
time from 2000 through 2004. Government
domestic debt, interest rates, inflation, and public
expenditures increased at an unsustainable pace.
This discouraged investment and development.
If government policies are not reflected in
annual budgets, donor assistance to support
budget initiatives is not appropriate.
Malawi has a PRSP that was developed in a consul￾tative fashion with donors, civil society, and NGOs.
The PRSP provides a good development policy
framework. However, it is not linked to the annual
budget. This disconnect between policy and budget
makes it hard to see whether development priorities
are being implemented. GBS donors have a diffi￾cult time in Malawi: multiyear strategies and poli￾cies are of little value, as they are not linked to
annual budget expenditures. 
Higher-level reforms are of little value by
themselves. To maintain momentum, it is
important to follow up on reforms with
implementation measures that ensure bene￾fits are realized.
For example, USAID sector program assistance
helped create policies and new organizational rela￾tionships that should improve Malawi’s natural
resource management. Policy reform was a success.
But there were implementation delays. In the field,
in Lengwe National Park and Mwabvi Wildlife
Reserve, government officials and local village lead￾ers were concerned that—after much preparation
for a new comanagement approach to environmen￾tal conservation—nothing on the ground had
changed. To deal with such problems, USAID used
policy dialogue to urge the government to put
funds in place to implement the reforms. 
Budget Discipline and Sound
Financial Systems
If donor funds are to support a country’s
annual development budget, donors need to
be confident in the budget process.
In Malawi a budget is developed each year, but it is
rarely reflected in actual expenditures. Faced with
annual revenue shortages, cuts and reallocations are
made throughout the year, though the government
does protect pro-poor expenditures. Budget trans￾parency, execution, and accounting are extremely
weak, so it is difficult to track expenditures. Over
the years, many donor-supported financial and
accounting systems have been launched, with few
results. With government plans to decentralize
operations to the 27 districts, it will be even harder
to track financial flows and results. 
What Conditions Favor the Success of General Budget Support and Sector Program Assistance? 7
2
3
1
4
3 PN-AAA-XXX, available online at www.dec.org.
5
The government’s financial and accounting sys￾tems must assure that funds are spent for their
intended purposes. Governance reforms can
improve service delivery and accountability. 
In Malawi budget execution and accounting are
extremely weak. The problems are due to inade￾quate financial systems, inappropriately trained
staff, and a lack of political and bureaucratic inter￾est in accountability. In 1996, the World Bank sup￾ported the introduction of an Integrated Financial
Management Information System, with limited
success. More recently, the Bank has worked with
the government on the introduction of a Country
Financial Accountability Assessment (financial
management), a Financial Accountability Action
Plan (to improve budget execution and tracking),
and a system of audits. 
In Malawi there is no shortage of donor-supported
acronyms for new systems to improve public
finance. What appears to be missing is government
willingness to use such tools to deal with a dysfunc￾tional financial system. A government must be
accountable to its own citizens, not just the donors.
Efforts are needed to encourage domestic oversight
by Parliament, NGOs, media, and civil society
through diagnostics based on public-services users
and tools to track public expenditures. 
Government Needs to Manage and
Own the Development Process
A government’s technical and management
capabilities are critical to policy and institu￾tional reform.
With a per capita GNI of $160, Malawi is very
poor. Almost by definition, low-income countries
lack skilled manpower. Compared to those of other
African countries, Malawi’s senior managers are
well trained and highly skilled. At mid- and lower
levels, skills are generally good. A looming problem
comes from high HIV/AIDS rates (15 percent of
adults), which could be a major threat to govern￾ment capacity. Donor assistance can help the gov￾ernment build the capacity to successfully manage
its own development efforts. In the case of Malawi,
this includes HIV/AIDS programs. 
Policy reform must have government 
ownership.
USAID’s environmental and natural resources 
policy reform program (NATURE) worked with
the government to develop a set of policy and 
institutional changes. It required actions on the
government’s part from several ministries and
departments. Rather than working with all of the
different ministries to get the reforms adopted,
USAID left it to the Department of Environmental
Affairs to take on the role of getting agreement
from all ministries. This clearly added time to the
process, but it was successful and helped the gov￾ernment gain ownership of the reform process. 
Corruption drains resources from develop￾ment; budget support (which is a cash 
transfer) is at high risk, and donors may
want to avoid such situations.
In Malawi, corruption is not at the level of many
other African countries, but it is visible at the 
highest political levels, where it sets a bad example
for government employees and creates cynicism
among the people of Malawi. The biggest danger,
based on experience in other countries, is that 
corruption starts small but tends to increase and
spread throughout the government. Corruption
then creates serious economic distortions and
destroys respect for the government.
8 Evaluation Brief No. 12
7
This brief and other relevant documents can be ordered from USAID’s Development Experience Clearinghouse (DEC). To order or download, go
to www.dec.org and enter the document identification number—PN-ADA-356—in the search box. The DEC may also be reached at 8403
Colesville Road, Suite 210, Silver Spring, MD 20910; telephone 301-562-0641; fax 301-588-7787; email docorder@dec.cdie.org.
Editorial, design, and production assistance was provided by IBI–International Business Initiatives, Arlington, VA, under contract no. HFM-C-00-
01-00143-00. For more information, contact IBI’s Publications and Graphics Support Project at 703-525-2277 or pgsp@ibi-usa.com.
6
8
9