International Monetary Fund: Few Changes Evident in Design of New
Lending Program for Poor Countries (08-MAY-01, GAO-01-581).	 
								 
The effectiveness and appropriateness of the International	 
Monetary Fund's lending programs to poor countries have been	 
widely debated, generally centering on whether the program have  
improved these countries' economies. In response to come of these
concerns, and as part of a concerted international effort to	 
reduce poverty, in September 1999, the Fund expanded the goals of
its lending program to its poorest members to include an explicit
focus on poverty reduction. In November 1999, to underscore this 
focus, the Fund renamed its concessional lending program from the
Enhanced Structural Adjustment Facility to the Poverty Reduction 
and Frowth Facility. GAO found that although the design of the	 
Poverty Reduction and Growth Facility does not differ		 
significantly from the Fund's previous program, certain elements 
of the new program are emphasized more now than in the past. The 
one major design change--getting countries to take ownership of  
their macroeconomic framework--is difficult to achieve for three 
reasons. First, the limited capacity of many recipient		 
governments to independently analyze and effectively negotiate	 
the macroeconomic framework reduces the opportunity for 	 
country-specific elements to be addressed. Second, the challenges
to effectively engaging nongovernmental organizations in a	 
dialogue on these very complex matters are significant. Finally, 
a national dialogue on the choice of effective policies is	 
hampered by the limited knowledge of all parties about how	 
different policies actually effect elements of the macroeconomic 
framework. GAO found few changes in the Fund programs of the	 
three countries reviewed that can be clearly attributed to the	 
changes that the Fund announced in 1999. Nonetheless, the three  
countries faced several difficulties in developing a nationally  
owned macroeconomic framework. First, each government has limited
capacity to independently analyze macroeconomic issues. Second,  
while all three governments have begun a dialogue with		 
nongovernmental organizations, many people that GAO spoke with	 
were unsure how to use this dialogue to address the countries'	 
complex macroeconomic policies and targets, other than the	 
composition and level of spending. Finally, GAO did not see	 
evidence of the changes in the three countries' Fund documents	 
that were called for under the new program. If the changes	 
announced by the Fund for its lending program to its poorest	 
members actually improve the overall effectiveness of a country's
development program, the likelihood of earlier graduation from	 
the Fund's program could increase. However, the impact of these  
changes on economic growth is unknown at this time.		 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-01-581 					        
    ACCNO:   A00892						        
  TITLE:     International Monetary Fund: Few Changes Evident in      
             Design of New Lending Program for Poor Countries                 
     DATE:   05/08/2001 
  SUBJECT:   Developing countries				 
	     Economic growth					 
	     Federal aid to foreign countries			 
	     Foreign loans					 
	     International economic relations			 
	     International organizations			 
	     Macroeconomic analysis				 
	     Albania						 
	     Benin						 
	     Heavily Indebted Poor Countries Debt		 
	     Initiative 					 
								 
	     Honduras						 
	     IMF Enhanced Structural Adjustment 		 
	     Facility Program					 
								 
	     IMF Poverty Reduction and Growth			 
	     Facility Program					 
								 
	     International Monetary Fund			 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Testimony.                                               **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-01-581
     
Report to the Chairman, Committee on Foreign Relations, U. S. Senate

United States General Accounting Office

GAO

May 2001 INTERNATIONAL MONETARY FUND

Few Changes Evident in Design of New Lending Program for Poor Countries

GAO- 01- 581

Page i GAO- 01- 581 IMF: New Lending Program Letter 1

Executive Summary 2

Chapter 1 Introduction 12 Financing of and Eligibility for the IMF?s
Concessional Loans 12 The IMF Focuses on Macroeconomic and Structural Issues
14 The IMF?s Increased Focus on Social Issues 15 Criticisms of the IMF 16
The New Facility Is to Focus More Explicitly on Poverty Reduction 17
Objectives, Scope, and Methodology 21

Chapter 2 The Design of the IMF?s New Program Differs Little From Its
Previous Program 24 Most of the Announced Modifications Represent a Change
in

Emphasis Rather Than a Change in Philosophy 24 Widely Accepted Goal of
Maintaining Macroeconomic Stability

Limits Potential Changes Under the Program 30 Developing a Nationally Owned
Macroeconomic Framework Is

Difficult 35

Chapter 3 Few Changes Evident in Three Countries? Programs 39 Inclusion of
Some Elements Difficult to Attribute to New Program 39 Difficulties
Establishing National Ownership of Macroeconomic

Framework 44

Chapter 4 The Effects of the New Program on Graduation and the Role of the
IMF 47 Strong, Sustained Economic Growth Essential to Reach

Consideration for Graduation 47 The IMF Has Been a Long- term Lender to Low-
income Countries 52

Appendix I Identifying the Impact of Different Policy Choices Is Difficult
55 Contents

Page ii GAO- 01- 581 IMF: New Lending Program Appendix II Key Events and IMF
Programs in Albania, Benin, and

Honduras 59

Appendix III Comments From the Department of the Treasury 66

Appendix IV Comments From the International Monetary Fund 67

Appendix V GAO Contact and Staff Acknowledgments 72

Tables

Table 1: The Fund?s Stated Focus on Poverty Reduction, Government
Leadership, and Civil Society Participation 26 Table 2: Inclusion of PRGF
Elements in Three Countries? IMF

Programs 40 Table 3: Projected and Previous Annual Growth Rates for 32 PRGF

Borrowers 50 Table 4: Number of Years 32 Countries Received IMF Resources 53
Table 5: Key Events in Albania, 1991- 2001 60 Table 6: Key Events in Benin,
1960s- 2001 62 Table 7: Key Events in Honduras, 1992- 2000 64

Figures

Figure 1: Disbursements of IMF Concessional Loans by Region, 1986- 2000 14
Figure 2: Elements of a Macroeconomic Framework in Support of

the PRGF?s Twin Goals of Poverty Reduction and Growth, in a Context of
Macroeconomic Stability 32 Figure 3: Complex Linkages Between Increased
Spending on

Education and Health and Goals 56

Page iii GAO- 01- 581 IMF: New Lending Program

Abbreviations

ESAF Enhanced Structural Adjustment Facility GDP Gross domestic product IDA
International Development Association IMF International Monetary Fund PFP
Policy framework paper PRGF Poverty Reduction and Growth Facility PRSP
Poverty Reduction Strategy Paper

Page 1 GAO- 01- 581 IMF: New Lending Program

May 8, 2001 The Honorable Jesse Helms Chairman, Committee on Foreign
Relations United States Senate

Dear Mr. Chairman: This report responds to your request that we (1) analyze
how the International Monetary Fund?s new Poverty Reduction and Growth
Facility is designed to be different from the Fund?s Enhanced Structural
Adjustment Facility, (2) assess what has actually changed in recipient
countries? programs, and (3) evaluate whether the changes implemented in the
new program increase the likelihood of recipient countries graduating from
(that is, no longer being eligible for) concessional borrowing in less than
15 years or have contributed to a more short- term lending focus for the
Fund.

We are sending copies of the report to the Honorable Paul H. O?Neill,
Secretary of the Treasury; the Honorable Colin L. Powell, Secretary of
State; the Honorable Horst Kï¿½hler, Managing Director of the International
Monetary Fund; the Honorable James D. Wolfensohn, President of the World
Bank; and other interested parties. Copies will also be made available to
others on request.

Please contact me at (202) 512- 4128 if you or your staff have any questions
concerning the report. An additional GAO contact and staff acknowledgments
are listed in appendix V.

Sincerely yours, Harold J. Johnson Director, International Affairs and Trade

United States General Accounting Office Washington, DC 20548

Executive Summary Page 2 GAO- 01- 581 IMF: New Lending Program

The effectiveness and appropriateness of the International Monetary Fund?s
lending programs to poor countries have been widely debated. 1 This debate
has centered on whether Fund programs have improved countries? economies and
whether the potential negative impacts of its programs on the poor have been
sufficiently considered. Some have also argued that the Fund should return
to its initial mandate of lending on a short- term basis to countries
experiencing crises and end its support of the longer- term reform programs
of low- income countries.

In response to some of these concerns, and as part of a concerted
international effort to reduce poverty, in September 1999 the Fund expanded
the goals of its lending program to its poorest members to include an
explicit focus on poverty reduction. In November 1999, to underscore this
focus, the Fund renamed its concessional lending program from the Enhanced
Structural Adjustment Facility to the Poverty Reduction and Growth Facility.
In December 1999, the Fund?s Executive Board asked Fund staff to begin
implementing the process quickly, recognizing that it would involve a
substantial degree of experimentation and innovation. 2

Under this new focus, the economic and structural measures in Fundsupported
programs are to emerge directly from each country?s own priorities for
reducing poverty, determined through a government- led, broad participatory
process that includes representatives of civil society (including the poor),
3 parliament, and donors, based on each country?s unique circumstances. In
this way, policies aimed at poverty reduction could have a direct impact on
the design of the macroeconomic

1 The International Monetary Fund, supported by its 183 member governments,
promotes international monetary cooperation and exchange rate stability and
provides lending to member countries that experience balance- of- payments
difficulties. The Fund lends money to low- income members on concessional,
or below- market, terms in order to support programs to strengthen their
balance- of- payments positions and to foster growth. A country?s balance-
of- payments accounts summarize its financial dealings with the outside
world.

2 The Fund operates under the authority of the Board of Governors, the
highest decisionmaking authority. General operations are delegated to a
smaller group of representatives, the Board of Executive Directors, who are
responsible for making policy decisions and approving loans. The Board
comprises 24 Executive Directors who are appointed or elected by one or more
member countries.

3 ?Civil society? refers to the nongovernmental segment of society and
includes churches, community groups, trade unions, business associations,
and organizations that advocate for specific causes, such as human rights
and environmental protection. Executive Summary

Purpose

Executive Summary Page 3 GAO- 01- 581 IMF: New Lending Program

framework (that is, economic policies, targets, and structural reforms) that
seeks to promote faster, sustainable economic growth.

The Chairman of the Senate Committee on Foreign Relations asked GAO to
evaluate what the Fund has changed about its lending program for its poorest
members since it announced the Poverty Reduction and Growth Facility;
specifically to (1) analyze how this new program is designed to be different
from the previous program, (2) assess what has actually changed in recipient
countries? programs, and (3) evaluate whether the changes implemented in the
new program increase the likelihood of recipient countries graduating from
(that is, no longer being eligible for) concessional borrowing in less than
15 years or have contributed to a more short- term lending focus for the
Fund.

To address all three issues, GAO analyzed a wide range of documents from the
Fund, the World Bank, 4 recipient and donor governments, U. N.
organizations, and nongovernmental organizations. GAO also interviewed
officials from these organizations in the United States and abroad. To
analyze whether the design of the new program differs from the old, GAO
examined Fund statements describing the key elements of the new program and
compared them to Fund statements and analyses regarding the Fund?s previous
program. GAO?s findings regarding program design derive from this analysis
and are not based on individual country experiences. To address the second
issue, GAO selected and visited three countries for case studies- Albania,
Benin, and Honduras- that met the following criteria as of September 30,
2000: they had a current Fund program and relatively consistent performance
under this program, were expected to have full poverty reduction strategies
by mid- 2001, and had been reviewed by the Fund under their current and
previous Fund programs. GAO did not generalize from the experience of these
countries to draw conclusions about overall implementation issues. Instead,
the case studies are examples of how countries are addressing the design
challenges identified in the first objective.

Since 1976, the Fund has provided loans on concessional terms 5 to eligible
low- income members in order to strengthen their balance- of- payments

4 The World Bank, supported by its 182 member governments, promotes economic
growth and the development of market economies by providing financing on
reasonable terms to countries that have difficulty obtaining capital.

5 The lending terms include a 5-ï¿½ year grace period, 10- year maturity, and
annual interest rate of 0.5 percent. Background

Executive Summary Page 4 GAO- 01- 581 IMF: New Lending Program

positions and support their reform programs. In order to receive loans,
countries agree to implement macroeconomic and structural reforms. 6 These
loans have been funded primarily with the profits from the sale of some of
the Fund?s gold holdings and contributions (loans or grants) from member
countries. 7 Eligibility for these concessional loans has been based mainly
on a country?s per capita income and eligibility for World Bank concessional
lending. The terms of the loans have remained the same, while the purpose of
the loans has expanded from balance- of- payments assistance to fostering
lasting growth that leads to higher living standards and a reduction in
poverty.

The Fund provides advice on macroeconomic issues such as achieving and
maintaining stability. 8 At the same time, the Fund seeks to integrate
social policies into its programs and advice, with the World Bank taking the
lead on these issues. As of February 21, 2001, 34 of the 77 countries
eligible for the Poverty Reduction and Growth Facility had current loan
commitments totaling more than $4 billion. Twenty- two of the 34 countries
are in subSaharan Africa.

Although the design of the International Monetary Fund?s Poverty Reduction
and Growth Facility does not differ significantly from the Fund?s previous
program, certain elements of the new program are emphasized more now than in
the past. GAO observed that most of the changes announced for the new
program were also pursued under the previous program. The one major design
change- getting countries to take ownership of their macroeconomic
framework- is difficult to achieve for three reasons. First, the limited
capacity of many recipient governments to

6 Programs cover a period of 3 years but can be extended to a fourth year.
Loans are disbursed either quarterly or semi- annually. The Fund makes the
first disbursement after it approves the arrangement. The Fund releases the
next disbursements after it completes reviews of the program, determining
that the country has satisfactorily met its performance requirements. These
requirements generally include macroeconomic indicators and important
structural measures.

7 These resources are separate from the Fund?s other resources such as the
quota contributions of its members. 8 Macroeconomic stability exists when
key economic relationships are in balance, for example, between domestic
demand and output, the balance of payments, fiscal revenues and expenditure,
and savings and investment. These relationships need not necessarily be in
exact balance. Imbalances such as fiscal and current account deficits or
surpluses are perfectly compatible with economic stability provided that
they can be financed in a sustainable manner. Results in Brief

Executive Summary Page 5 GAO- 01- 581 IMF: New Lending Program

independently analyze and effectively negotiate the macroeconomic framework
reduces the opportunity for country- specific elements to be addressed.
Second, the challenges to effectively engaging civil society in a dialogue
on these very complex matters are significant. Finally, a national dialogue
on the choice of effective policies is hampered by the limited knowledge of
all parties about how different policies actually affect elements of the
macroeconomic framework. Considering the above difficulties, civil society
may not be able to influence the macroeconomic framework through the initial
poverty reduction strategy; however, civil society may help improve the
allocation of resources and increase the amount of resources donors are
willing to provide by helping establish priorities for poverty reduction.
Even if national ownership increases, given the need for poor countries to
maintain macroeconomic stability, which is essential for economic growth and
poverty reduction, the actual policies and targets within the macroeconomic
framework are not likely to be altered substantially from the past.

GAO found few changes in the Fund programs of the three countries reviewed-
Albania, Benin, and Honduras- that can be clearly attributed to the changes
that the Fund announced in 1999. Most elements of the Fund?s new program
were also included in countries? previous programs, making it difficult to
determine to what extent these elements reflect the announced changes in the
Fund?s approach rather than the direction the country?s program was moving
in prior to the new program. Nonetheless, the three countries faced several
difficulties in developing a nationally owned macroeconomic framework.
First, each government has limited capacity to independently analyze
macroeconomic issues. Second, while all three governments have begun a
dialogue with civil society, many people that GAO spoke with were unsure how
to use this dialogue to address the countries? complex macroeconomic
policies and targets, other than the composition and level of spending.
Finally, GAO did not see evidence of changes in the three countries? Fund
documents that were called for under the new program. These changes include
showing that before reforms were implemented, their social impacts were
assessed and measures to alleviate any negative impacts were put in place.

If the changes announced by the Fund for its lending program to its poorest
members actually improve the overall effectiveness of a country?s
development program, the likelihood of earlier graduation from the Fund?s
program could increase. However, the impact of these changes on economic
growth is unknown at this time. GAO?s analysis shows that most of the
current recipients of the Fund?s concessional assistance will require
strong, sustained economic growth to reach the point of graduation from

Executive Summary Page 6 GAO- 01- 581 IMF: New Lending Program

concessional Fund assistance. To reach this point within 15 years, the 32
countries that borrowed from the Fund?s concessional facility in 2000 must
average a real per capita income growth in excess of 6 percent annually
during that entire period, a growth rate that significantly exceeds the
countries? average growth rate of negative 1 percent over the last 15 years.
Unlike the Fund?s nonconcessional facilities that are to lend resources
based on short- term conditions, the concessional facility does not assume
that countries will have only a temporary need for assistance. During the
past 15 years, the Fund has functioned as a long- term lender that supports
countries? reform programs, and indications are that it intends to retain
this role.

The Treasury said that GAO?s report addresses a number of important issues
concerning the Fund?s concessional lending to poor countries. The Fund did
not disagree with GAO?s findings regarding the design of the Poverty
Reduction and Growth Facility but stated that GAO?s report could have been
more positive in its judgments.

The design of the Poverty Reduction and Growth Facility does not differ
significantly from the Enhanced Structural Adjustment Facility because the
new program includes elements that have been pursued under the previous
program for a number of years. One change in the new program- that the
macroeconomic targets and policies in Fund programs will emerge from a
government- led process involving civil society and donors- could be a major
departure from how the macroeconomic framework has been traditionally
chosen. However, even this builds on features that the Fund has discussed
and pursued for some time.

This commonality does not necessarily mean that the new program is identical
to the previous program, since the new program envisions consolidating the
program elements into a single framework and giving Principal Findings

The Design of the Fund?s New Program Differs Little From Its Previous
Program

Most of the Announced Modifications Represent a Change in Emphasis Rather
Than a Change in Philosophy

Executive Summary Page 7 GAO- 01- 581 IMF: New Lending Program

them greater prominence. GAO believes that such a change represents a shift
in emphasis rather than a change in the Fund?s stated philosophy, with signs
of this shift evident over the past few years. Yet the Fund has had
difficulties achieving some of these program elements. For example, since at
least 1987, ownership by the recipient governments of their Fund programs
has been seen as an important element in successful program implementation.
However, the main vehicle to achieve this- the policy framework papers- has
been generally developed by the staff of the Fund and not by the countries
themselves. Under the new program, the governments are to write the new
poverty reduction strategies themselves. According to Fund staff, the Fund?s
Executive Board could endorse strategies that contain elements that the Fund
does not agree with. However, the significance of this change is uncertain,
since the Board must endorse a country?s overall strategy in order for the
country to borrow from the Fund.

The Fund and the World Bank consider macroeconomic stability to be a
necessary prerequisite for economic growth and poverty reduction, although
not sufficient on its own to achieve those goals. Countries that experience
macroeconomic instability, such as high inflation rates, have tended to have
low or even negative economic growth rates. This concern over the negative
effects of macroeconomic instability underlies the Fund?s continuing goal
that a country?s macroeconomic framework should work to maintain stability,
once achieved. However, policies that are overly concerned with
macroeconomic stability may turn out to be too austere, lowering economic
growth from its optimal level and impeding progress on poverty reduction.
According to Fund and World Bank documents, there is a ?substantial gray
area? between those policies that may be considered too austere and those
that cause macroeconomic instability. Presumably, one goal of including the
macroeconomic framework within the national poverty reduction dialogue would
be to explore this gray area to establish an effective mix of policies
consistent with the medium- term goals of the country.

Based on GAO?s analysis of numerous documents and discussions with Fund and
World Bank officials, it is difficult to determine whether in fact there is
a ?substantial? range of macroeconomic policy targets to be discussed and
explored within this so- called ?gray area.? This is due to two factors.
First, precise identification of the bounds of the gray area is beyond the
current understanding of the economics profession. For example, many
economists, including some at the Fund, think that inflation above a 7 to 11
percent range is risky, whereas others think the level can be between 20 and
40 percent before it starts to endanger Widely Accepted Goal of

Maintaining Macroeconomic Stability Limits Potential Changes Under the
Program

Executive Summary Page 8 GAO- 01- 581 IMF: New Lending Program

economic growth. Second, the harsh economic realities confronting these very
poor countries also work to limit the choice of policies or the amount that
spending can be prudently increased over a short period of time. Both of
these factors are strongly influenced by the desire to ensure that whatever
macroeconomic framework is agreed to, it does not put the country at greater
risk of macroeconomic instability.

Governments face several challenges in developing a macroeconomic framework
for which they are expected to take ownership. First, many governments have
limited technical capacity relative to the substantial complexities inherent
in establishing macroeconomic policies and targets. For example, governments
are expected to more openly discuss the selection, impacts, and trade- offs
of various macroeconomic policies (i. e., lower budget deficits versus
higher spending on poverty reduction). Yet, there are questions about the
extent to which many governments are capable of effectively engaging in such
discussions. Second, the challenges in establishing a participatory process
and the complexities of macroeconomic issues may limit the extent to which
civil society can influence the macroeconomic framework in the near term.
This can be especially difficult in countries that lack a democratic or
representative tradition and thus have few existing means for getting
citizen or nongovernmental organizations? input or for electing
representatives. Finally, even if the capacity of the national governments
were improved and civil society were effectively engaged in a dialogue on
the macroeconomic framework, national ownership would be hampered by the
current limitations in economic knowledge on how different policies actually
affect elements of the macroeconomic framework. The World Bank and others
are attempting to develop models that may help explain the impact of various
policies, but the process is slow due to technical complexities and limited
country- specific data. Considering the above difficulties, civil society
may not be able to influence the macroeconomic framework through the initial
poverty reduction strategy; however, civil society may help improve the
allocation of resources and increase the amount of resources donors are
willing to provide by helping establish priorities for poverty reduction.
Developing a Nationally Owned

Macroeconomic Framework Is Difficult

Executive Summary Page 9 GAO- 01- 581 IMF: New Lending Program

GAO found a few changes in the Fund programs for Albania, Benin, and
Honduras that can be clearly attributed to the changes announced by the Fund
in 1999. These few changes are (1) Albania?s establishment of a
participatory approach for preparing a national development strategy and (2)
fewer structural conditions in Benin. However, the significance of these
changes for the Fund program is not clear, since Albania?s participatory
process does not describe how Albania will include civil society in
discussing the macroeconomic framework, and Benin still has to meet two of
the three conditions dropped from the Fund program in order to receive debt
relief under the Heavily Indebted Poor Countries Initiative. 9 The absence
of clear change is due, in part, to the inclusion of many of the same
elements in countries? previous Fund programs. For example, Albania and
Benin have focused on poverty reduction since at least 1997.

For all three countries, establishing national ownership of their
macroeconomic policies and targets has been challenging for several reasons.
First, the three governments? ability to analyze macroeconomic issues is
limited. For example, according to a government official in Albania, since
the departure of an employee of the central bank in 1997, no one has been
able to operate and maintain their macroeconomic model. Second, while all
three national governments have organized a participatory process for
preparing their countries? poverty reduction strategies, they have found it
difficult to determine how to include civil society in discussing
macroeconomic policies and targets. For example, in Honduras, government and
donor officials told GAO that there were no civil society organizations
conducting analyses of macroeconomic issues. Finally, while countries? Fund
program documents are to include information on the social impact of
significant economic adjustments before adjustments are implemented, GAO did
not find such information.

9 The Heavily Indebted Poor Countries Initiative is a comprehensive approach
for providing debt relief to the poorest and most indebted countries in the
world. For more information on this initiative, see Developing Countries:
Debt Relief Initiative for Poor Countries Faces Challenges (GAO/ NSIAD- 00-
161, June 29, 2000). Few Changes Evident

in Three Countries? Programs

Inclusion of Some Elements Difficult to Attribute to New Program

Difficulties Establishing National Ownership of Macroeconomic Framework

Executive Summary Page 10 GAO- 01- 581 IMF: New Lending Program

If the changes announced by the Fund for its lending program to its poorest
members, particularly national ownership of the macroeconomic framework,
improve the overall effectiveness of a country?s development program, the
likelihood of earlier graduation from the Fund?s program could increase.
However, the actual impact of these changes on economic growth is unknown at
this time and, as previously discussed, there are many challenges and
obstacles to establishing national ownership.

GAO?s analysis showed that most of the 32 countries that borrowed
concessional resources from the Fund in 2000 will need to achieve strong,
sustained economic growth to reach eligibility for graduation within the
next 15 years. These countries would have to achieve an annual average
growth rate of 6 percent, which is substantially greater than the negative 1
percent growth rate the countries averaged over the previous 15 years. These
countries are projected to have an average of 2.5 percent growth over the
next 15 years. At those projected growth rates, only four additional
countries would reach the eligibility threshold by the end of the 15- year
period. Given that the current annual per capita income levels of these
countries is generally quite low (an average of $427), the average number of
years required for these countries to reach the graduation threshold at
their projected per- capita income growth rates is 59, with Niger requiring
366 years.

Over the last 15 years, the Fund has functioned as a long- term lender to
support poor countries? reform programs, and it has indicated that it
intends to retain this role. The underlying problems that Fund assistance is
designed to help address in poor countries often take a long time to
resolve. Unlike the Fund?s nonconcessional resources, there is no
presumption under the Poverty Reduction and Growth Facility that countries
will have only a temporary need for assistance. The Fund has wide latitude
to lend to poor countries, given the countries? underlying vulnerabilities
and low standards of living. For example, in recent years Benin has had a
very stable macroeconomic environment, with low inflation, a balance- of-
payments surplus, low budget deficits, and good economic growth. Within this
healthy macroeconomic context, in July The Effects of the

New Program on Graduation and the Role of the Fund

Strong, Sustained Economic Growth Essential to Reach Consideration for
Graduation

The Fund Has Been a Longterm Lender to Low- income Countries

Executive Summary Page 11 GAO- 01- 581 IMF: New Lending Program

2000 the Fund and Benin negotiated another program for an additional 3
years, with the objectives of achieving high and sustainable economic growth
and reducing poverty while maintaining macroeconomic stability. The 32
countries that received Fund assistance in 2000 had received assistance from
the Fund for an average of 7 of the last 10 years. According to a Fund
official, the Fund is likely to continue to provide lowincome countries with
concessional resources as long as they have a need for those resources, the
resources are available to be lent, and the countries pursue effective
policies.

GAO received written responses on this report from the Department of the
Treasury and the International Monetary Fund. The Treasury said that GAO?s
report addresses a number of important issues concerning the Fund?s
concessional lending to poor countries. Comments received and GAO?s
evaluation of them are reprinted in appendixes III and IV. These
organizations, along with the World Bank, also separately provided technical
comments that GAO discussed with relevant officials and included in the text
of the report, where appropriate.

The Fund did not disagree with GAO?s findings regarding the design of the
Poverty Reduction and Growth Facility but stated that GAO?s report could
have been more positive in its judgments. The Fund said that GAO?s report
discounts the process of public involvement in the formulation of key
macroeconomic policy choices. GAO disagrees with this characterization of
the report. The report recognizes the significance of the participatory
process and, in fact, points out that civil society may help improve the
allocation of resources and increase the amount of resources donors are
willing to provide by helping establish priorities for poverty reduction.
The Fund stated that further GAO investigation would have identified actual
changes in Poverty Reduction and Growth Facilitysupported programs and that
the Fund knows change is taking place in many of these programs. However,
the Fund provided no evidence to support this assertion nor did it provide
any examples of actual changes or identify countries where change is taking
place. The Fund said that the GAO analysis could be more tightly focused on
whether outcomes are likely to be better under the new program than under
the previous program. GAO reports that the impact of the new program on
countries? economic growth rates is unknown at this time and that there are
many challenges and obstacles to establishing national ownership. Agency
Comments

and GAO?s Evaluation

Chapter 1: Introduction Page 12 GAO- 01- 581 IMF: New Lending Program

Since 1976, the International Monetary Fund (IMF) has provided loans on
concessional (below- market) terms to eligible low- income members of the
Fund in order to strengthen their balance- of- payments positions and
support their reform programs. In order to receive loans, countries agree to
implement reform programs to address their underlying problems. The IMF has
primarily focused on macroeconomic and structural issues, such as those
intended to promote a market- based economy. In recent years, the role of
the IMF in providing long- term assistance to poor countries as well as the
appropriateness and effectiveness of its programs have been the subject of
considerable debate. In response to some of these criticisms, and as part of
a concerted international effort to reduce poverty, in September 1999 the
IMF expanded the goals of its lending program to its poorest members to
include an explicit focus on poverty reduction. In December 1999, the IMF?s
Executive Board asked IMF staff to begin implementing the revised program
quickly, recognizing that it would involve a substantial degree of
experimentation and innovation. 1 The IMF said in September 2000 that it
expected to see measurable progress in incorporating the new features of the
concessional lending program by September 2001.

The IMF has provided concessional financing through four facilities since
1976. 2 The terms of the IMF?s loans have remained the same over this
period, while the purpose of the loans has expanded from providing balance-
of- payments assistance to now include strengthening a country?s balance of
payments and fostering durable growth, leading to poverty reduction. The
IMF?s concessional lending has been financed primarily with the profits from
the sale of some of the IMF?s gold holdings and contributions (loans or
grants) from member countries. These resources

1 The IMF operates under the authority of the Board of Governors, the
highest decisionmaking authority. General operations are delegated to a
smaller group of representatives, the Board of Executive Directors, who are
responsible for making policy decisions and approving loans. The Board
comprises 24 Executive Directors who are appointed or elected by one or more
member countries.

2 These facilities are the Trust Fund, which operated between 1976 and 1981;
the Structural Adjustment Facility, which operated between 1986 and 1994;
the Enhanced Structural Adjustment Facility (ESAF), which operated between
1987 and 1999; and the Poverty Reduction and Growth Facility (PRGF), which
was established in 1999. The legal authority for the IMF to act as an
administrator of such resources derives from its Articles of Agreement,
Article V, Section 2( b), which empowers it, if requested, to ?perform
financial and technical services, including the administration of resources
contributed by members, that are consistent with the purposes of the Fund.?
Chapter 1: Introduction

Financing of and Eligibility for the IMF?s Concessional Loans

Chapter 1: Introduction Page 13 GAO- 01- 581 IMF: New Lending Program

are separate from the IMF?s other resources such as the quota contributions
of its members. 3 Eligibility for these concessional loans has been based
mainly on a country?s per capita income and eligibility to borrow from the
World Bank?s concessional lending window, the International Development
Association, but it can also take into account other considerations such as
a country?s balance- of- payments and external debt situation. 4 The amount
lent under the IMF?s concessional facility is based on the member?s balance-
of- payments needs, the strength of its adjustment program, its outstanding
use of IMF credit, and its record of such use in the past. The initial
access limit under a 3- year arrangement is 140 percent of the member?s
quota, or, under exceptional circumstances, up to 185 percent of quota, as
of February 21, 2001.

Countries eligible for these concessional loans are among the poorest in the
world, with many classified by the United Nations as being in its lowest
category of human development, based on life expectancy, literacy, and
annual per capita income. Many depend on development assistance from
governments, multilateral organizations, and nongovernmental organizations
and have significant development needs.

Since 1986, a majority of the IMF?s concessional resources have been
disbursed to Africa. Over the latter part of this period, Asia reduced its
use of IMF concessional resources, while Europe increased its use as part of
the transition from centrally planned economies in the early 1990s. (See
fig. 1.)

3 The characterization of the arrangements involving the IMF?s concessional
resources differs from those involving its nonconcessional resources.
Whereas the IMF?s concessional resources are considered loans, use of its
nonconcessional resources are considered ?purchases,? with repayments
considered ?repurchases.? Nonconcessional resources are to address short-
term balance- of- payments problems, such as those that occurred in Asia in
1997. See International Monetary Fund: Observations on the IMF?s Financial
Operations (GAO/ NSIAD/ AIMD- 99- 252, Sept. 30, 1999) for a description of
the

nature of the IMF?s nonconcessional resources, and International Monetary
Fund: Approach Used to Establish and Monitor Conditions for Financial
Assistance (GAO/ GGD/ NSIAD- 99- 168, June 22, 1999) for a description of
the IMF?s lending programs. 4 This threshold was based on a 1998 per capita
income of $895 or less.

Chapter 1: Introduction Page 14 GAO- 01- 581 IMF: New Lending Program

Figure 1: Disbursements of IMF Concessional Loans by Region, 1986- 2000

Note: The spike in disbursements in 1995 was primarily due to a significant
assistance package worth $1.3 billion provided to Zambia. This package was
part of an effort by Zambia to clear its arrears to the IMF.

Source: GAO analysis of IMF data. Amounts adjusted to constant 2000 U. S.
dollars.

As of February 21, 2001, 34 of the 77 countries eligible for the Poverty
Reduction and Growth Facility had current loan commitments totaling more
than $4 billion. Twenty- two of the 34 countries are in sub- Saharan Africa.

In its lending programs to low- income countries, the IMF primarily focuses
on macroeconomic issues, such as promoting international monetary
cooperation, a balanced growth of international trade, and a stable system
of exchange rates; and structural measures such as financial sector reform.
When countries borrow from the IMF, they agree to implement The IMF Focuses
on

Macroeconomic and Structural Issues

0.0 0.2

0.4 0.6

0.8 1.0

1.2 1.4

1.6 1.8

2.0 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
2000

Dollars (in billions)

Europe Africa

Asia Western Hemisphere

Regions

Middle East A

Chapter 1: Introduction Page 15 GAO- 01- 581 IMF: New Lending Program

reform programs as a condition for access to financial assistance. 5 This

?conditionality? aims to alleviate the underlying difficulties that led to
the countries? economic problems and ensure repayment to the IMF. Not all of
a country?s policies or reform efforts are necessarily included as
conditions of the IMF arrangement. The actions the IMF judges to be
particularly important for achieving program objectives will become (1)
?performance criteria? that generally must be met for members to qualify for
disbursements or (2) ?benchmarks? that help to measure progress in meeting
objectives but do not generally affect disbursements.

The IMF has focused on advising countries on how to achieve and maintain
macroeconomic stability, which is widely viewed as a precondition for, but
not a guarantee of, economic growth and poverty reduction. According to IMF
and World Bank staff, macroeconomic stability exists when key economic
relationships are in balance, for example, between domestic demand and
output, the balance of payments, and fiscal revenues and expenditure. To
meet the goal of achieving or maintaining macroeconomic stability, IMF
programs may target factors such as the rate of inflation, debt
sustainability, and the level of international reserves.

The IMF?s primary mandate is to focus on macroeconomic issues; however,
according to IMF staff papers, the IMF has become more involved in areas
where the World Bank has the lead role. For example, the IMF said it has
worked to ensure that social policies are integrated into IMF- supported
programs and advice, thus fulfilling its primary mandate while also
contributing to sustainable economic and human development. In this way, the
IMF?s contribution to social development is mainly indirect, and its role in
social policy advice is limited. Nevertheless, the IMF?s involvement in
social issues has increased. During the 1950s and 1960s, when the IMF
provided financial assistance primarily to industrialized countries, its
policy advice focused mainly on macroeconomic policies. With the shift to
lending to developing countries since the 1970s and economies in transition
since the late 1980s, greater

5 Programs cover a period of 3 years but can be extended to a fourth year.
Loans are disbursed either quarterly or semi- annually. The IMF makes the
first disbursement when its Executive Board- its primary decision- making
body- approves the arrangement. The IMF releases the next disbursements
after it completes reviews of the program, determining that the country has
satisfactorily met its performance requirements. These requirements
generally include macroeconomic indicators and important structural
measures. The IMF?s Increased

Focus on Social Issues

Chapter 1: Introduction Page 16 GAO- 01- 581 IMF: New Lending Program

consideration has been given to the complementarity of macroeconomic
policies and structural reforms and to the formulation of policies in a
medium- term context. With this broadening focus, the link between economic
and social policies has also increasingly been recognized.

The role of the IMF in providing long- term concessional lending and the
appropriateness and effectiveness of its concessional loan programs have
been the subject of considerable debate. For example, some critics believe
the IMF has overstepped its original mission by including conditions related
to social development strategies. In response, the IMF says its increasing
emphasis on structural issues has reflected a growing understanding that
countries? balance- of- payments problems cannot be resolved if a country
suffers from deep- seated structural weaknesses. Critics also contend that
IMF programs impose undue hardships on the poor. They point out that IMF
programs often require that governments cut expenditures and reduce budget
deficits in order to meet the IMF?s macroeconomic goals. They argue that
such cuts often result in reductions in spending on health, education, and
other social programs vital to the poor.

The IMF has acknowledged that, in certain cases in the past, programs for
the poor have been excessively reduced. To lessen this potential, the IMF
said that it now pays more attention to social issues and to social safety
nets and, if necessary, requires that countries maintain minimum spending
levels for social programs despite the need for a general reduction in
government spending. In 1998, the IMF- commissioned external evaluators of
ESAF recommended that IMF programs for low- income countries be better
integrated with policies to fight poverty and that recipient countries
should take more ownership of their program. 6 In response, the IMF began to
implement these changes; this policy has continued under the new PRGF.

6 Kwesi Botchwey and others, Report of the Group of Independent Persons
Appointed to Conduct an Evaluation of Certain Aspects of the Enhanced
Structural Adjustment Facility (IMF: Washington, D. C., Jan. 13, 1998).
Criticisms of the IMF

Chapter 1: Introduction Page 17 GAO- 01- 581 IMF: New Lending Program

In response to some of the above criticisms, and as part of a concerted
international effort to reduce poverty, in September 1999, the IMF expanded
the goals of its lending program to its poorest members to include an
explicit focus on poverty reduction. Both the ESAF and PRGF highlight sound
macroeconomics and rapid, sustainable economic growth- based on robust
private sector activity and investments- as critical to reducing poverty.
Under the PRGF, the IMF will continue to advise on and support policies to
this end, including prudent macroeconomic management, freer and more open
markets, and a stable and predictable environment for private sector
activity. According to IMF documents, the new approach also recognizes the
increasing evidence that entrenched poverty and severe inequality in
economic opportunities and assets can themselves be impediments to growth.
According to these documents, growth- oriented policies should be
implemented in a framework in which the pressing need to reduce poverty is
also a central objective. IMF staff documents highlight the following as the
key features of the PRGF. 7

1. The IMF program is to emerge from the country?s own strategy for growth
and poverty reduction.

 The broadest and most fundamental changes to the IMF?s work arise from the
fact that the targets and policies embodied in IMF- supported programs will
emerge directly from the country?s own poverty reduction strategy paper
(PRSP). The country and its people will need to take the lead. The
government will prepare the poverty reduction strategy based on a process
involving the active participation of civil

7 See IMF, Poverty Reduction Strategy Papers- Operational Issues (IMF:
Washington, D. C., Dec. 13, 1999); Poverty Reduction and Growth Facility--
Operational Issues (IMF: Washington, D. C., Dec. 13, 1999); Concluding
Remarks by the Chairman on Poverty Reduction Strategy Papers- Operational
Issues and Poverty Reduction and Growth Facility-- Operational Issues (IMF:
Washington, D. C., Dec. 21, 1999); and Key Features of IMF Poverty Reduction
and Growth Facility (PRGF) Supported Programs (IMF:

Washington, D. C., Aug. 16, 2000). The New Facility Is to

Focus More Explicitly on Poverty Reduction

Chapter 1: Introduction Page 18 GAO- 01- 581 IMF: New Lending Program

society, nongovernmental organizations, donors, and international
institutions. 8

 The participatory process is to facilitate open debate on issues such as
the social impact of policy measures and the pace and sequencing of reforms.
Discussions on the macroeconomic framework are to be more open and
iterative.

 Macroeconomic policies will need to be better integrated with social and
sectoral objectives to ensure that plans are mutually supportive and
consistent with a common set of objectives to spur growth and reduce
poverty. To ensure this consistency, the poverty reduction strategy will set
out a coherent and comprehensive strategy.

 The bottom- up approach will be reflected in the design of the
macroeconomic framework, including the level and composition of government
expenditures and the fiscal and external deficits. Key social and sectoral
policies, infrastructure projects, institutional reforms, and other measures
aimed at reducing poverty will be costed, prioritized, and incorporated
within the macroeconomic framework.

 Reconciling macroeconomic stability with spending levels needed to achieve
poverty goals will generally be a matter of raising an adequate level of
external grants or highly concessional loans. One key component will be to
mobilize external support, and IMF staff will need to intensify efforts to
identify sustained increases in resources for countries where these can be
used most effectively.

8 The World Bank?s Poverty Reduction Strategy Sourcebook defines
participation as ?a

process through which stakeholders influence and share control over
development initiatives and the decisions and resources that affect them.?
The major stakeholders involved in poverty reduction include the poor and
vulnerable groups, the general public, organized civil society, the private
sector, government agencies, representative assemblies, and donors. As the
Bank staff reported, participation is not a uniform process. The staff noted
that ?[ Participation] is likely to involve a cycle of participatory
dialogue, analysis, actions, and feedback within existing political and
governance structures that is designed to bring the views of all levels of
civil society, from communities to the private and public sectors, into
government policy- making and program implementation, at both the national
and local levels.? The goals are to reach the maximum level of participation
feasible within a particular country and to increase accountability,
transparency, and efficiency of governance structures in promoting
development and reducing poverty. Participation includes the following
continuum of approaches: information sharing, consultation, collaboration,
and empowerment.

Chapter 1: Introduction Page 19 GAO- 01- 581 IMF: New Lending Program

2. Budgets are to be more pro- poor and pro- growth.

 Government spending should shift toward activities that demonstrably
benefit the poor, while tax reforms should improve both equity and
efficiency.

3. Appropriate flexibility is to be ensured in fiscal targets.

 The program should be flexible enough to react to commonly experienced
shocks, such as deteriorating terms of trade or poor harvests, or to spend
newly available aid when it is clear that it could be used productively.

4. Analysis of the social impacts of major reforms is to be conducted before
the reforms are put in place (normally by the World Bank), and measures to
counteract negative impacts should be incorporated in the IMF?s program. 9

 The IMF should demonstrate that the effects of substantial
macroadjustments or structural reforms on different groups have been
considered. The program should also highlight countervailing measures to
offset temporary adverse effects on the poor. If a technical impact analysis
is needed, the World Bank should lead that effort, but IMF documents should
indicate what work was done and how it influenced policies.

5. Greater emphasis will be accorded to good governance.

 All government activities should have greater transparency, effective
monitoring procedures, anticorruption initiatives, effective public resource
management, accountability, and the involvement of all sectors of society in
monitoring relevant aspects of the program.

6. Programs are to contain more selective structural conditionality. 9
According to World Bank staff, ideally social impact analysis should be
undertaken as an integral part of the government- led preparation of PRSPs.
When such analysis has not been adequately done as part of the PRSP process
for policy measures supported by the IMF?s program, and further technical
impact analysis is needed, the World Bank should normally lead that effort.

Chapter 1: Introduction Page 20 GAO- 01- 581 IMF: New Lending Program

 Program documents are to discuss only those areas where the IMF had
primary responsibility (and in these areas conditionality would be used
sparingly). For example, the IMF is to focus on structural conditions that
may impact the macroeconomic framework, such as the privatization of
significant government entities. Conditionality covering areas within the
primary mandate of the World Bank would be the responsibility of the Bank
and would not be expected to appear in IMF financial arrangements.

According to IMF documents describing the PRGF, to help ensure that the
above key features are addressed in country- specific programs, program
documents are specifically to include the following. This information is to
provide a benchmark for future progress reviews.

 Documents should demonstrate that the distributional effects of
substantial macroeconomic adjustments or structural reforms have been
considered. For example, privatization of a state- owned utility may impact
(positively or negatively) employment levels and the cost and quality of
services. Or, tax increases may affect some vulnerable groups more than
others. Normally, the World Bank would undertake these analyses, drawing to
the extent possible on analysis already conducted during the preparation of
the PRSP, but PRGF documents should state what analysis was done and how it
influenced the policy choices.

 Documents should highlight measures designed to ease any temporary adverse
impacts. Negative impacts- such as the loss of jobs- may be discussed along
with mitigating actions, such as a job retraining program or temporary
social assistance for laid- off workers.

 Documents should state how fiscal targets would be modified in the event
of key shocks, such as a significant drop in the price of a major export.
Where relevant, programs should explicitly indicate how the fiscal
objectives have been, or will be, influenced by actual or likely shocks.
Several factors need to be weighed in determining the appropriate response
to an adverse shock, such as the amount and terms of financing and the
impact of higher spending on domestic demand, the real monetary exchange
rate, and the country?s competitiveness. Inevitably, questions will arise
regarding the amount of domestic financing and the appropriate inflation
target. IMF and World Bank documents state that since inflationary financing
represents a tax whose incidence falls principally on the poor, the risks
here need to be weighed carefully. According to the IMF, sacrificing low
inflation to finance additional expenditures is not an effective means to
reduce poverty, particularly in cases where inflation is above

Chapter 1: Introduction Page 21 GAO- 01- 581 IMF: New Lending Program

single- digit levels. All of these issues are to be discussed in the context
of the participatory process leading to the PRSP and elaborated in the
document itself.

 Country program documents should present clearer information to signal
financing needs. Programs could be presented in ways that give clearer
signals to donors about the connection between financing and the poverty and
growth goals. PRSPs may identify (1) additional priorities that could be
funded if donors provided more aid and (2) the areas in which spending cuts
would be made in the event of financing shortfalls. This change would
require better measurement and tracking of government expenditures.

 Documents should indicate how, in the course of the PRSP preparation
process, choices were made regarding the balance between the public and
private sectors in the allocation of total credit. It is generally desirable
that governments in low- income countries avoid or strictly limit their use
of relatively expensive domestic credit. But there may be situations in
which urgent public spending needs exceed the resources available from
revenues and external assistance, and a case could be made for drawing on
some of the domestic credit that would otherwise flow to the private sector.
This is another area in which flexibility can be exercised, depending on how
the relative needs of the two sectors are prioritized.

The Chairman of the Senate Committee on Foreign Relations asked us to
evaluate what the IMF has changed about its lending program for its poorest
members since it announced the PRGF, specifically to (1) analyze how this
new program is designed to be different from the previous program, (2)
assess what has actually changed in recipient countries? programs, and (3)
evaluate whether the changes implemented in the new program increase the
likelihood of recipient countries graduating from (that is, no longer being
eligible for) concessional borrowing in less than 15 years or have
contributed to a more short- term lending focus for the IMF.

For our first objective, we analyzed documents from the IMF describing the
design of its concessional lending programs between 1986 and 2001. To
describe the key elements of a macroeconomic framework, we studied IMF
documents on financial programming and staff research papers (workpapers,
discussion papers, and technical papers) and attended a Objectives, Scope,

and Methodology

Chapter 1: Introduction Page 22 GAO- 01- 581 IMF: New Lending Program

seminar at the IMF on financial programming. 10 Our research also included
an analysis of the current state of the development of macroeconomic models.
We also read research papers by other experts on this topic. We met with IMF
and World Bank officials to discuss the formulation of a country?s
macroeconomic framework and their understanding of how a broad- based,
participatory approach could be expected to influence the macroeconomic
framework. We also reviewed research conducted by staff from the IMF, the
World Bank, academia, consultant groups, and nongovernmental organizations
to identify empirical evidence supporting the extent to which specific
policies affect economic growth. To address all three issues, we analyzed a
wide range of documents from the Fund, the World Bank, 11 recipient and
donor governments, U. N. organizations, and nongovernmental organizations.
We also interviewed officials from these organizations in the United States
and abroad. Our findings regarding program design derive from this analysis
and are not based on individual country experiences.

To address the second issue, we selected and visited three countries for
case studies- Albania, Benin, and Honduras. These countries met the
following criteria as of September 30, 2000: they had a current Fund program
with no significant disruptions; were expected to have full poverty
reduction strategy papers by mid- 2001; and been reviewed under their
current and previous IMF programs, which provided us with documents for
comparing the PRGF to the ESAF. We met with and analyzed information on
countries? IMF programs and efforts to develop poverty reduction strategies
from officials of the IMF, the World Bank, and other multilateral
organizations; the recipient governments; the United States and other donor
countries; and nongovernmental organizations in the United States and
abroad. These officials also provided information on the capacity of
governments and civil society to influence the formulation of the country?s
macroeconomic framework. We did not generalize from the experience of these
countries to draw conclusions about overall

10 Financial programming is the framework used by the IMF to determine a
country?s macroeconomic program. The framework consists of several basic
accounting identities (relationships that are true by definition) and a few
behavioral relationships (such as a country?s demand for imports). To ensure
consistency of the outcome, judgments are made regarding underlying
relationships, such as the relationship between investment expenditures and
import requirements.

11 The World Bank, supported by its 182 member governments, promotes
economic growth and the development of market economies by providing
financing on reasonable terms to countries that have difficulty obtaining
capital.

Chapter 1: Introduction Page 23 GAO- 01- 581 IMF: New Lending Program

implementation issues. Instead, the case studies are examples of how
countries are addressing the design challenges identified in the first
objective.

For our third objective, to determine the likelihood of the 32 countries
that received PRGF disbursements in 2000 graduating from concessional
borrowing in 15 years or less, we utilized two databases. Using the per
capita income data from the World Bank?s World Development Indicators 2000
CD- ROM, we determined countries? past growth rates and calculated the per
capita income growth rate required for them to reach the International
Development Association?s (IDA) income threshold in 15 years. 12 Using 15-
year forecasted growth rates estimated by Standard & Poor?s DRI, a leading
provider of economic information, for each of the 32 countries, we
determined what each country?s projected per capita income would be in 15
years and how many years it would take each country, at the DRI- projected
growth rate, to reach the IDA threshold. 13 To evaluate whether the PRGF has
contributed to a more short- term lending focus for the IMF, we reviewed IMF
documents and data on its concessional lending programs.

We performed our work from August 2000 through March 2001 in accordance with
generally accepted government auditing standards.

12 The IDA income threshold is generally used to determine whether a country
is eligible for concessional lending. 13 DRI is a unit of Standard & Poor?s,
which is a division of The McGraw Hill Companies. It is a leading provider
of industry and economic data, forecasting, and consulting services. Its
World Forecast Database includes historical and forecast data for 170
countries.

Chapter 2: The Design of the IMF?s New Program Differs Little From Its
Previous Program

Page 24 GAO- 01- 581 IMF: New Lending Program

We found that the design of the new PRGF differs little from the IMF?s
previous program. Although certain elements of the new program have greater
emphasis placed on them now than in the past, we observed that most of the
announced changes for the new program were pursued under the previous
program, and the one major design change- national ownership of effective
macroeconomic policies- is difficult to achieve. Given the need for poor
countries to maintain macroeconomic stability, which is essential for
economic growth and poverty reduction, the policies and targets within the
macroeconomic framework are not likely to be altered substantially from the
past. Our analysis shows that national ownership of the macroeconomic
framework is difficult to achieve for three reasons: (1) the limited
capacity of many recipient governments to independently analyze and
effectively negotiate the macroeconomic framework reduces the opportunity
for country- specific elements to be addressed; (2) the challenges to
effectively engaging civil society in a dialogue on these very complex
matters are significant; and (3) a national dialogue on the choice of
effective policies is hampered by the limited knowledge of all parties,
including the IMF, on how different policies actually affect elements of the
macroeconomic framework.

Our analysis of IMF documents shows that the design of the Fund?s new
lending program for low- income countries- the PRGF- does not differ
significantly from its previous program- the ESAF- because it includes
concepts that have been pursued for a number of years. As shown in table 1,
according to the IMF, it has focused on key issues such as reducing poverty
and increasing the recipient government?s role in developing its Fund
programs since at least the late 1980s. Chapter 2: The Design of the IMF?s
New

Program Differs Little From Its Previous Program

Most of the Announced Modifications Represent a Change in Emphasis Rather
Than a Change in Philosophy

Chapter 2: The Design of the IMF?s New Program Differs Little From Its
Previous Program

Page 25 GAO- 01- 581 IMF: New Lending Program

[Blank page]

Chapter 2: The Design of the IMF?s New Program Differs Little From Its
Previous Program

Page 26 GAO- 01- 581 IMF: New Lending Program

Table 1: The Fund?s Stated Focus on Poverty Reduction, Government
Leadership, and Civil Society Participation

Sizable reduction in the incidence of poverty is the highest priority for
the international development community. A broad consensus is emerging on
strategies to achieve this goal.

Policy framework papers identify policies that help and those that may hurt
the poor. Essential ingredients for

successful structural adjustment programs include integration into program
design of

* poverty reduction objectives  ways to mitigate adverse effects on the
most vulnerable groups, preferably through income- generating programs.

Civil society participation

The IMF Executive Board said the explicit focus on alleviating structural
problems was particularly important, since many countries suffered for many
years from low rates of economic growth and declining per capita incomes.

The IMF Executive Board  welcomed the increased attention being paid to the
impact of IMF programs on income distribution and the poorest population
groups

 supported occasional use of measures to ease the impact of reform on these
groups.

Focus on poverty reduction

Government leadership in

developing strategy

According to the IMF Board  country authorities should play a greater role
in developing PFPs.

 PFPs should include more discussion of country authorities? priorities.
Successful reform and adequate safeguards for IMF resources require the
borrower to be fully committed to the reform program.

IMF programs need to pay due regard to countries? domestic social and
political objectives.

The IMF Board  stressed that the member country should play the leading
role in preparing the PFP, so that the document would be an authentic
expression of the country?s objectives and policies

 emphasized the importance of achieving consensus within the member
country?s government on the content of

an adjustment program. In some cases, additional time might be needed to
permit full discussion of major issues before the PFP was drafted.

The IMF Board said that focusing on programs? impacts on income distribution
and the poorest population groups enhanced programs? chances of success by
minimizing public resistance to them.

In a March 1989 review of concessional lending programs, the IMF Board
stressed that the involvement of member countries? authorities in program

design be intensified. The Board emphasized the importance of allowing
sufficient time for political consensus to be reached in borrowing
countries.

Other essential ingredients for success include well- designed programs with
broad public understanding/ support that take into account the country?s
deep- seated structural problems and social, demographic, and political
environment.

Governments of developing countries have primary responsibility for
achieving goal of poverty reduction. This goal could be most effectively
achieved through national development strategies that

 include sound macroeconomic and structural policies

 encourage sustainable growth that increases income earning opportunities
for the poor

 develop the human resources of the poor, particularly through broad access
to education, health, and family planning services.

Developing countries? experience in the 1970s shows that growth founded on
inflation and excessive external borrowing and not built on broad- based and
creative participation of the largest part of the population is deprived of
its essential driving force. Involvement of the poor in designing and
executing projects and programs is key to effective poverty reduction
efforts.

1987 1988 1989 1990 a b b c A

Chapter 2: The Design of the IMF?s New Program Differs Little From Its
Previous Program

Page 27 GAO- 01- 581 IMF: New Lending Program

a International Monetary Fund, Annual Report of the Executive Board for the
Financial Year Ended April 30, 1987 (Washington, D. C.: IMF, 1987) and
International Monetary Fund, Annual Report of the Executive Board for the
Financial Year Ended April 30, 1988 (Washington, D. C.: IMF, 1988). b
International Monetary Fund, Annual Report of the Executive Board for the
Financial Year Ended April 30, 1989 (Washington, D. C.: IMF, 1989) including
press communiquï¿½s of the Joint Ministerial Committee of the Boards of
Governors of the Bank and the Fund on the Transfer of Real Resources to
Developing Countries (Development Committee). The Development Committee,
which is composed of 24 members who are usually ministers of finance or
development, advises the IMF?s and the World Bank?s Boards of Governors.
Legend

PFP = Policy framework paper The IMF focused more on the social issues of

adjustment. Short- term focus: mitigate adverse effects on vulnerable
groups. Longterm poverty reduction is best achieved by sustained and broad-
based growth and improvements in level and quality of government spending on
social services. The IMF focused more on composition and efficiency of
public expenditures. The IMF continued to find ways,

commensurate with its mandate as a monetary institution, to contribute to
reducing poverty by

 emphasizing sound macroeconomic and structural policies, which promote
sustainable growth worldwide

 discussing impacts of economic policies for poor groups and appropriate
policy mix to achieve reform.

Explicit focus on poverty reduction within a country?s overall development
strategy. Growthoriented policies should be implemented in a framework in
which the pressing need to reduce poverty is also a central objective.

All IMF programs resulted from stronger collaboration among the governments,
the IMF, and the World Bank through the PFP.

Political commitment to reform was critical.

Vested interests were often against reforms; ways to ease the impact on
these groups and persuade them of the longer- term benefits of reforms were
important. Declining terms of trade eroded support

of reforms. Countries, the IMF, and others developed mix of

policies in close consultation. The IMF programs have increasingly paid
attention to the mix and sequencing of policies, with a view to minimizing
adverse effects on the poor. This helps increase the political
sustainability of economic reforms. Ultimately, the choice of social and
economic policies belongs to the government.

It is critical that key players-- especially at the grassroots level-- have
a stake in economic policy- making. Strategy of high- quality economic
growth-- which

is key to poverty alleviation-- includes participatory development through
active involvement of all groups in society.

The country and its people will need to take the lead.  The government will
prepare the poverty

reduction strategy paper based on a process involving the active
participation of civil society, nongovernmental organizations, donors, and
international institutions.

 The most fundamental changes to the IMF?s work arise from the fact that
the targets and policies in IMF programs will emerge directly from the
country's poverty reduction strategy.

 Discussions on the macroeconomic framework will be more open and
iterative.

 The costs of poverty- reduction programs will more directly influence the
design of the macroeconomic framework.

Broad participation and greater ownership to ensure that civil society is
involved and that country authorities are in the driver's seat.

d ef 1992- 93 1995 1996 A

Chapter 2: The Design of the IMF?s New Program Differs Little From Its
Previous Program

Page 28 GAO- 01- 581 IMF: New Lending Program

c International Monetary Fund, Annual Report of the Executive Board for the
Financial Year Ended April 30, 1990 (Washington, D. C.: IMF, 1990);
International Monetary Fund and World Bank, Finance and Development
(Washington, D. C.: IMF and World Bank, Sept. 1990); and International
Monetary Fund, Annual Report of the Executive Board for the Financial Year
Ended April 30, 1991

(Washington, D. C.: International Monetary Fund, 1991), including press
communiquï¿½s of the Joint Ministerial Committee of the Boards of Governors of
the Bank and the Fund on the Transfer of Real Resources to Developing
Countries (Development Committee). d International Monetary Fund, Annual
Report of the Executive Board for the Financial Year Ended

April 30, 1992 (Washington, D. C.: International Monetary Fund, 1992) and
International Monetary Fund, Economic Adjustment in Low- income Countries:
Experience Under the Enhanced Structural Adjustment Facility, Occasional
Paper 106 (Washington, D. C.: IMF, Sept. 1993). e International Monetary
Fund, Social Dimensions of the IMF?s Policy Dialogue, Pamphlet Series No. 47
(Washington, D. C.: IMF, Mar. 1995). f The Poverty Reduction and Growth
Facility (PRGF)- Operational Issues (Washington, D. C.: IMF and

World Bank, Dec. 13, 1999); Social Policy Issues in IMF- supported Programs:
Follow- up on the 1995 World Summit for Social Development (Washington, D.
C.: IMF, Mar. 16, 2000); Key Features of IMF Poverty Reduction and Growth
Facility (PRGF) Supported Programs (Washington, D. C.: IMF, Aug. 16, 2000);
and International Monetary Fund, News Brief No. 00/ 81 (Washington, D. C.:
IMF, Sept. 7, 2000).

Sources: IMF documents listed in above table notes.

According to our analysis, one change- that the macroeconomic targets and
policies in IMF programs will be the responsibility of the government and
emerge from a process involving civil society and donors- could be a
significant departure from how targets and policies have been traditionally
chosen. However, even this approach builds on and includes features that
have been discussed and pursued for some time. As shown in table 1, staff
documents indicate that the IMF has focused on the importance of government
leadership and domestic public support for reform programs since at least
the early 1990s.

The commonality of these concepts over time does not necessarily mean that
the new program is identical to the previous program, since the new program
envisions consolidating these concepts into one framework and giving them
greater prominence. We believe that this represents a continued increase in
emphasis rather than a change in the Fund?s stated philosophy, with IMF
documents placing increased focus on these topics since the external review
of the ESAF in January 1998. 1 For instance, under the PRGF, program
development is to begin with the country?s priorities (as described in the
country?s poverty reduction strategy paper) instead of the IMF?s
macroeconomic targets, as was done in the past. While this represents a
shift in the IMF?s starting point, a similar approach was available under
the previous facility- if a country identified needs

1 Botchwey and others, Report of the Group of Independent Persons.

Chapter 2: The Design of the IMF?s New Program Differs Little From Its
Previous Program

Page 29 GAO- 01- 581 IMF: New Lending Program

that were greater than its available financing, then the government, the
Fund, or others could seek additional donor and creditor support.
Furthermore, according to IMF documents, since at least 1998, countries that
achieved macroeconomic stability were permitted to increase spending on
priority areas over previous program levels, if the spending were financed
through donor resources.

The IMF has had difficulties achieving some of the goals listed in table 1
in the past, and it is uncertain whether it will be able to do so in the
future. For example, since at least 1987 getting governments to take
ownership of their IMF programs has been seen as an important element in the
programs? successful implementation. However, the main vehicle to achieve
this- the policy framework papers- has been generally developed by the staff
of the IMF and not by the countries themselves. According to an external
evaluation of the ESAF commissioned by the IMF,

?although initially the PFP [policy framework paper] had held great promise
as an instrument of a genuine three- way dialogue between the government,
the Fund, and the Bank, it has become a rather routine process whereby the
Fund brings uniform drafts (with spaces to be filled) from Washington, in
which even matters of language and form are cast in colorless stone? the
general yearning therefore was for the realization of a potential that never
was- a truly country- specific PFP, agreed on the basis of a government- led
consultation process.? 2

In responding to this evaluation, IMF staff noted that a wider range of
initiatives to promote ownership is clearly needed. Under the new program,
the governments are to write the PRSP themselves. According to Fund staff,
the IMF Executive Board could endorse PRSPs that contain elements that the
Fund does not agree with. However, the significance of this change is
uncertain since the IMF?s Executive Board must endorse a country?s overall
PRSP in order for the country to borrow from the Fund.

2 Botchwey and others, Report of the Group of Independent Persons.

Chapter 2: The Design of the IMF?s New Program Differs Little From Its
Previous Program

Page 30 GAO- 01- 581 IMF: New Lending Program

According to the IMF and the World Bank, macroeconomic stability is
considered to be a necessary prerequisite for economic growth and poverty
reduction, although not sufficient on its own to achieve those goals. 3
Countries that experience macroeconomic instability- such as high inflation,
volatile and over- valued exchange rates, or excessive fiscal deficits- have
tended to have slow or even negative economic growth rates. 4 According to
the IMF and the World Bank, macroeconomic instability can also place a heavy
burden on the poor, for example, inflation can place a disproportionate
burden on those in the lower income brackets. This concern over the negative
effects of macroeconomic instability underlies the continuing goal that a
country?s macroeconomic framework should work to maintain stability, once
achieved.

The IMF has been accused in the past of having an overly austere approach to
macroeconomic policy, sacrificing potential economic growth and poverty
reduction in order to reduce a country?s inflation rate and deficit level.
The impact of too much austerity would be to lower economic growth from its
optimal level and impede progress on poverty reduction. In recent documents,
the IMF presents the view that it is now open to alternative approaches, as
long as they are consistent with

3 Macroeconomic stability exists when key economic relationships are in
balance, for example, between domestic demand and output, the balance of
payments, fiscal revenues and expenditure, and savings and investment. These
relationships need not necessarily be in exact balance. Imbalances such as
fiscal and current account deficits or surpluses are perfectly compatible
with economic stability provided that they can be financed in a sustainable
manner.

4 The IMF and the World Bank PRSP Sourcebook identifies macroeconomic
instability as a situation in which a country has large current account
deficits financed by short- term borrowing, high and rising levels of public
debt, double- digit inflation rates, and stagnant or declining output.
Widely Accepted Goal

of Maintaining Macroeconomic Stability Limits Potential Changes Under the
Program

The Importance of Maintaining Macroeconomic Stability

Chapter 2: The Design of the IMF?s New Program Differs Little From Its
Previous Program

Page 31 GAO- 01- 581 IMF: New Lending Program

macroeconomic stability and growth. In discussing the range of macroeconomic
policies available, IMF and World Bank documents refer to a ?substantial
gray area? between those policies that may be considered too austere and
those that cause macroeconomic instability; in other words, the gray area
represents the range of optimal policy options. Figure 2 illustrates the
macroeconomic framework, with the four key targets that are instrumental in
pursuing poverty reduction and growth, in the context of macroeconomic
stability.

Chapter 2: The Design of the IMF?s New Program Differs Little From Its
Previous Program

Page 32 GAO- 01- 581 IMF: New Lending Program

Figure 2: Elements of a Macroeconomic Framework in Support of the PRGF?s
Twin Goals of Poverty Reduction and Growth, in a Context of Macroeconomic
Stability

Source: GAO analysis of IMF and World Bank documents.

Chapter 2: The Design of the IMF?s New Program Differs Little From Its
Previous Program

Page 33 GAO- 01- 581 IMF: New Lending Program

Presumably, one goal of including the macroeconomic framework within the
national poverty reduction dialogue would be to explore this gray area to
establish an effective mix of policies consistent with the medium- term
goals of the country.

Based on our analysis of numerous documents and discussions with IMF and
World Bank officials, it is difficult to determine how ?substantial? or wide
is the range of macroeconomic policy targets within this so- called

?gray area.? This is due to two factors: Precise identification of the
bounds of the gray area is beyond the current understanding of the economics
profession, and the harsh economic realities confronting these very poor
countries also work to limit the choice of policies or amount of spending
that can be used effectively. Both of these factors are strongly influenced
by the desire to ensure that whatever macroeconomic framework is agreed to,
it does not put the country at greater risk of macroeconomic instability. 5

Although there is general agreement within the economics profession that
certain policies will likely lead to macroeconomic instability (such as a
rapid increase in the money supply, leading to high inflation), there is
little agreement on what constitutes the best approach. For example, many
economists, including some at the IMF, think that inflation above a 7 to 11
percent range is risky, whereas others think the level can be between 20 and
40 percent before it starts to endanger economic growth. The uncertainty
created by this lack of precision may increase the likelihood that the more
conservative approach will be agreed upon to avoid the

5 The focus of the macroeconomic framework for countries that are
experiencing instability has been to reestablish stability. Most of the
countries that are current borrowers from the PRGF program are considered to
have achieved or nearly achieved macroeconomic stability. The Fund does not
explicitly classify countries as macroeconomic stable or unstable, nor does
it provide threshold levels or the ?gray areas? separating stability from
instability for relevant macroeconomic variables. However, the PRSP
Sourcebook provides country data and criteria and concludes that ?many
developing countries are presently in a state of macroeconomic stability.?
Countries are classified over the recent economic period 1994- 99 as good
performers or poor performers for three macroeconomic variables: real gross
domestic product (GDP) average annual growth rate, average annual inflation
(GDP deflator), and average annual primary fiscal budget surplus/ deficit as
a percent of GDP. (The threshold levels for good performance are real GDP
growth rate greater than 2 percent, inflation below 20 percent, and primary
fiscal surplus/ deficit as percent of GDP greater than -3 percent.) Using
the same variables and criteria for the 32 countries that received PRGF
disbursements in 2000, we found that 69 percent of the countries are
classified as good performers for at least two of the variables, and 41
percent for all three variables. The Area Available for

Negotiation Is Limited

Chapter 2: The Design of the IMF?s New Program Differs Little From Its
Previous Program

Page 34 GAO- 01- 581 IMF: New Lending Program

perils of macroeconomic instability. Whereas an overly conservative approach
may both lower economic growth and reduce spending on poverty reduction, the
negative consequences of macroeconomic instability are much greater and are
likely to result in even lower economic growth and require greater austerity
to recover from it. For example, policies identified under the poverty
reduction strategy may include measures such as a significant increase in
spending for primary education and health, beyond what domestic revenue can
fund, what the donors are willing to finance, and what could be used
effectively given some countries? limited capacity. However, domestic
financing of these priority areas would increase the budget deficit, the
risk of high inflation, and an unsustainable debt burden in the future. How
much of an increase is affordable is difficult to determine, but the fear of
macroeconomic instability would limit the amount of new spending in these
areas.

These concerns are magnified for the very poorest countries, given the
underlying fragility of their economies. In addition to domestic factors
that should be considered for maintaining macroeconomic stability, such as
the identification of how much spending can be prudently increased over a
short period of time, consideration should also account for the
vulnerability of these economies to external shocks. This vulnerability is
due to many factors, including a reliance on just a few basic commodities
for much of their export income, weak revenue- generating capacity, and
substantial dependence on imported oil. This tension was evident in the
circumstances of Honduras between 1993 and 1994. In the period leading up to
the presidential elections of November 1993, domestic spending rapidly
increased, more than doubling the fiscal deficit from 4. 8 percent of gross
domestic product to 10.7 percent. The country then experienced a severe
drought in 1994, creating a serious electricity shortage and dramatically
lowering the volume of its leading exports- coffee and bananas. The combined
effect of the increased domestic spending and drought was to increase
inflation from 12.9 percent in 1993 to 28.4 percent in 1994, to lower per
capita growth from 6.2 percent in 1993 to negative 1.5 percent in 1994, and
to reduce Honduras?s international reserve levels to less than 1 month of
imports.

Chapter 2: The Design of the IMF?s New Program Differs Little From Its
Previous Program

Page 35 GAO- 01- 581 IMF: New Lending Program

The development of a nationally owned macroeconomic framework faces several
constraints. The limited capacity of many recipient governments to
independently analyze and effectively negotiate the macroeconomic framework
reduces the opportunity for country- specific elements to be brought into
the dialogue. The complex nature of this subject matter presents significant
challenges to an effective civil society dialogue. Furthermore, even if
these challenges were overcome, a national dialogue on the choice of
effective policies is complicated by the lack of knowledge by all parties,
including the IMF, about how different policies actually affect the
macroeconomic framework.

Borrower governments are responsible for preparing the poverty reduction
strategy, negotiating the IMF program, and deciding among difficult trade-
offs. However, the limited technical capacity within the government and the
substantial complexities inherent in establishing macroeconomic policies and
targets have constrained the involvement of governments in establishing
their previous macroeconomic programs, with many macroeconomic policies and
targets based on the technical analyses of the Fund and others. The PRGF and
PRSP are to increase the role of borrower governments by calling for a
nationally owned program developed with civil society participation. Under
these programs, governments are expected to (1) more openly discuss the
selection, impacts, and trade- offs of various macroeconomic policies with
civil society; (2) assess country- specific circumstances that affect
economic growth and poverty reduction; and (3) evaluate and respond to the
information received from civil society. Yet there are questions as to
whether many governments are in a position to engage effectively in such
discussions, according to some IMF and World Bank officials. Efforts to
increase borrower governments? ability to develop and assess macroeconomic
issues have been ongoing since at least the late 1980s. 6 However, these
efforts have not increased capacity sufficiently, according to some donors
and creditors. This is of particular concern, given the increased emphasis
on country ownership and participation. Without such

6 The objective of the Social Dimensions of Adjustment program was to help
participating African countries to integrate poverty and social concerns in
the design and implementation of their adjustment programs so as to mitigate
the burden on the poor in the process of structural adjustment. Its mandate
was to strengthen the capacity of African governments to design appropriate
programs and projects in this regard. The World Bank, the African
Development Bank, and the U. N. Development Program jointly administered the
program. Developing a

Nationally Owned Macroeconomic Framework Is Difficult

The Importance of Increased Capacitybuilding for Recipient Governments

Chapter 2: The Design of the IMF?s New Program Differs Little From Its
Previous Program

Page 36 GAO- 01- 581 IMF: New Lending Program

increased capacity and the willingness of donors and creditors to accept
government views, technical analyses will remain outside the governments?
hands, and the goal of country- owned macroeconomic programs focused on
reducing poverty may not be achieved.

The limitations on flexibility resulting from the need to maintain
macroeconomic stability do not mean that there is no role for country
involvement in the process. Although the IMF?s financial programming
approach has been applied to all member countries, its relevance is improved
by using country- specific information. While the iterative approach of the
staffs of the IMF and the World Bank attempts to capture these local
differences as accurately as possible, there have been instances in the past
where they misread a country?s underlying circumstances, resulting in
mistakes in the macroeconomic framework. For example, according to an expert
on economic development in Africa, the IMF made mistakes in the programs of
Zambia and Malawi due to not understanding the countries? individual
circumstances. 7 Active involvement on the part of technically skilled
representatives of each country would presumably help reduce imprecision in
the formulation of their macroeconomic framework through greater
understanding of the country?s unique circumstances.

As discussed previously, the targets and policies of the PRGF program are
expected to emerge directly from the country?s poverty reduction strategy.
The challenges in establishing a participatory process and the complexities
of macroeconomic issues may limit the extent to which civil society can
influence the macroeconomic framework in the near term. The IMF and the
World Bank have urged countries to adopt key concepts such as ?ownership?
and ?civil society participation? in discussing macroeconomic policies.
However, neither the Fund nor the World Bank has defined these concepts.
Given the desire of the Fund and the World Bank to allow for country-
specific interpretations, countries are expected to define these concepts
for themselves. As we previously reported, just establishing a participatory
process itself is challenging. 8 For example, countries face challenges in
determining which groups represent civil

7 Bruce R. Bolnick, The Role of Financial Programming in Macroeconomic
Policy Management, Harvard Institute for International Development
Discussion Paper No. 720 (Cambridge, MA: Sept. 1999). 8 See Developing
Countries: Debt Relief Initiative for Poor Countries Faces Challenges (GAO/
NSIAD- 00- 161, June 29, 2000). Civil Society Unlikely to

Significantly Affect the Macroeconomic Framework in the Near Ter m

Chapter 2: The Design of the IMF?s New Program Differs Little From Its
Previous Program

Page 37 GAO- 01- 581 IMF: New Lending Program

society, since stakeholders include those directly affected by the project
or strategy, such as the poor, as well as those who are indirectly affected,
such as the borrower governments at the local level, nongovernmental
organizations, and private sector organizations. Identifying representatives
can be especially difficult in countries that lack a democratic or
representative tradition and thus have few existing means for getting
citizen or nongovernmental organizations? input or for electing
representatives. These concerns are magnified when the process is to involve
the poor or groups that have traditionally been excluded, such as women and
indigenous populations. There is also concern that government officials will
not support a participatory process if it is perceived as diluting their
power or alienating influential constituencies.

To effectively involve civil society- however defined- in discussing
macroeconomic policies and targets, efforts may be needed to educate them
and their representatives about the macroeconomic framework and build their
organizational and financial capacity to participate. Specialized
information and skills are needed to assess the trade- offs between, for
instance, a higher fiscal deficit and a higher rate of inflation or to
determine the impact of a floating exchange rate on economic growth.
Moreover, as previously discussed, within the academic literature there is
limited knowledge about how much flexibility exists in the macroeconomic
framework with, for example, uncertainty about how much spending can be
increased without jeopardizing macroeconomic stability. Given the complexity
of the issues, participation in influencing macroeconomic policy could be
limited to a few informed constituencies or experts, such as representatives
from academia and the business community and others who can assess trade-
offs. Considering the above difficulties, civil society may not be able to
influence such macroeconomic targets as inflation or exchange rates through
the initial poverty reduction strategy; however, civil society may help
improve the allocation of resources and increase the amount of resources
donors are willing to provide by helping establish priorities for poverty
reduction.

Even if the capacity of the national governments were improved and civil
society were effectively engaged in a dialogue on the macroeconomic
framework, national ownership would be hampered by the current limitations
in economic knowledge. An assumption of the PRGF program is that the
macroeconomic framework is designed to support the twin goals of sustained
economic growth and poverty reduction within the context of stability.
National ownership of the macroeconomic framework could then reflect
agreement on the broad macroeconomic targets and the National Dialogue

Hampered by Limited Knowledge About the Impact of Policies

Chapter 2: The Design of the IMF?s New Program Differs Little From Its
Previous Program

Page 38 GAO- 01- 581 IMF: New Lending Program

policies that support those goals. The expectation of the IMF that countries
will assume ownership of their macroeconomic frameworks creates the
impression that these countries can hold a meaningful debate on the
alternative policy choices that can be implemented, with the policy mix most
compatible with national interest embraced. However, a meaningful national
dialogue on the range of policies that are available is constrained by the
limited knowledge of all parties, including the IMF, about how different
policies actually affect elements of the macroeconomic framework. (See app.
I for a fuller discussion of how limitations in the existing body of
economic knowledge can work to constrain national dialogue.)

The ability to analyze trade- offs among different policy choices requires
an underlying model that explains the relationships between these policies
and the macroeconomic framework. For example, if as part of the national
dialogue the country sought to consider a slowing or even abandonment of its
privatization efforts due to concern over lost employment and potential
price increases, it would have no way of gauging the impact of this policy
on future economic growth. The expectation of the IMF and others is that
privatization will lead to increased investment and economic growth.
However, there are cases in which the expected growth and investment have
not materialized. Attempts to construct models that explain the impact of
these various policy choices have not been successful, with the current
expectation that such models need to be tailored to each country?s
individual circumstances in order to be meaningful. The World Bank and
others are attempting to develop these models, but the process is slow due
to technical complexities and limited, country- specific data.

Chapter 3: Few Changes Evident in Three Countries? Programs

Page 39 GAO- 01- 581 IMF: New Lending Program

We found a few changes in the IMF programs of the three countries we
reviewed- Albania, Benin, and Honduras- that can be clearly attributed to
the changes announced by the IMF in 1999. Many elements of the PRGF were
also included in the countries? previous programs, making it difficult to
determine to what extent their presence reflects the announced changes in
the IMF?s approach rather than the direction the country?s program was
moving in anyway. Our review of these countries? efforts highlighted the
difficulties countries face in developing a nationally owned macroeconomic
framework. First, each government has limited capacity to analyze
macroeconomic issues independently. Second, while all three governments have
begun to establish a participatory process and promote dialogue with civil
society, many people that we spoke with were unsure how to use this dialogue
to address a country?s complex macroeconomic policies and targets, other
than the composition and level of spending. Third, we did not see evidence
of changes in country- specific program documents that were called for under
the PRGF, such as the inclusion of assessments of the social impacts of
reforms and measures to alleviate any negative impacts, which could help
facilitate participants? analysis of issues.

The changes in the IMF programs of the three countries we reviewed that can
be clearly attributed to the changes the IMF announced in 1999 are (1)
Albania?s establishment of a participatory approach for preparing a national
development strategy and (2) fewer structural performance criteria in Benin.
For example, in Benin, between 1998 and 2000 the number of structural
performance criteria decreased from five to two as IMF conditions on the
civil service wage scale and privatization of stateowned- enterprises were
eliminated. 1 However, the significance of these changes for the IMF program
is not clear. For instance, Albania?s participatory process does not
describe how Albania will include civil society in discussing the
macroeconomic framework, and Benin still has to meet two of the three
structural performance criteria dropped from the IMF program if it is to
receive full debt relief under the Heavily Indebted

1 Benin met its two remaining performance criteria by the end of August
2000, according to an IMF document. Chapter 3: Few Changes Evident in Three

Countries? Programs Inclusion of Some Elements Difficult to Attribute to New
Program

Chapter 3: Few Changes Evident in Three Countries? Programs

Page 40 GAO- 01- 581 IMF: New Lending Program

Poor Countries Initiative. 2 The inclusion of PRGF- type elements in the
three countries? current and previous IMF program is illustrated in table 2.

Table 2: Inclusion of PRGF Elements in Three Countries? IMF Programs Key
elements of PRGF (per IMF) Albania Benin Honduras

Greater degree of government ownership and civil society participation

 Under the PRGF/ PRSP, Albania established its first participatory process
that includes civil society representatives in preparing a national
development strategy.

 Government officials told us (1) they believe their current IMF program is
more of a government- led program than the one negotiated in 1998 because
they believe they have increased their capacity to negotiate and (2) the IMF
program they hope to negotiate in 2001 will be a country- owned program that
is linked to the Growth and Poverty Reduction Strategy (Albania?s PRSP).

 Several issues- such as civil service reform and privatization of the
cotton industry- have been points of contention in Benin?s past and present
IMF programs. The first PRGF review notes some of these divergences in
views, as well as the agreement between the World Bank and the government of
Benin to have consultants assess the impact of the different cotton
privatization scenarios.

 Broad participatory process is not new for Benin. Benin undertook
participatory processes in 1994 and 1998, including one to create a poverty
reduction action plan based on full participation of the vulnerable groups.

 There has been very little progress in civil society participation in the
poverty reduction strategy. No consultations had taken place at year- end
2000, though 12 regional workshops were planned, according to the Interim-
PRSP. This is partly due to the presidential elections that took place on
March 4, 2001, and were expected to be ongoing through the end of March
2001.

 The government of Honduras has previously worked with civil society to
define priorities for the country. After Hurricane Mitch, the government
prepared a plan aimed at promoting economic revival and job creation, which
involved analysis, discussion, and consultation with representatives of
Honduran civil society.

 In 2000, the government held 13 regional meetings (including in very
remote areas) to present and discuss the draft poverty reduction strategy
and met with some nongovernmental organizations in the capital, and members
of parliament. Besides resistance toward privatization of electricity
distribution, macroeconomic issues that the IMF deals with were not
prominent in these discussions. Instead, interest groups generally pushed
their sector- specific issues.

Increased focus on poverty reduction Reducing poverty has been an

objective of Albania?s IMF program since 1997, when it outlined medium- and
short- term strategies for alleviating poverty.

Throughout the 1990s, Benin increased its focus on poverty reduction. To
combat widespread poverty, the government undertook national programs and
projects at the sector level specifically to address the vulnerable
populations.

Since 1995, poverty in Honduras has been discussed more in IMF documents.

Budgets more propoor and pro- growth Since 1998, the government and

the IMF have sought to Budgeted expenditures for education and health in
2000 and 2001 are Targeted increases in

expenditures since 1998 are

2 The Heavily Indebted Poor Countries Initiative is a comprehensive approach
for providing debt relief to the poorest and most indebted countries in the
world. For more information on this initiative, see Developing Countries:
Debt Relief Initiative for Poor Countries Faces Challenges.

Chapter 3: Few Changes Evident in Three Countries? Programs

Page 41 GAO- 01- 581 IMF: New Lending Program

Key elements of PRGF (per IMF) Albania Benin Honduras

 increase social spending, with spending on health and education increasing
as a percentage of GDP since 1998. The government projects spending on
health and education to increase from 4.7 percent of GDP in 1998 to 7.1
percent in 2003.

 reduce the domestically financed deficit to free up credit for private
sector. In 1998, credit to the private sector grew by 15 percent. In 2000,
it is projected to grow by 39 percent. The government projects public
investment expenditures (i. e., infrastructure) to increase from 5.3 percent
of GDP in 1998 to 8.7 percent in 2003.

greater than in the recent past but have increased by less than 0.5 percent
of GDP. To receive debt relief, Benin must increase basic health
expenditures including reproductive health, child immunization, and HIV/
AIDS prevention. A key objective of the 1996- 99 program was to raise
economic growth to 6 percent. The government?s main concern was to create
growthenhancing policies that would allow for substantial poverty
alleviation. High and sustainable growth rates of 5- 6 percent continue to
be a goal in the current program.

taking place within the context of efforts to recover from Hurricane Mitch,
which makes it difficult to attribute projected increases solely to the
PRGF. Pro- poor: More spending for social expenditures (such as rural
development and health) and targeting of subsidies to the poor. Social
expenditures to increase from about 8 percent in 1999 to about 11 percent of
GDP in 2000. Pro- growth: Elimination of remaining export taxes may benefit
small banana producers. To support the growth and recovery after Hurricane
Mitch, public sector investment is programmed to increase 6.3 percent of GDP
in 1999 to 7.9 percent in 2000.

Flexibility in fiscal targets The overall fiscal deficit as a

percent of GDP has remained relatively large during 1997- 2000, averaging 11
percent. While the domestically financed budget deficit has declined from
10.7 percent in 1997 to 3.3 percent in 2000, foreign financing rose over
this period. Whereas the PRGF document projects a decrease in the overall
deficit to 7.1 percent in 2003, the government of Albania projected the
overall deficit to be 9.4, according to the Medium Term Expenditure
Framework finalized in December 2000. An important reason for the difference
between these two projections stems from different expectations on future
foreign financing.

Deficits slightly larger than the deficit in 2000 are projected, partly
reflecting increased expenditures budgeted for health and education.
However, these higher deficit levels may not materialize because of concerns
about the government?s ability to use additional resources effectively.

For the most part, Honduras has been meeting fiscal targets for the last few
years. After Hurricane Mitch struck, the IMF program targeted the central
government?s 1999 fiscal deficit to increase to 8.6 percent of GDP, but the
actual deficit was about 1 percent. For 2000, the IMF programmed a deficit
of 6.8 percent of GDP.

Analysis of social impacts of major reforms conducted and countervailing
measures included

None mentioned. Minor anecdotes but no social impact assessments discussed
in current or previous IMF program.

No social impact studies cited. April 2000 document said, ?The

nature and poverty impact of social policies will be reviewed in the process
of preparing the final PRSP.? Greater emphasis on good governance The 1997
program identified the

government?s limited/ weak (but improving) ability to formulate and
implement macroeconomic and

Ministries have had difficulties implementing public investment projects,
especially those domestically financed. Weak planning, budgeting,

The January 1995 policy framework paper stated that Honduras?s public sector
has a shortage of technical and

Chapter 3: Few Changes Evident in Three Countries? Programs

Page 42 GAO- 01- 581 IMF: New Lending Program

Key elements of PRGF (per IMF) Albania Benin Honduras

structural measures as the main risk to the program and stated that
obtaining future concessional IMF lending depended on improving capacity.

and audit mechanisms, and corruption within the public sector, especially
the ministries of health and education, continue to be the major
constraints.

managerial staff. The ministries, the public enterprises, and the central
bank required institution building and technical assistance in many areas,
including project preparation and evaluation, budgetary design and control,
and tax policy and administration. To assist the government in these areas,
since at least 1994 the IMF has provided technical assistance. Good
governance and transparency became major policy issues in Honduras after
Hurricane Mitch. IMF documents emphasized the importance of further efforts
in those areas. More selective structural conditionality

Structural performance criteria stayed roughly the same. Number of
structural performance

criteria decreased from 5 to 2 between 1998 and 2000 as conditions on civil
service reform and cotton privatization were eliminated. After end- August
2000, Benin?s program had no structural performance criteria. However,
adopting a privatization strategy for the cotton sector is a condition for
Benin to receive debt relief.

It appears that the number of structural performance criteria has increased,
from two in November 1999 to three in April 2000 to five in October 2000.

Sources: IMF, World Bank, and recipient government documents.

(For additional information on the three countries? programs, see app. II.)
We found that many elements of the PRGF were also included in the three
countries? previous IMF programs. In those cases, it is difficult to
determine to what extent their continued inclusion reflects the announced
changes in the IMF?s approach rather than the direction the country?s
program was moving in anyway. For example, Albania and Benin have focused on
reducing poverty for several years.

 One objective of Albania?s 1997 program with the IMF was to address the
impoverishment of the Albanian population resulting from the collapse of
financial pyramid schemes, in which an estimated two- thirds of Albania?s

Chapter 3: Few Changes Evident in Three Countries? Programs

Page 43 GAO- 01- 581 IMF: New Lending Program

population was involved. 3 The program outlined a two- stage strategy for
alleviating poverty. In the short term, the government would temporarily
expand social safety nets targeted toward the poorest families, including
faster disbursements of public assistance and initiation of job- creating
public works projects, which were to be funded through cuts in spending in
other areas and external assistance. Over the medium term, poverty was to be
alleviated through high, sustainable growth driven by private sector
development.

 Benin has been moving in the direction of increased attention toward
poverty reduction throughout the 1990s. In an effort to combat Benin?s
widespread poverty, the government has undertaken a number of programs and
projects. In 1994, Benin adopted the national program on the Social
Dimensions of Development. This program was to explicitly integrate the
social dimensions of development with the macroeconomic and sectoral
policies, formulate and implement a poverty reduction action plan identified
by full participation of the most vulnerable groups, and periodically
monitor the population?s living conditions to get at the root causes of
poverty. In 1998, the National Community Development Program was to
supplement the Social Dimensions program using an approach that addressed
basic needs identified by the beneficiaries themselves.

Consistent with the goals of the PRGF program, the budgets of all three
countries indicate plans to increase spending in priority social areas such
as health and education. However, in Albania and Benin, increased government
spending in social areas has been an objective of their IMF programs since
at least 1998. Importantly, as the IMF notes, the effectiveness of higher
spending and deficit targets depends on the governments? ability to use
resources well and to obtain financing on reasonable terms, when the
governments cannot collect sufficient revenues. In all three countries, IMF
program documents have raised concerns about the countries? ability to use
resources effectively. In the past, due to the government?s weaknesses in
effectively planning, executing, and monitoring programs and budgets, Benin
has been unable to fully spend available funds, including in priority areas
such as primary health and education. Similarly, in December 2000, Albania?s
Ministry of Finance reported that in all five sectors reviewed- health,
education,

3 In a typical pyramid scheme, a fund or company attracts contributors by
offering them very high returns; these returns are paid to the first
contributors out of the funds received from those who contribute later. The
scheme is insolvent from the day it begins. The scheme flourishes initially
as news about the high returns spreads and more contributors are drawn in.

Chapter 3: Few Changes Evident in Three Countries? Programs

Page 44 GAO- 01- 581 IMF: New Lending Program

labor and social protection, transport, and public works- insufficient
capacities for policy planning and program management constrained the more
effective utilization of budgetary resources. The ministry expressed the
concern that if the substantial delays that have beset the implementation of
many donor- financed projects in recent years were to continue, donor
agencies may reduce their planned level of funding.

For all three countries, establishing national ownership of their
macroeconomic policies and targets has been challenging for several reasons.
First, the governments? ability to analyze macroeconomic issues is limited.
Second, while all three national governments have organized a participatory
process for preparing their countries? poverty reduction strategies, they
have found it difficult to determine how to include civil society in
discussing macroeconomic policies and targets. Finally, while countries?
PRGF program documents are to include information on the social impact of
economic adjustment policies, we did not find such information.

The three governments? ability to analyze macroeconomic issues is limited.
For example, although government officials in Albania told us that their
capacity to negotiate with the IMF has increased, this capacity is still
limited. A recent World Bank report concluded that the Ministry of Finance
needs to strengthen its expenditure policy and analytical capacities in
order to enhance the policy debate on public spending and the formulation of
economic policies. 4 The report further stated that the current framework
was largely based on the parameters guiding Albania?s PRGF arrangement and
that ?in the future it is expected that this framework will increasingly
reflect the government?s own forecasts and analysis.? A central bank
official in Albania told us that in 1997 a former employee developed a model
to analyze macroeconomic policies, but since his departure no one has been
able to operate or maintain the model. As for Benin, based on our
discussions with government and donor officials there, it was not evident
that the government is attempting to develop alternative macroeconomic
scenarios based on different policy options. Finally, the January 1995
policy framework paper for Honduras stated that the central bank required
assistance in financial programming,

4 See Albania Public Expenditure and Institutional Review: A Briefing Note,
World Bank/ Europe and Central Asia Region (Washington, D. C.: World Bank,
Mar. 1, 2001). Difficulties

Establishing National Ownership of Macroeconomic Framework

Limited Government Capacity

Chapter 3: Few Changes Evident in Three Countries? Programs

Page 45 GAO- 01- 581 IMF: New Lending Program

open market, and foreign exchange market operations; and the strengthening
of balance- of- payments, money, banking, and national accounts statistics.
According to IMF staff, there are Honduran government officials who are
capable of devising a macroeconomic framework and discussing macroeconomic
policies and trade- offs; however, due to limited budgets and lack of
specialists, they sometimes need external technical assistance.

All three countries we reviewed have taken initial steps to implement the
most significant change resulting from the PRGF- that macroeconomic targets
and policies will be nationally owned based on a participatory process. All
three national governments have organized a participatory process and
initiated a dialogue with civil society for preparing their countries?
poverty reduction strategies. However, many people said that they were
unsure how to use this dialogue to address the country?s macroeconomic
targets and policies other than in discussions involving the budget.
Determining how, in practice, to include civil society in discussing
macroeconomic policies and targets has been challenging due to the
following.

 Significant time and resources are needed to establish a participatory
process for addressing the numerous elements of the poverty reduction
strategy. For example, in Albania, identifying civil society representatives
involved significant effort, because civil society does not have a history
of being organized and, according to many people we spoke with, remains
fractured. Yet, with the help of the Carter Center, civil society has begun
to organize itself by sector, provide input to working groups in four key
sectors- agriculture/ rural, education, health, and labor and social
affairs- and assist government working groups in developing and writing
sector- specific strategies that will prioritize needs and feed into the
government?s budget. 5

 Limited capacity exists in- country for addressing complex macroeconomic
issues. Although there is ample donor assistance in Benin, we were not told
of any research organizations that are assisting Benin in the development of
its macroeconomic framework. Donors in Benin indicated that they prefer to
stand back and let the government and the people

5 In 1982, former U. S. President Jimmy Carter and Rosalynn Carter founded
the Carter Center, a nonprofit nongovernmental organization that seeks to
promote peace and health worldwide. Unclear How to Involve

Civil Society in Discussing Many Macroeconomic Issues

Chapter 3: Few Changes Evident in Three Countries? Programs

Page 46 GAO- 01- 581 IMF: New Lending Program

decide for themselves how to integrate the sectoral priority areas, such as
health and education, into the overall poverty reduction strategy.
Government and donor officials in Honduras told us that there are no
independent think tanks analyzing macroeconomic issues. 6

In practice, civil society participation in the three countries has focused
on discussing poverty reduction goals and priorities for the composition and
level of spending- particularly within specific sectors- to achieve those
goals. However, despite the importance of these discussions, they only
impact one element of the macroeconomic framework.

Under the PRGF, countries? program documents are to include information on
the social impact of adjustment; however, we did not find such information.
Such information is to help inform the range of policies considered and the
measures designed to ease any temporary adverse impacts. In 1989, the IMF
and the World Bank considered inclusion of measures for mitigating adverse
effects on the most vulnerable groups to be an essential ingredient for
successful structural adjustment. Unlike the lack of knowledge on trade-
offs and linkages discussed in chapter 2, such an analysis is doable and,
according to the IMF, the IMF and the World Bank have been undertaking such
analyses for some time and including this information in country- specific
programs. Although we found limited evidence of such analyses in pre- PRGF
program documents, we did not find such information in the documents for the
three countries? current programs. This information is not difficult to
include in program documents relatively quickly, since it does not require
the preparation of the poverty reduction strategy and was called for under
the IMF?s previous program. Despite the limited technical capacities of
member governments and civil society, such information is valuable in
helping to inform the discussion on the effects of certain policies,
especially on the poor

6 In commenting on a draft of this report, IMF headquarters staff said that
there is at least one independent, donor- funded entity in Benin (Cellule
d?analyse politique ï¿½conomique)

and two groups in Honduras (the economics school at the public university
and the Chamber of Commerce) that engage in macroeconomic analysis. Lack of
Information on the

Social Impact of Adjustment Limits Discussion

Chapter 4: The Effects of the New Program on Graduation and the Role of the
IMF

Page 47 GAO- 01- 581 IMF: New Lending Program

If the changes the IMF announced for its lending program to its poorest
members work to actually improve the overall effectiveness of a country?s
development program, the likelihood of earlier graduation from the Fund?s
program could increase. However, the actual impact of these changes on
economic growth is unknown at this time and, as previously discussed, there
are many challenges to achieving national ownership of effective
macroeconomic policies. Nonetheless, our analysis shows that most of the
current recipients of the IMF?s concessional assistance will require strong,
sustained economic growth to reach the point where they would be considered
for graduation from concessional Fund assistance. To reach this point within
15 years, we found that the 32 countries that borrowed from the IMF?s
concessional facility in 2000 must average a real per capita income growth
rate in excess of 6 percent annually during that entire period. This
significantly exceeds the countries? average growth rate of negative 1
percent over the last 15 years. Unlike the IMF?s nonconcessional facilities
that are to lend resources based on short- term conditions, the concessional
facility does not assume that countries will have only a temporary need for
concessional assistance. During the last 15 years, the IMF has functioned as
a long- term lender that supports countries? reform programs, and
indications are that it intends to retain this role.

Since the beginning of the IMF?s Structural Adjustment Facility in 1986, six
countries have graduated from program eligibility. Three of those countries
(the Dominican Republic, the Philippines, and St. Kitts and Nevis) were
graduated from eligibility for concessional lending in 1995. According to
IMF documents, this decision was based on three factors: (1) the countries
were no longer eligible for IDA funds; (2) their balance- ofpayments and
debt positions had improved substantially; and (3) they had not used, nor
were they expected to use, PRGF resources in the near future. In 2000,
China, Egypt, and Equatorial Guinea also graduated from PRGF eligibility,
with similar considerations given for that decision. In the case of China,
although its per capita income was still below the IDA Chapter 4: The
Effects of the New Program

on Graduation and the Role of the IMF Strong, Sustained Economic Growth
Essential to Reach Consideration for Graduation

Graduation Eligibility Influenced by a Country?s Income Level

Chapter 4: The Effects of the New Program on Graduation and the Role of the
IMF

Page 48 GAO- 01- 581 IMF: New Lending Program

threshold (its per capita income equaled $750 in 1999), it graduated from
both IDA and PRGF eligibility because it was considered creditworthy, with a
substantial foreign exchange reserve and a sustainable external debt
obligation. Some countries, such as Bolivia and Macedonia, have remained
eligible for borrowing under the IDA and PRGF even after their per capita
income exceeded this threshold because they were determined to lack
creditworthiness under the World Bank?s market- based lending facility. 1
Macedonia is considered a candidate for future graduation from both IDA and
PRGF eligibility, according to the Fund.

Countries that are only eligible for IMF nonconcessional resources can be
added to the PRGF eligibility list if their economic situations worsen.
According to IMF documents, as of December 2000, two countries (Uzbekistan
and Turkmenistan) had per capita income levels below the IDA threshold and
were under consideration for IDA eligibility and could potentially be made
PRGF- eligible in due course. Indonesia?s per capita income has also fallen
substantially over the last several years, and Indonesia is now deemed
eligible for IDA resources. However, according to IMF documents, Indonesia
is not considered eligible for PRGF resources at this time.

If the changes announced by the IMF for its lending program to its poorest
members work to improve the overall effectiveness of a country?s development
program, the likelihood of earlier graduation could increase. According to
an IMF official, the poverty reduction strategy is likely to improve a
country?s overall economic growth rate by having the country?s national
dialogue focus increased attention on ways to accelerate its growth rate and
through increased coordination among the donors. However, the actual impact
of these changes on economic growth is unknown at this time and, as
previously discussed, there are many challenges to achieving national
ownership of effective macroeconomic policies.

To better understand the likelihood that current borrowers will graduate
from eligibility from the IMF?s concessional lending facility, we analyzed
borrowers? per capita income growth rates in the previous 15 years and

1 This facility is the World Bank?s International Bank for Reconstruction
and Development loan facility. Most Current Borrowers

Require Strong, Sustained Economic Growth to Become Eligible for Graduation
Within 15 Years

Chapter 4: The Effects of the New Program on Graduation and the Role of the
IMF

Page 49 GAO- 01- 581 IMF: New Lending Program

the impact of projected economic growth over the next 15 years. 2 We
estimated per capita income growth rates based on forecasts provided by
Standard & Poor?s DRI for each of the 32 countries that received assistance
from the PRGF in 2000. 3 As is demonstrated in table 3, the DRIprojected
growth rates are considerably greater than the growth rates that most of
these countries experienced over the previous 15 years.

2 We focused on countries that received PRGF disbursements in calendar year
2000. Our analysis focused only on per capita income in our consideration of
graduation. This was due to three reasons: (1) per capita income is the
primary factor in determining eligibility; (2) forecasts of creditworthiness
and balance- of- payment positions would be very speculative and difficult
to implement; and (3) as discussed previously, although some countries
graduate prior to reaching that threshold, others do not graduate until they
have exceeded that level for some time. Thus, we consider the IDA per capita
income threshold to be a reasonable proxy for graduation eligibility. We
used the 1998 IDA threshold of per capita income of $895 or less, which
corresponds to the 1998 per capita income levels used in the analysis.

3 DRI is a unit of Standard & Poor?s, which is a division of The McGraw Hill
Companies. It is a leading provider of industry and economic data,
forecasting, and consulting services. Its World Forecast Database includes
historical and forecast data for 170 countries.

Chapter 4: The Effects of the New Program on Graduation and the Role of the
IMF

Page 50 GAO- 01- 581 IMF: New Lending Program

Table 3: Projected and Previous Annual Growth Rates for 32 PRGF Borrowers
Country name

1998 per capita

GNP Forecasted

real per capita GDP growth rates,

2001- 2015 a (in percent)

Projected per capita GNP in 15 years, using

forecasted real per capita GDP growth

rates (in 1998 U. S. dollars)

Number of years required to reach IDA

threshold of $895, using forecasted real

per capita GDP growth rates

Rate of growth required to reach

IDA threshold of $895 per capita GNP in 15 years

(in percent) Real per

capita GNP growth rates

over past 15 years (in percent)

Real per capita GNP

growth rates over past 5 years

(in percent)

Albania $810 4.57% $1,582.5 2 0.67% 2.69% 4.67% Benin 380 2.26 531.2 38 5.
88 0. 22 2. 38 Bolivia 1,010 3.31 1,645.0 -4 -0.80 1.61 2.23 Burkina Faso
240 2.19 332.3 61 9. 17 0. 85 2. 15 Cambodia 260 4.14 477.5 30 8. 59 1. 96
1. 52 Cameroon 610 1.88 806.3 21 2. 59 -4.02 0.81 Chad 230 2.19 318.4 63 9.
48 0. 17 1. 40 Djibouti 789 0.91 904.2 14 0. 84 -4.12 -3.36 Gambia, The 340
1.87 448.7 52 6. 67 -0.57 -0.71 Ghana 390 2.94 601.9 29 5. 69 1. 49 1. 51
Guinea- Bissau 160 1.61 203.3 108 12.16 0.44 -3.15 Guyana 780 3.12 1,237.3 4
0.92 2.82 7.44 Honduras 740 1.78 964.0 11 1. 28 0. 55 0. 91 Kenya 350 2.73
524.0 35 6. 46 0. 16 1. 82 Kyrgyz Republic 380 3.20 609.9 27 5. 88 -5.71
-1.44 Macedonia, FYR 1, 290 4.28 2,417.8 -9 -2.41 -1.40 1. 10 Madagascar 260
1.29 315.3 96 8. 59 -1.28 -0.24 Malawi 210 2.92 323.2 50 10.15 0.25 2.80
Mali 250 2.18 345.4 59 8. 87 -0.10 1.55 Mauritania 410 1.96 548.6 40 5. 34
0. 11 2. 19 Moldova 380 3.50 636.8 25 5. 88 -8.29 -7.28 Mongolia 380 3.44
631.4 25 5. 88 -1.20 2.18 Mozambique 210 4.68 417.0 32 10.15 2.39 6.42
Nicaragua 370 2.96 573.2 30 6. 07 -2.45 5.70 Niger 200 0.41 212.6 366 10.51
-1.69 0. 60 Rwanda 230 -0.63 209.3 b 9.48 -3.62 0. 96 Sï¿½o Tomï¿½ and Principe
270 0.87 307.4 138 8.32 -1.19 -0.69 Senegal 520 2.78 785.1 20 3. 69 -0.09
2.64 Tajikistan 370 2.67 549.4 34 6. 07 -11.89 -6.78 Tanzania 220 3.02 343.6
47 9. 81 0. 27 1. 22 Uganda 310 2.71 463.2 40 7. 32 2. 70 4. 92 Zambia 330
2.32 465.7 44 6. 88 -1.07 -1.04 Average $427 2.50% $648 59 6.13% -0.94%
1.08% Median $360 2.69% $528 34 6.26% .01% 1.45%

Chapter 4: The Effects of the New Program on Graduation and the Role of the
IMF

Page 51 GAO- 01- 581 IMF: New Lending Program

Legend FYR = Former Yugoslav Republic GNP = Gross national product

Note: Data from World Bank?s World Development Indicators 2000, CD- ROM. The
past 15- year growth rate is based on real per capita GNP from 1983 to 1998,
except for the following countries for which the data were available for
only the following years; Albania - last 8 years; Bolivia - last 10 years;
Cambodia - last 11 years; Djibouti - based on DRI GDP data, not GNP; Kyrgyz
Republic- last 12 years; Macedonia - last 6 years; Moldova - last 6 years;
Sï¿½o Tomï¿½ and Principe - last 12 years; Tajikistan - last 12 years; and
Tanzania - last 10 years. a Forecasted real per capita GDP growth rates are
calculated from Standard & Poor?s DRI?s forecasted

real GDP and population growth rates. b Given the projected negative growth
rate for Rwanda over the next 15 years, its projected GNP per capita does
not converge toward the IDA threshold. For the purpose of calculating an
average duration until graduation, we assigned Rwanda the same value (366)
as the next highest estimate.

Source: GAO analysis using data from the World Bank and Standard & Poor?s
DRI.

Although the DRI- projected country growth rates are generally higher than
prior growth rates, they are only sufficient to bring four countries-
Albania, Djibouti, Guyana, and Honduras- to per capita income levels that
would exceed the IDA eligibility threshold by the end of the 15- year
period. 4 An additional 26 countries would still have per capita income
levels that were below the graduation threshold. Given that the current per
capita income levels of these countries is generally quite low (an average
of $427), the average number of years required for these countries to reach
the graduation threshold at their projected per capita income growth rates
is 59, with Niger requiring 366 years.

To reach the eligibility threshold within 15 years, these countries must
average a real per capita gross national product growth rate in excess of 6
percent annually over that entire period. The poorest country, GuineaBissau,
must average a real growth rate in excess of 12 percent a year to reach the
threshold for graduation within 15 years. Such high, sustained

4 As of the end of 2000, the per capita income level of Bolivia and
Macedonia already exceeded the IDA eligibility threshold.

Chapter 4: The Effects of the New Program on Graduation and the Role of the
IMF

Page 52 GAO- 01- 581 IMF: New Lending Program

growth levels would greatly exceed the rates projected by DRI and those that
have prevailed in the past. 5

The underlying problems that IMF assistance is designed to help address in
poor countries often take a long time to resolve. In an August 2000 speech,
the Managing Director of the IMF said that the efforts to combat poverty
?will inevitably be an arduous and often lengthy process.? According to an
IMF official, the IMF is likely to continue to provide a lowincome country
with resources as long as it has a need for such resources, resources are
available to be lent, and the country pursues effective policies.

Unlike the IMF?s nonconcessional resources, there is no presumption under
the PRGF that countries will have only a temporary need for concessional
assistance. Each program under the PRGF is scheduled to last for 3 years,
but countries can receive follow- on programs over time, if needed. The
purpose of PRGF assistance is to strengthen a country?s balance- of-
payments position and foster lasting economic growth. This purpose extends
beyond temporary balance- of- payments difficulties, which are the focus of
IMF nonconcessional assistance. They also allow the IMF wide latitude to
lend to poor countries, given the countries? underlying vulnerabilities and
low standards of living. For example, in recent years Benin has had a very
stable macroeconomic environment, with low inflation, a balance- of-
payments surplus, low budget deficits, and good economic growth. Within this
healthy macroeconomic context, in July 2000 the IMF and Benin negotiated
another program for an additional 3 years, with the objectives of achieving
high and sustainable economic

5 It should be noted that this average projected level of growth is actually
lower than the levels called for by the World Bank and the IMF for many of
these countries over the next 15- 20 years. For example, in a speech given
in February 2001, the Managing Director of the IMF endorsed a plan proposed
by African leaders that called for a sustained growth rate in Africa of at
least 7 percent. Such high growth levels are needed to achieve the
reductions in poverty called for by the donors in Copenhagen in 1995. To the
extent that such high growth rates are achieved, graduation from PRGF
eligibility would be realized sooner. The IMF Has Been a

Long- term Lender to Low- income Countries

Chapter 4: The Effects of the New Program on Graduation and the Role of the
IMF

Page 53 GAO- 01- 581 IMF: New Lending Program

growth and reducing poverty while maintaining macroeconomic stability during
2000- 03. 6

As shown in table 4, the 32 countries that received IMF concessional
assistance in 2000 also received assistance for about 9 years since the
inception of the Structural Adjustment Facility in 1986 and for about 7 of
the last 10 years. 7

Table 4: Number of Years 32 Countries Received IMF Resources Country

Number of years countries received IMF resources,

1986- 2000 Number of years countries

received IMF resources, 1991- 2000

Albania 8 8 Benin 9 8 Bolivia 13 9 Burkina Faso 9 9 Cambodia 5 5 Cameroon 10
8 Chad 10 7 Djibouti 6 6 Gambia, The 9 4 Ghana 12 7 Guinea- Bissau 8 6
Guyana 11 10 Honduras 8 7 Kenya 8 5 Kyrgyz Republic 8 8 Macedonia, Former
Yugoslav Republic 7 7 Madagascar 10 5 Malawi 11 8 Mali 13 9 Mauritania 13 8
Moldova 7 7 Mongolia 8 8

6 According to IMF staff, the new PRGF arrangement was necessary for Benin
to benefit from full debt relief. In addition, given the vulnerability of
the macroeconomic framework to frequent internal and external shocks, an
IMF- supported program was considered necessary to consolidate the achieved
macroeconomic stability and for the Fund to play its catalytic role in
securing other donors? assistance.

7 During 1986- 2000, 28 of these countries received nonconcessional IMF
assistance in addition to concessional assistance.

Chapter 4: The Effects of the New Program on Graduation and the Role of the
IMF

Page 54 GAO- 01- 581 IMF: New Lending Program

Country Number of years countries

received IMF resources, 1986- 2000

Number of years countries received IMF resources,

1991- 2000

Mozambique 13 9 Nicaragua 5 5 Niger 10 5 Rwanda 6 6 Sï¿½o Tomï¿½ and Principe 2
1 Senegal 13 8 Tajikistan 5 5 Tanzania 11 7 Uganda 13 9 Zambia 5 4 Average
8. 9 6.8 Median 9 7

Source: GAO analysis of IMF data.

Fourteen of these countries received IMF assistance for 10 or more of the
last 15 years, with Guyana receiving assistance continually over the last 10
years. In August 2000, the Managing Director of the IMF stated the
institution?s commitment to remain engaged in the world?s poorest countries.
Over the last 15 years, the IMF has functioned as a long- term lender to
these countries to support their reform programs, and indications are that
it intends to continue to do so.

Appendix I: Identifying the Impact of Different Policy Choices Is Difficult

Page 55 GAO- 01- 581 IMF: New Lending Program

As part of the goal of establishing national ownership of the macroeconomic
framework, the government and civil society may seek to discuss the impact
of different policy options, to help establish a consensus for reform.
However, as discussed in chapter 2, meaningful national dialogue on the
range of policies that are available is constrained by the limited knowledge
of all parties, including the International Monetary Fund (IMF), on how
different policies actually affect elements of the macroeconomic framework.
To help illustrate these complexities, we describe the possible impacts of a
policy that has been prominently mentioned- increased spending on education
and health- on the goals of economic growth and poverty reduction in the
context of maintaining macroeconomic stability.

As part of the change from the Enhanced Structural Adjustment Facility
(ESAF) to the Poverty Reduction and Growth Facility (PRGF), the IMF called
for budgets of countries to be more ?pro- poor.? However, it is difficult to
evaluate the impact of this strategy, according to the IMF, because of the
limited knowledge of the relationship between expenditures and actual
results. In addition, for the policy to be effective, it often requires
supporting structural and governance reforms that should also be taken into
account. One example of increased pro- poor budgets that has been discussed
is a greater allocation of spending to the education and health sectors.
While such a policy choice may be appropriate for a country to pursue, the
actual impact on economic growth, poverty reduction, and macroeconomic
stability is uncertain, as demonstrated in figure 3. Appendix I: Identifying
the Impact of

Different Policy Choices Is Difficult The Impact of Increased Spending on
Education and Health Is Uncertain

Appendix I: Identifying the Impact of Different Policy Choices Is Difficult

Page 56 GAO- 01- 581 IMF: New Lending Program

Figure 3: Complex Linkages Between Increased Spending on Education and
Health and Goals

Source: GAO analysis of IMF, World Bank, and academic documents.

Appendix I: Identifying the Impact of Different Policy Choices Is Difficult

Page 57 GAO- 01- 581 IMF: New Lending Program

The impact of increased education and health spending on poverty reduction
depends chiefly on two factors: the effectiveness of the increased spending
and its allocation. As discussed in chapter 3, poor countries have had
difficulties in recent years in effectively utilizing resources, including
those provided by donors. If the money is not well spent or not spent at
all, the impact on poverty reduction will be negligible. For example, if the
money is used to build new schools and clinics, but there are not enough
teachers and doctors to fill them, then the improvement of health and
education will likely be small. In addition, despite the stated goal of
increased expenditures to the poor, the increased expenditures could be
allocated instead to services primarily used by the middle class and the
wealthy. In such a case, relative poverty may in fact rise as income
inequality worsens. However, if the concerns over effectiveness and
allocation are sufficiently addressed, the increased spending could work
successfully to influence indicators of poverty, such as educational
attainment and child mortality. Poor countries have confronted these
concerns for some time, and the actual impact of increased spending on
poverty reduction remains difficult to determine.

The impact of increased spending on economic growth is even more difficult
to determine. Whereas a healthier and better- educated population may
contribute to increased economic growth, the actual benefit is uncertain,
since this reallocation of resources may divert spending away from other
uses that would have a clearer and more immediate impact on productivity.
The benefits of increased health and education spending may take years to
materialize and require complementary resources (such as land and capital)
to achieve the greatest impact on productivity. In simple terms, a country
is not likely to benefit much from greatly increasing the productivity of
its labor force if there is a scarcity of opportunity for laborers to
utilize their skills effectively. As shown in chapter 4, most PRGF countries
have had little or no growth in the previous 15 years. Thus, the
identification of investments that would be most productive for these
economies has proven very challenging.

As discussed in chapter 2, the consideration of policy choices should also
account for their impact on macroeconomic stability. Countries that
experience macroeconomic instability have tended to have low or even
negative economic growth, which also negatively impacts their effort to
reduce poverty. In assessing the increased expenditures on education and
health, the issue is whether this budgetary allocation is consistent with
maintaining macroeconomic stability. However, as discussed in chapter 2,
Impact on Economic

Growth Impact on Macroeconomic Stability

Appendix I: Identifying the Impact of Different Policy Choices Is Difficult

Page 58 GAO- 01- 581 IMF: New Lending Program

it is not possible to establish a definitive boundary between macroeconomic
stability and instability.

The increased expenditure on education and health could raise the risk of
macroeconomic instability if it negatively impacts the inflation rate, debt
sustainability, and the deficit. If these increased expenditures represent
borrowed resources, then all three factors could be negatively impacted. We
found in a recent report that the efforts to reallocate Heavily Indebted
Poor Country debt relief to poverty spending could only increase spending if
the countries borrowed these resources. 1 Efforts to finance this increased
spending through donor- financed concessional borrowing are likely to
present the least threat to macroeconomic stability in the short run.
However, as we found in the recent report, unless countries achieve strong,
sustained economic growth, they are likely to have problems with debt
sustainability in the future. Thus, these macroeconomic stability concerns
have to be integrated into the trade- offs when assessing the choice of the
best policy.

Given the challenges that confront government and civil society in choosing
the correct policies to achieve the country?s goals, it is not clear

?how? or ?if? a specific macro policy instrument, such as targeted
expenditures on the poor, will lead to the ultimate goals- growth and
poverty reduction in the context of macroeconomic stability. The difficulty
of specifying imperfectly understood linkages, the numerous possible
outcomes, and the challenges in assessing the trade- offs involved make it
difficult to choose the appropriate macro policy stance (e. g., a larger
fiscal deficit) and to select the specific macro policy instrument (e. g.,
targeted expenditures on the poor). Given the complexity of the links
between macroeconomic policies and poverty, the IMF and the World Bank have
suggested that to assist countries in assessing these trade- offs and the
distributional implications of their macroeconomic policies, it would be
particularly useful for them to have a quantitative model that identifies
the critical relationships. These models, however, are presently only at a
nascent stage of development.

1 See Developing Countries: Debt Relief Initiative for Poor Countries Faces
Challenges (GAO/ NSIAD- 00- 161, June 29, 2000). The Need for Improved

Economic Models

Appendix II: Key Events and IMF Programs in Albania, Benin, and Honduras

Page 59 GAO- 01- 581 IMF: New Lending Program

Albania, Benin, and Honduras have borrowed from the IMF?s concessional
lending facilities since at least the early 1990s. They each have
experienced internal or external shocks such as civil unrest, declines in
the world price of a major export, and natural disasters. Tables 5- 7
provide information on key events and the IMF- supported programs in
Albania, Benin, and Honduras. Appendix II: Key Events and IMF Programs

in Albania, Benin, and Honduras

Appendix II: Key Events and IMF Programs in Albania, Benin, and Honduras

Page 60 GAO- 01- 581 IMF: New Lending Program

Table 5: Key Events in Albania, 1991- 2001

IMF Board approved new, 3- year ESAF arrangement of about $59.2 million in
July 1993. IMF staff said Albania?s performance was impressive during the
first 2 years but deteriorated subsequently so that an agreement on the
third annual arrangement could not be reached. IMF Board approved a

1- year stand- by (nonconcessional) arrangement of about $28. 2 million.

Key social, economic, and political

events

Signs of an impending crisis: About two- thirds of Albania?s population
contributed to pyramid schemes, totaling almost half of the country?s GDP. a

Albania made little progress in structural reforms. Growth remained rapid.
However, it was supported by pyramid schemes and election- driven loose
fiscal policy that permitted a near tripling of annual inflation. Steep
macroeconomic

recovery. Average GDP growth neared double digits, inflation fell to single
digits, and external imbalances fell sharply. Impressive economic

performance reflected reforms such as privatization of agriculture and
financial discipline. Yet, limited progress was made in reforming the
banking sector and building the government?s institutional capacities.
Democratically elected

government took office, moved quickly to restore civil order, and embarked
on an ambitious economic reform program. Collapse of Albania?s Communist

regime was accompanied by economic and political chaos. Deteriorating living
conditions and growing awareness of political alternatives contributed to
widespread social unrest, public disorder, and severe disruptions in
government functions and institutions. Albania became dependent on emergency
food aid.

Albania joined the IMF and the World Bank.

IMF program 1993- 95 1996 1992 1991

A

Appendix II: Key Events and IMF Programs in Albania, Benin, and Honduras

Page 61 GAO- 01- 581 IMF: New Lending Program

a According to IMF staff, in a typical pyramid scheme, a fund or company
attracts contributors by offering them very high returns; these returns are
paid to the first contributors out of the funds received from those who
contribute later. The scheme is insolvent from the day it begins. The scheme
flourishes initially as news about the high returns spreads and more
contributors are drawn in.

Source: GAO review of IMF and Albanian government documents.

199

20, 000 refugees fled escalating conflict in Kosovo for northern Albania.

Civil unrest occurred in September after a prominent opposition politician
was shot; the Prime Minister resigned. New government reaffirmed the
economic policies of the IMFsupported program. Collapse of pyramid schemes

triggered deep economic and social crisis, bringing the economy to a
standstill. Albania descended into near civil war in which about 2,000
people were killed.

The government resigned. The interim government made impressive progress in
restoring order and stabilizing the economy. New government elected in July.

Crisis in Kosovo erupted when the NATO campaign against Yugoslavia began.
Refugees poured into Albania; by mid- May about 430,000 largely destitute
refugees had flooded into Albania.

In the fall, an internal power struggle forced the Prime Minister to resign;
the transfer of power was smooth.

In May 2000, the government completed its interim poverty reduction
strategy. In November, the government launched a participatory process for
developing its poverty reduction strategy .

Local elections in October were relatively peaceful but remain contested by
opposition party.

1997 1998 1999 2000- 01

IMF approved emergency post- conflict assistance totaling about $12.1
million under a 6- month program.

Immediate priorities: restore security, achieve macroeconomic stabilization,
and rebuild institutional capacity to help resume growth. Resuming
sustainable, job- creating growth was a key priority. Pyramid- scheme crisis
impoverished the population; government developed short- term and medium-
term poverty alleviation strategies.

Risks to program: Authorities? limited capacity to formulate and implement
macroeconomic and structural measures and the difficulty of projecting key
economic variables in uncertain circumstances.

IMF approved new, 3- year ESAF arrangement of about $47.9 million.

Main objectives: restart rapid growth and reduce inflation further-- to
generate employment and reduce poverty-- and reduce the current account
deficit to levels consistent with mediumterm viability.

Reforms in 1999 to allow for increased spending on government priorities-
infrastructure and social services, especially health and education.
Government reaffirmed short- and mediumterm poverty alleviation strategies.

Despite disturbances, Albania generally met the macroeconomic targets,
although progress on structural reforms was delayed. Refugees stretched the
already thin administrative capacity of the government, risking a breakdown
in law and order and social stability. Refugees placed considerable strain
on the social and economic infrastructure, budget, and balance of payments.
In response, the IMF increased ESAF resources by about $13. 3 million.

Fiscal deficit remained on track, with disappointing tax collection (partly
due to continued problems in customs administration) offset by significant
underspending, particularly on public investment.

Albania met all performance criteria between January 2000 and January 2001.

Cautious policies and generous external assistance allowed the economy to
weather the Kosovo refugee crisis. Economy still suffers from acute
structural weaknesses (i. e., no reliable electricity); poor governance; and
problems enforcing law and order, with graft reportedly remaining
widespread.

Significant progress made in improving tax collection and implementing
structural reforms, including two major privatizations, in 2000.

Legend GDP = Gross domestic product NATO = North Atlantic Treaty
Organization

A

Appendix II: Key Events and IMF Programs in Albania, Benin, and Honduras

Page 62 GAO- 01- 581 IMF: New Lending Program

Table 6: Key Events in Benin, 1960s- 2001 Key social, economic, and
political

events

Adjustment strategy- strengthened by currency devaluation-- increased the
profitability of cotton but also led initially to slowdown in economic
activity and jump in the inflation rate to 54 percent.

Country remained one of the poorest in the world on a per capita income
basis. Management deficiencies cited in ministries of education and health.

Insurance and cotton ginning sectors opened to private sector investment.
GDP growth resumed in 1990,

the inflation rate was low, the fiscal balance was in surplus, confidence in
the banking system was restored, and reserves increased. Peaceful transfer
of power in

1991 presidential elections. Significant effort to contain the wage bill
through voluntary departure from civil service. During this period, under

President Kï¿½rï¿½kou, Benin was a Marxist- Leninist state. Policies pursued in
the 1980s led to economic stagnation characterized by significant
imbalances, debt, and a severe financial crisis.

By 1986, Benin?s health care system ranked among the poorest in Africa. The
banking system collapsed in 1988.

Fundamental political changes occurred as Benin abandoned Marxist policies
by the end of 1989, becoming a multiparty democracy in the following year.
Benin became a leading

innovator in primary health care, designing-- without outside help-- a
primary health care system involving rural populations and covering the
entire country, according to a World Bank report.

IMF program IMF Board approved a 3- year ESAF arrangement of about $72. 5
million in 1993. Thrust of adjustment programs was to develop a market-
based economy: reduce public sector, promote private sector, improve public
finances, reform taxes, reform regulatory system, liberalize prices, and
privatize state- owned entities. IMF Board approved a

3- year structural adjustment loan in 1989 of about $28.1 million. Benin
joined the IMF and World

Bank in 1963. IMF approved annual

loan under ESAF.

A 1994- 95 1990- 93 1980s 1960s- 70s

Appendix II: Key Events and IMF Programs in Albania, Benin, and Honduras

Page 63 GAO- 01- 581 IMF: New Lending Program

Source: GAO review of IMF and Benin government documents. Government adopted
national

plan for combating poverty, 1998- 2002.

Government unable to fully spend budget appropriations.

Opposition encountered for civil service reform and privatization of the
cotton sector and utilities.

Severe drought in early 1998 impaired cotton production and resulted in
power shortages, which negatively impacted economic performance in 1998.
Mathieu Kï¿½rï¿½kou elected President.

His main concern was to devise growth- enhancing policies to allow for a
substantial and visible poverty alleviation and improvement in basic
services.

Petroleum- distributing sectors opened to private sector.

Given its heavy reliance on cotton production and exports, Benin was
vulnerable to trade risks and environmental deterioration.

Legislative elections in March 1999 resulted in opposition parties gaining a
slight majority.

Deterioration in cotton sector due to drop in international prices,
compounded by a fall in production and weak management.

Banking system remained heavily exposed to cotton sector; thus, vulnerable
to adverse developments in cotton sectors.

PRSP process launched in November 1999 with IMF and World Bank workshop for
government officials.

In June 2000, the government completed its interim poverty reduction
strategy paper and was declared eligible for debt relief of $265 million
(net present value) under the HIPC Initiative.

Low absorptive capacity continued to lead to social and public investment
underspending.

Government partially reformed cotton sector, but progress on privatization
was slow.

Concerns about grave financial problems and mismanagement related to
petroleum privatization uncovered by audits in two banks.

Presidential elections held March, 2001.

Benin?s economic and financial performance remained satisfactory. Surplus in
overall balance of payments due to a high level of external financial
assistance.

IMF approved 3- year PRGF arrangement for about $35. 7 million in July 2000.

Key objectives: medium- term program aimed at achieving high and sustainable
economic growth of 5- 6 percent, maintaining financial stability, and
reducing poverty. First review under the PRGF in December 2000 noted that
structural reforms were delayed. Progress made in implementing

structural measures, but there were delays.

Fall in cotton production and export price, lack of market flexibility, and
weak management capacity were expected to have broad repercussions for rest
of economy, including expected reduction in GDP growth.

Policies: Fiscal still on track. Liberalize cotton ginning enterprise,
complete civil service reform, and continue to reduce public sector
involvement in utility companies and the main port. Focus on poverty
alleviation and social policies by increasing spending for health and
education. In December 1998, the IMF

concluded that Benin had satisfactorily implemented the program despite
protracted discussions due to disagreement on key reforms and delays in
implementing structural reforms, notably in the liberalization of the cotton
sector and the divestiture program. 1998- 99 program: emphasis

continued from previous program, which aimed at increased spending in social
sectors and public investment. Structural policies focused on strengthening
public resource management, civil service wage reform, privatization,
expanding social programs, and reducing poverty. IMF Board approved 3- year
ESAF

arrangement of about $39.5 million. Key objectives: raise economic growth to
5. 5 percent, limit inflation to 3 percent, attain sustainable balance of
payments, and alleviate poverty.

According to IMF staff, Benin made progress in several areas including a
reduction in domestic and external financial imbalances, the acceleration of
growth, and increased private and public investment

Legend GDP = Gross domestic product HIPC = Heavily Indebted Poor Countries
PRSP = Poverty Reduction Strategy Paper

1997- 98 1999 2000- 01 1996 A

Appendix II: Key Events and IMF Programs in Albania, Benin, and Honduras

Page 64 GAO- 01- 581 IMF: New Lending Program

Table 7: Key Events in Honduras, 1992- 2000 Key social, economic, and
political

events IMF program

Economic situation improved with growth rate of about 4. 5 percent, but the
inflation rate declined only slightly to 27 percent. Economic activity
expanded

with a growth of real GDP of 6 percent, but the overall financial situation
deteriorated sharply in the run- up to the presidential elections in
November. Financial imbalances widened sharply, resulting in the fiscal
deficit reaching 10.7 percent of GDP. Volume of coffee and banana exports
(Honduras?s main export crops) declined. The inflation rate grew to 10. 3
percent by June 1993.

Economic situation deteriorated further in 1994. Economic activity was
affected adversely by major electricity shortages (partly related to a
drought) and a reduction in construction activity. Plunging world coffee
prices and a severe drought that damaged agriculture also contributed to
decline in economic activity. Government authorities

introduced a medium- term economic program for 1992- 95 that aimed to
consolidate the stabilization effort and deepen structural reforms initiated
in 1990. Remaining interest subsidies were eliminated, and the
liberalization of price, import, and export controls in agriculture and
forestry improved resource allocation in these sectors. Economic activity
recovered (real GDP rose by 5.6 percent), with inflation declining sharply
to 6. 5 percent, and structural reform efforts continued. Fiscal position
weakened, however.

Key structural policies, including progress toward the privatization of
public utilities, recapitalization of the central bank, and strengthening
the finances of the social security system, were delayed or deferred.

Real GDP growth remained at 3. 7 percent, while the annual inflation rate
was 26 percent in December.

IMF Board approved new, 3- year ESAF program of about $57 million and
disbursed about $9. 6 million.

IMF Board approved second ESAF disbursement of $9. 5 million in July 1993.

IMF Board approved second annual ESAF arrangement for about $30.9 million in
January, increased the total access under the 3- year arrangement to about
$82. 4 million, and extended the ESAF program through July 1997.

Negotiations on a third annual arrangement were not concluded, because
government authorities were not prepared to enter into commitments that
would extend into the presidential primaries in late 1996.

1992 1993- 94 1995 1996 A

Appendix II: Key Events and IMF Programs in Albania, Benin, and Honduras

Page 65 GAO- 01- 581 IMF: New Lending Program

Source: GAO review of IMF and Honduras government documents. Macroeconomic
performance

improved, but there was little progress in carrying out structural reforms,
including privatization of electricity distribution and issuance of bids for
half of shares of HONDUTEL (the state- owned telecommunications company).
Real GDP growth was 5.1 percent. The annual inflation rate declined to 13
percent in December.

New President took office in January and declared that the government?s main
objective was to reduce poverty through policies aimed at achieving faster
economic growth and providing broader and more effective social services.
Hurricane Mitch hit in October, causing the deaths of close to 7,000 people,
displacing nearly 2 million people, destroying several towns and villages,
and damaging infrastructure and the economy. Estimates of the damage to the
economy indicate that direct losses of inventories and fixed assets amounted
to about 100 percent of annual GDP.

Reconstruction efforts and improvements in external environment helped bring
about a recovery in economic activity during the second half of the year,
containing the decline in real GDP to 1.9 percent for the year as a whole.
The end- year inflation rate fell to 10.9 percent, the lowest since 1992. In
December, Honduras was declared eligible for debt relief under the Heavily
Indebted Poor Countries Initiative. The government initiated work on its
poverty reduction strategy.

Recovery from the hurricane continued. A pick- up in key exports (bananas,
coffee, and maquila) and agricultural production, especially during the
second quarter, helped lift real GDP growth to nearly 5 percent in 2000.

The government completed its interim poverty reduction strategy in March and
began a participatory process for developing a full poverty reduction
strategy.

Commitment period for ESAF arrangement expired in July. The government
authorities? economic program for 1997 was monitored by IMF staff.

In December the IMF Board approved Honduras?s request for emergency
financial assistance of about $66 million to support the government?s
economic recovery program and associated relief and reconstruction efforts
in the aftermath of Hurricane Mitch.

IMF Board approved a new, 3- year arrangement under ESAF for about $215
million in March.

IMF Board completed first PRGF review and approved $22.5 million
disbursement in December.

IMF completed second review under the PRGF and approved a $21 million
disbursement in June.

Legend GDP = Gross domestic product

1997 1998 1999 2000

Appendix III: Comments From the Department of the Treasury Page 66 GAO- 01-
581 IMF: New Lending Program

Appendix III: Comments From the Department of the Treasury

Appendix IV: Comments From the International Monetary Fund

Page 67 GAO- 01- 581 IMF: New Lending Program

Appendix IV: Comments From the International Monetary Fund

Note: GAO comments supplementing those in the report text appear at the end
of this appendix.

See comment 2. See comment 1. See comment 1.

Appendix IV: Comments From the International Monetary Fund

Page 68 GAO- 01- 581 IMF: New Lending Program

See comment 7. See comment 6. See comment 5.

See comment 4. See comment 3.

Appendix IV: Comments From the International Monetary Fund

Page 69 GAO- 01- 581 IMF: New Lending Program

See comment 8.

Appendix IV: Comments From the International Monetary Fund

Page 70 GAO- 01- 581 IMF: New Lending Program

The following are GAO?s comments on the International Monetary Fund?s letter
dated April 24, 2001.

1. The IMF said that we understate the extent to which the Poverty Reduction
Strategy Paper (PRSP) approach influences the design of PRGF- supported
programs and that other groups have recognized that the new approach
represents a fundamental change in the design of PRGF- supported programs.
We disagree with the IMF?s characterization. We do recognize the extent to
which PRSPs could potentially affect the IMF program by getting countries to
take ownership of their macroeconomic policies and targets. Our report
clearly states that this change could be a major departure from how the
macroeconomic framework has traditionally been chosen.

2. The IMF stated that the requirement for borrowers to prepare an interim-
PRSP represents a major change from the way in which concessional assistance
was previously provided by the international financial institutions,
including the IMF. However, IMF guidance documents do not indicate how, or
if, the interim PRSP is to influence the PRGF policies and targets.
Therefore, our review focused on how the full PRSP- rather than the interim
PRSP- is designed to influence the PRGF.

3. The IMF said our report discounts the process of public involvement in
the formulation of key macroeconomic policy choices on the grounds that it
is ?impossible for the poorer countries to develop national professional
capacity to debate the issues.? We disagree with the IMF?s characterization
of our report. We do not discount the significance of the participatory
process. Although our report finds that national ownership of the
macroeconomic framework will be difficult to achieve, we recognize that
civil society may help improve the allocation of resources and increase the
amount of resources donors are willing to provide by helping establish
priorities for poverty reduction. Furthermore, we state that given the
complexity of the issues, civil society participation in influencing
macroeconomic policy could be limited to a few informed constituencies or
experts, such as representatives from academia and the business community
and others who can assess trade- offs.

4. The IMF said that the PRSP now forms the underpinnings not only of the
PRGF, but also of the World Bank?s concessional lending programs, and is
also being adopted as a common framework by a number of important bilateral
agencies. The IMF concludes that this is a significant difference from the
policy framework paper (PFP), which directly influenced only the ESAF
operations of the IMF. IMF and World Bank documents show, however, that the
PFP was intended to GAO Comments

Appendix IV: Comments From the International Monetary Fund

Page 71 GAO- 01- 581 IMF: New Lending Program

underpin the World Bank?s and bilaterals? programs, as well as the IMF?s.
For example, according to The World Bank Annual Report 1987, the PFPs were
to serve as an overall framework for countries?

adjustment programs that were supported by the World Bank?s adjustment-
lending operations. Also, in 1991, the World Bank?s Operational Manual
directed that PFPs should indicate how the Bank?s

assistance strategy supports and complements the country?s own approach to
reducing poverty. Moreover, in February 1988 the Fund and the World Bank
convened a seminar for senior officials of major bilateral and multilateral
agencies to discuss issues related to enhancing the usefulness of the PFPs
in coordinating aid and mobilizing resources for low- income countries.

5. The IMF said that there is substantial diversity in macroeconomic
frameworks in PRGF- supported programs and that the report?s conclusion
about the lack of scope for macroeconomic flexibility should be
reconsidered. Based on our analysis, we found that it is difficult to
determine whether there is a substantial range of macroeconomic policy
targets that are acceptable within the context of macroeconomic stability.
Furthermore, in our discussions with IMF staff on acceptable inflation
rates, we were told that the IMF expects countries to strive for single-
digit rates.

6. The IMF stated that further GAO investigation would have identified
actual changes in PRGF- supported programs and that they ?know

change is taking place in many PRGF- supported programs.? However, the IMF?s
response to our report provided no evidence to support this assertion, nor
did it provide any examples of actual changes or identify countries where
change is taking place.

7. The IMF said that we assessed two of our three country case studies
against final rather than the more relevant interim standards and implied
that we dismissed the potential of the PRGF changes. Our report did not
dismiss the potential of the PRGF but instead described some of the
challenges countries face in reaching that potential. We reported countries?
progress toward meeting the final standards and noted that certain changes
announced by the IMF- such as social impact assessments- were also called
for under the IMF?s previous concessional lending program.

8. The IMF said that the GAO analysis could be more tightly focused on
whether outcomes are likely to be better under the PRGF than under the ESAF.
We reported that the impact of the PRGF on countries? economic growth rates
is unknown at this time and that there are many challenges and obstacles to
establishing national ownership.

Appendix V: GAO Contact and Staff Acknowledgments

Page 72 GAO- 01- 581 IMF: New Lending Program

Thomas Melito, (202) 512- 9601 Cheryl Goodman, Bruce Kutnick, R. G.
Steinman, Lee Kaukas, and Rona Mendelsohn made key contributions to this
report. Appendix V: GAO Contact and Staff

Acknowledgments GAO Contact Staff Acknowledgments

(711563)

The first copy of each GAO report is free. Additional copies of reports are
$2 each. A check or money order should be made out to the Superintendent of
Documents. VISA and MasterCard credit cards are also accepted.

Orders for 100 or more copies to be mailed to a single address are
discounted 25 percent.

Orders by mail:

U. S. General Accounting Office P. O. Box 37050 Washington, DC 20013

Orders by visiting:

Room 1100 700 4 th St., NW (corner of 4 th and G Sts. NW) Washington, DC
20013

Orders by phone:

(202) 512- 6000 fax: (202) 512- 6061 TDD (202) 512- 2537

Each day, GAO issues a list of newly available reports and testimony. To
receive facsimile copies of the daily list or any list from the past 30
days, please call (202) 512- 6000 using a touchtone phone. A recorded menu
will provide information on how to obtain these lists.

Orders by Internet

For information on how to access GAO reports on the Internet, send an email
message with ?info? in the body to:

Info@ www. gao. gov or visit GAO?s World Wide Web home page at: http:// www.
gao. gov

Contact one:

 Web site: http:// www. gao. gov/ fraudnet/ fraudnet. htm

 E- mail: fraudnet@ gao. gov

 1- 800- 424- 5454 (automated answering system) Ordering Information

To Report Fraud, Waste, and Abuse in Federal Programs
*** End of document. ***