Linking Poverty Reduction Strategies and Debt Relief Initiative: An assessment of the World Bank's / IMF agenda to reduce poverty and debt

TABLE OF CONTENTS

LIST OF ABBREVIATIONS

LIST OF FIGURES

LIST OF TABLES

LIST OF ANNEXES

1. INTRODUCTION
1.1 Poverty: the global challenge of the 21st century
1.2 Aim of the analysis and chapter overview

2. UNDERSTANDING POVERTY
2.1 A definition of development and poverty
2.1.1 Measuring development and poverty
2.1.1.1 Measurements at different aggregation levels.
2.1.1.2 Single Indicators: Classification of developing countries and the international poverty lines
2.1.1.3 Composite Indicators: HDI and HPI
2.1.2 The multidimensional view of poverty
2.1.2.1 The conception
2.1.2.2 Understanding and measuring the dimensions of poverty
2.1.2.3 Pervasive poverty
2.2 Towards an agenda on poverty reduction
2.2.1 The World Bank’s thinking of poverty reduction in historical perspective
2.2.1.1 The 1950s: trickle-down growth and investments
2.2.1.2 The 1960s: approaching the poor
2.2.1.3 The 1970s: waging the war on poverty
2.2.1.4 The 1980s and 1990s: free markets and adjustment
2.2.2 Today’s framework for poverty reduction
2.2.2.1 The current understanding.
2.2.2.2 The Millennium Development Goals

3. A FRAMEWORK FOR ACTION: CDF AND PRSP
3.1 The concept of a Comprehensive Development Framework (CDF)
3.2 Poverty Reduction Strategy Papers (PRSP)
3.2.1 Background and conception
3.2.2 Developing a PRSP
3.2.2.1 The formulation of a poverty reduction strategy
3.2.2.2 The participatory process
3.2.3 Status of implementation

4. DEBT RELIEF AND THE HIPC-INITIATIVE
4.1 Background: Debt overhang theory
4.2 The amount of debt in HIPC
4.3 The HIPC-Initiative
4.3.1 Background and conception
4.3.2 Status of implementation

5. ASSESSING THE INITIATIVES
5.1 Achievements and challenges of the PRSP approach
5.1.1 Reviewing the PRSP formulation process
5.1.1.1 Analysis and monitoring
5.1.1.2 Ownership and participation
5.1.2 Reviewing the contents of PRSPs
5.1.3 Synopsis
5.2 Achievements and challenges of the HIPC-Initiative
5.3 Linking debt relief and poverty reduction

6. RHETORIC VS. REALITY: THE INITIATIVES, THE POVERTY AGENDA AND THE MDG
6.1 Does the PRSP approach follow the current agenda on poverty reduction?
6.2 The route to poverty reduction and the MDG

7. CONCLUDING REMARKS

ANNEXES

REFERENCES

LIST OF ABBREVIATIONS

illustration not visible in this excerpt

LIST OF FIGURES

Figure 1: Indicators at different aggregation levels

Figure 2: Dimensions of poverty

Figure 3: An International Poverty Trap

Figure 4: The CDF matrix

Figure 5: The process of developing a poverty reduction strategy

Figure 6: Participation in the PRSP

Figure 7: The debt Laffer curve

Figure 8: Total public external debt in the 41 HIPC Countries 1981-1999 by creditor group (billion of US dollar)

Figure 9: External debt in selected country groups 1981–1999 as percentage of debt-to- exports and debt-to-GNP.

Figure 10: Debt stock reduction in US$ billions

Figure 11: Debt service and social expenditure ratios for the 26 decision point countries (weighted averages in percent)

Figure 12: Direction of conditionality and incentives

LIST OF TABLES

Table 1: The World Bank’s classification of economies

Table 2: Population below $1 a day (1993 PPP), 1987-1998

Table 3: PRSP status of implementation by country group (as of February 2003)

Table 4: Progress towards selected MDG

LIST OF ANNEXES

Annex 1: Calculating the HDI and HPI

Annex 2: The Millennium Development Goals

Annex 3: The Comprehensive Development Framework and its wider context

Annex 4: Overview of the HIPC-Initiative

1.INTRODUCTION

1.1 Poverty: the global challenge of the 21st century

The fight against poverty has always been at the core of development economics. Reviewing achievements of past decades, one has to acknowledge that great progress has been made towards improving the conditions of people living in what is called ‘the developing world’: life expectancy rose, infant mortality rates dropped and primary school enrolment rates doubled – just to mention a few examples. Or, how the World Bank puts it

rather simplistically on its website: “The developing world today is healthier, wealthier, better fed, and better educated.”¹

Nevertheless, besides these encouraging results, the progress has been spread unevenly in different parts of the world. Thus poverty still remains one of the greatest global challenges today. It is estimated that 1.2 billion people live on less than one dollar a day – that represents every one in four persons in the developing world – and 2.8 billion, around half of the total world population, live on less than two dollars a day.² Looking beyond merely consumptive measures of poverty, the poor countries still see low levels of human development and have to face diverse challenges such as HIV/Aids, armed or ethical conflicts, famines and environmental degradation.

The 1990s can be seen as a period of stocktaking about development: a series of reassessments about achievements and shortcomings concerning poverty brought about new insights and strategies on the understanding of poverty and on the mechanisms needed for poverty reduction. The key players in this process are certainly international organizations like the World Bank, OECD and United Nations (UN). The current understanding of poverty was influentially shaped by the World Development Report (WDR) 2000/2001 of the World Bank, which broadened the definition of poverty towards a multi-dimensional concept. With this came also several new strategic proposals towards poverty reduction. As a yardstick, the international community committed itself to halve

the proportion of people living in poverty until 2015 as it is incorporated in the eight Millennium Development Goals (MDG).³ World Bank president James D. Wolfensohn made a proposal for a Comprehensive Developing Framework (CDF) – an analytical tool designed for developing countries to align their macroeconomic, social, structural and human development aspects with the efforts of the international donor community. An integral part of the CDF was the development of Poverty Reduction Strategy Papers

(PRSP) by the developing countries governments. A PRSP should lay out a framework for poverty reduction, be developed in a participatory manner and be owned by the national government. The PRSP has to reflect the multi-dimensional, comprehensive view of poverty and is expected to increase effectiveness through national ownership and results- orientation.

A special aspect in the fight against poverty is the indebtedness of developing countries. Theory suggests that highly indebted countries suffer from a debt overhang that contributes to low growth and creates disincentives towards adjustment and reform. Furthermore, high debt repayments take away resources needed for social and economic development, thus feeding a vicious circle of poverty and indebtedness. Realizing that many of the poorest countries have debt levels in excess of their repayment capacity, the World Bank and the International Monetary Fund (IMF) proposed the Highly Indebted

Poor Countries (HIPC)-Initiative in 1996.⁴ It provides debt relief for countries that have a

level of ‘unsustainable debt’ according to World Bank/IMF criteria and/or that are eligible for World Bank/IMF concessional assistance. A condition for debt relief under the HIPC- Initiative is the formulation of a PRSP by the national government. This is meant to ensure that funds freed through debt relief are channeled towards poverty reduction. With this, a combination of debt relief and poverty reduction demanded by several critics and Non Governmental Organizations (NGO) has been established.

1.2 Aim of the analysis and chapter overview

The aim of this analysis is to provide a description of the current understanding of poverty reduction in the international community and to assess two strategies developed for the fight against poverty. For that purpose this study illustrates today’s conception of poverty and gives an extensive review of the PRSP approach and the HIPC-Initiative. Its focus is on the achievements and challenges of the two strategies. It seeks to evaluate the benefits and shortcomings inherent in the formulation and implementation process. Besides it is argued in which way the link between debt relief and poverty reduction represents a feasible solution. Finally this study will combine the insights gained from the discussion of the agenda and the initiatives. By investigating whether the new strategies represent a comprehensive approach consistent with the current agenda it is possible to make a judgment about their potential for poverty reduction.

This study does not intend to provide a discussion about the appropriateness of the international agenda or the chosen polices. For example it will assess to what extent the understanding of poverty relies on growth considerations and how this is reflected in the strategies. But it does not seek to evaluate if economic growth is an adequate means of poverty reduction. At center of this study is the question about the soundness of the approaches vis-à-vis what is expected from the poverty reduction process and the way it is carried out.

The most influential players in the development business are multilateral organizations. This analysis almost entirely focuses on the viewpoints and activities pursued by the World Bank and – in part – by the IMF. Although this is to some extent a simplification, as it neglects the viewpoints of other important players such as the UN, the OECD or national development agencies, this approach can be justified by two main arguments.

First, the World Bank is the leading organization committed to economic development and poverty reduction and has taken a foremost role concerning these issues. Due to its size and influence, the World Bank’s thinking and policies have strong impact on debate and consensus in the field of economic development.⁵ The understanding of poverty and poverty reduction has been shaped influentially by the World Bank in past and present; today’s consensus, as it is formulated in the WDR 2000/2001 has been adopted by

several other agencies. Second, the initiatives assessed in this study originated from the World Bank/IMF. The CDF/PRSP process has been developed by the World Bank, which also fosters its implementation. The HIPC debt relief initiative has been put forth jointly by the IMF and the World Bank. By focusing on the World Bank/IMF thinking and putting it in perspective with its operations, it is possible to provide a comprehensive analysis.

The second chapter places its focus on examining the meaning of poverty and poverty reduction. A definition of poverty and development is supplied at the beginning. It will become obvious that both terms describe very complex matters: on the one hand, both can be defined on purely monetary matters based on consumption data where clear, internationally accepted base lines exist. On the other hand they appear as multi- dimensional phenomena that include a range of economical, social, political and institutional factors. The definition presented in this study provides an analysis of different measures and conceptions along the various dimensions and is followed by a presentation of the World Bank’s thinking on poverty reduction from past to present. A review starting in the 1950s shows the trains of thought and the evolvement of the conception of poverty reduction. Today’s agenda, which is illustrated afterwards, combines several elements of past strategies with the multi-dimensional understanding of poverty. Thus chapter two presents a framework for poverty reduction that is comprehensive in scope and internationally accepted. This thinking builds the basis for both the CDF/PRSP process and the HIPC-Initiative which are exhibited in detail in the following chapters.

Chapter three describes the World Bank’s strategies for poverty reduction: the Comprehensive Development Framework (CDF) and the Poverty Reduction Strategy Papers (PRSP). The PRSP are at the heart of this chapter as they lay out the exact ways in which a reduction of poverty in each country can be achieved. Operational issues are discussed in detail by reviewing the World Bank’s guidelines on how to prepare a PRSP and how to conduct the participatory processes demanded in the formulation process. The end of the chapter gives a short overview on the status of implementation.

Chapter four illustrates a special issue in the fight against poverty: the indebtedness of developing countries and the demand for debt relief. First, economic theory is provided that clarifies the interaction between debt and poverty and which explains the rationale for debt relief. Next the focus is placed on the group of Highly Indebted Poor Countries (HIPC) and evidence of the amount of debt stock in these countries is supplied. In order to reduce this debt burden the World Bank and the IMF have designed an initiative that

provides debt relief on the condition that the countries prepare a PRSP. The conception of the HIPC-Initiative is presented subsequently. At the end of the chapter, realization and impacts of the initiative are illustrated.

Chapter five examines the challenges and achievements of the PRSP process and the HIPC-Initiative. The PRSPs are critically assessed by reviewing the process of strategy formulation and the policy contents. It is possible to identify two major obstacles that constrain the impact of the PRSPs: a capacity-problem and a lack of strategic focus. The HIPC-Initiative is appraised afterwards. It becomes obvious that it suffers from methodological errors and overstated assumptions. As a final point, the connection between debt relief and poverty reduction is assessed. This is acknowledged as a means of conditionality that creates positive incentives for implementation. However, the flaws in the two approaches noted before point at several tensions that inhibit the potential of the connection.

After providing information on today’s view of poverty and on the initiatives established to fight poverty, chapter six combines the insights from the four previous chapters. First it is asked if and how the initiatives described in chapter three, four and five adhere to the understanding of poverty reduction set out in chapter two. It is concluded that the country strategies rather than the process itself fail to live up to the comprehensive understanding. This is attributed to the lack of strategic focus observed earlier. At the end it is possible to provide a judgment about the benefits arising from the new strategies. The main achievement of the two approaches is that they resulted in concerted efforts of all actors to achieve comprehensive poverty reduction. As both processes are still in their starting phase, it is hardly possible to make a final conclusion. What is done in this study is that the progress towards the MDG is evaluated and taken as an outlook for poverty reduction.

Chapter seven concludes this analysis with a summary of the main findings and shows up potentials for further research.

Chapter two UNDERSTANDING POVERTY

2.1 A definition of development and poverty

The terms ‘development’ and ‘poverty’ are abstract expressions that offer a wide range of interpretations and understandings. Both seem to be related, like it is explained by Duth/Körner/Michaelowa: “Whereas in the past development had been regarded as a

necessary condition for poverty reduction, today overcoming poverty [...] seems a necessary condition for development.”⁶ In order to make a discussion about poverty and poverty reduction strategies, the attempt has to be made to lay out a framework that forms a basis of common understanding of those focal terms. Chapter 2.1.1 aims at identifying ‘who the poor are’ by providing an overview on measures and classification schemes, then in chapter 2.1.2, the attempt is made to clarify ‘what poverty is’.

2.1.1 Measuring development and poverty

2.1.1.1 Measurements at different aggregation levels

Development itself can be seen as a goal that countries wish to achieve, thus it describes a process of change in development levels over time.⁷ When development is seen as a goal, one is able to identify broad objectives that have to be pursued towards goal attainment. Two conceptions are provided exemplary. The first by Todaro identifies the following three objectives of development:⁸

- to increase the availability and widen the distribution of basic life-sustaining goods;
- to raise levels of living;
- to expand the range of economic and social choices.

Second, Nohlen and Nuscheler formulate five elements as goals of development: growth, labor, equality/justice, participation and independence.⁹

It becomes clear from both models that development consists of a multitude of goals and inherent factors that influence outcomes. Some of the elements formulated above are in itself again very complex or diverse matters which require further reasoning, such as the terms ‘levels of living’, ‘growth’ or ‘equality’. Furthermore, this list implies new questions regarding the appropriateness of selection, completeness, reinforcements or

conflicts, weight and timing of goals.¹⁰

In the case that the development process is regarded, the question arises of how to describe the level of development or the standard of living at a given level of time. For this purpose, indicators are needed that allow the comparability and classification of different stages of development. Figure one shows an overview of classes of indicators used to measure development and poverty.

Figure 1: Indicators at different aggregation levels

illustration not visible in this excerpt

Source: OECD (2001), p. 42.

Indicators are important means of identifying levels of development and patterns of poverty, of monitoring policy outcomes and goal attainments. Nevertheless, one has to keep in mind that an indicator is based on a statistical quantity, but is then put in a wider

relationship with a theoretical background that allows it to explain something that it did not measure:¹¹ for example is the infant mortality rate used as an indicator for the level of health, or the school enrolment rate as an indicator for the level of education. Thus indicators are based on assumptions about their explanatory power (validity) and are subject to measurement and data collection problems.

According to the framework described in figure one, the next two chapters present commonly used single indicators and composite indexes that are utilized to measure levels of development and poverty. Discrete indicators reflect the various dimensions of development and poverty and are dealt with in the chapter 2.1.2.

2.1.1.2 Single Indicators: Classification of developing countries and the international poverty lines

Single indicators provide “a bird’s eye view measure”¹², because they allow to distinguish between major groups of developed and developing countries. The most commonly used single indicators are consumption or income measures, which are widely available and easy to quantify. With their help one can build a classification system of countries and conduct overall comparisons and monitoring.

Table one shows the classification system of economies used by the World Bank that groups countries into four categories dependent on their per capita Gross National Income (GNI).¹³ It has to be noted that this classification of the World Bank is not the only system, though it is widely used: for example the UN and OECD have different categories and classification schemes.¹⁴

Table 1: The World Bank’s classification of economies

illustration not visible in this excerpt

1 in 2000; 2 economies with population in excess of 30,000 in 2001 Source: Data from World Bank (2002a), p. 231, 233, 241.

The low and middle-income countries are regarded as developing countries. This definition uses the per capita levels of income as a measure for development on the assumption that a growth of national income is associated with progress in development. Although economic growth generally goes together with improvements of social and human indicators, taking it as the only proxy for development disregards the fact that many

other factors influence the development of countries and the well-being of people.¹⁵

Furthermore, several methodological problems occur if income is used as an indicator.¹⁶ Income statistics do not account for self-consumptive production, barter trade and the informal sector – although such non-market activities occur frequently in developing countries. Data might be flawed due to problems of collection, wrong estimations (of for instance population size) and the conversion in US dollar.

A general measurement for the comparison of poverty is provided with the international poverty lines developed by the World Bank. They are based on consumption data and define thresholds below which private consumption is considered inadequate to maintain a minimally adequate standard of living. The international poverty lines are:¹⁷

- $1 a day (in 1993 Purchasing Power Parity (PPP)) as the lower poverty line estimated for low income countries;
- $2 a day (in 1993 PPP) as the upper poverty line estimated for lower-middle income countries.

The international poverty lines are not set arbitrarily. They are based on a regression model covering household consumption data of 33 developing countries.¹⁸ The model is tested against median poverty lines found in developing countries, which confirm the validity of the poverty lines. By comparing poverty estimates of different regions from different times, the international poverty lines can be used to identify levels and track global progress of absolute poverty as it is shown in table two.

Table 2: Population below $1 a day (1993 PPP), 1987-1998

illustration not visible in this excerpt

Source: Chen/Ravallion (2000), p. 27 also World Bank (2000a), p. 23.

Obviously the total share of people living in poverty has fallen from 28 to 24 percent while the absolute number remained roughly the same. Large regional differences can be noted: only in South Asia and the Middle East& North Africa the number and the share of people under $1 a day has fallen whereas everywhere else the number of people rose. In Sub-Saharan Africa (SSA) over 46 percent of the population is income poor. These findings suggest that low advancement has been made towards reducing poverty which are further accompanied by rising inequalities, despite aggregate economic growth in the

developing world.¹⁹

The international poverty lines are an easy-to-use base line of the scope of poverty in the world. But their explanatory power in conjunction with poverty estimates is limited to an overview of broad trends and their construction is not problem-free. The World Bank admits that the survey of consumption data varies between countries and over time which makes the comparison more difficult.²⁰ Furthermore, converting household consumption data into poverty estimates requires several assumptions which affect the outcome. Kircher criticizes that this measure relies only on resource considerations thus neglecting other dimensions of well-being.²¹ He adds that the construction is to that extent arbitrary as the choice of countries is limited and the identification of typical lines for certain group of countries might not be representative in all cases. Other authors pose more fundamental critique on the international poverty lines. Reddy/Pogge reject in their paper the validity of the international poverty lines and the World Bank’s estimations.²² Similar to Kircher they criticize that a poverty line has to measure essential commodities or capabilities that affect

poverty – and not only income. More important, they see the conception as flawed due to the conversion in PPP. The set of PPP used is not relevant for poverty assessments as its commodities are neither consumed by the poor nor does a change in prices impact on poor households in practice (but due to the conception impacts on the level of poverty lines). Bhalla also basically refuses the soundness of the poverty lines based on methodological

errors, especially the PPP projections.²³ Somehow surprisingly, although the authors start

with analogous assumptions they arrive at absolutely diametric conclusions. Whereas Reddy/Pogge estimate that the procedural distortions are likely to result in an understatement of the extent of poverty, Bhalla assumes an overstatement.²⁴ The latter author estimates for 2000 the share of people living on $1 a day at 650 million – nearly 45 percent lower than the official World Bank figures.

With regard to these arguments it has to be acknowledged that the poverty lines cannot be used in a country-level analysis, thus national poverty lines need to be formulated by each country for the purpose of measurement and policy or program design. This calls for the adoption of the conception of ‘relative poverty’ which differs from the definition of ‘absolute poverty’ underlying the international poverty lines in that respect that it regards poverty as dependent on changes in the society: for example what is

considered a minimally adequate level of consumption changes over time or might be different between regions.²⁵

It can be concluded that single indicators have the purpose of providing an overview of levels of development or poverty and to allow for classifications. Their explanatory power regarding policy action is low. Apart from technical problems in measurement, the main criticism is that a single quantity – here: consumption or income – is used as a proxy for development and poverty while neglecting other influential factors. Composite indicators try to solve that shortcoming by adding other variables into a measure for development and poverty.

2.1.1.3 Composite Indicators: HDI and HPI

The relevance of composite indicators has its origin in the fact that the traditionally used income or consumption indicators are insufficient in representing levels of development and poverty as they are biased towards economic criteria and disregard socio-economic aspects.

In its 1990 Human Development Report (HDR) the UNDP introduced the concept of the Human Development Index (HDI). The concept of the HDI is systematically and comprehensive in that regard that it focuses on the ends (human development) rather than on the means (economic growth) of the development process and gives direct value to those factors that influence the standard of living beyond income. The HDI encompasses

the following components:²⁶

- longevity (measured by life expectancy at birth);
- knowledge (measured by a weighted average of adult literacy and mean years of schooling);
- standard of living (measured by GDP per capita (PPP, US$)).

The HDI for each country is then calculated as a relative index value between zero and one taking the simple averages of the measures, with the value of one indicating highest achievements in human development.

Similarly, the UNDP constructs the Human Poverty Index (HPI) as a measure of depreciation. The HPI-1 is applied to developing countries and encompasses the same components as the HDI (longevity, knowledge, standard of living) but uses alternative indicators; the HPI-2 is calculated for OECD countries, measuring longevity, knowledge, standard of living and social exclusion.²⁷

HDI and HPI can be used to gain new insights regarding the extent of development and poverty in the world. The UNDP groups countries according to their HDI value and compares the ranking with a ranking according to GDP per capita.²⁸ In several cases, the HDI-rank deviates from the GDP-rank of a country. This illustrates that some countries are more/less successful in transforming income and growth into socioeconomic development outcomes which gives further insights towards priorities of government spending or the quality of economic policy.²⁹ Furthermore, this points to the fact that income is an

incomplete measure of development. The HPI is used in a similar fashion to extent the view on poverty beyond the focus on consumption or income.

Composite indexes like HDI and HPI are improvements over single measures due to their broader focus. Nevertheless they are also subject to criticism. The composition and weighting is to some extent arbitrary and not backed up by theoretical considerations.³⁰ Due to missing data, the indicators cannot be calculated for all countries and have to focus on indicators with good availability.³¹ Furthermore, the HDI also neglects important variables of development such as environmental quality or political variables.

Apparently composite indicators have the potential to overcome some failures inherent in the use of single indicators. By combining social and economical variables they give a more comprehensive view on the state and progress of countries concerning development and poverty. Like single indicators, composite indicators provide a view on average trends and patterns, thus their use in policy advice on national or even regional level is limited. The largest explanatory power can be gained, when composite indicators are used in conjunction with other measures, or when they are disaggregated into several components or regions.

2.1.2 The multidimensional view of poverty

2.1.2.1 The conception

The preceding chapters have shown how development and poverty is measured at a broad level. Single and composite indicators can be utilized to gain an overview on trends and progress, but both have been criticized on the grounds that they can hardly be used for policy formulation since they encompass only few variables. It is accepted in the international community that a conception of poverty has to include multiple dimensions of depreciation and has to address the interactions and reinforcements that exist between the

dimensions.³² A comprehensive scheme is provided by the OECD and shown in figure

two.

Figure 2: Dimensions of poverty

illustration not visible in this excerpt

Source: OECD (2001), p. 39.

This conception of poverty defines poverty as the lack of economic, human, socio- cultural, political and protective capabilities. The arrows indicate that the dimensions are interrelated and each one affects the others while it itself is affected by them. Gender and environment are included as external factors highlighting the fact that “poverty, gender and environment are mutually reinforcing, complementary and cross-cutting facets of sustainable development.”³³

The World Bank provides a similar conception in its WDR 2000/2001:

“This report accepts the now traditional view of poverty […] as encompassing not only material deprivation […] but also low achievements in education and health. […] This report also broadens the notion of poverty to include vulnerability and exposure to risk -and voicelessness and powerlessness.”³⁴

It identifies three main dimensions of poverty that coincide with those formulated by the OECD:³⁵

1. Lack of income and assets to attain basic necessities (representing the economic and human dimension of the OECD conception);
2. Sense of voicelessness and powerlessness (representing the political and socio- cultural dimension of the OECD conception);
3. Vulnerability to adverse shocks (representing the protective dimension of the OECD conception).

Both definitions go beyond a merely monetary definition and include several non-resource based measures of depreciation. The next chapter gives a wider insight into the dimensions.

2.1.2.2 Understanding and measuring the dimensions of poverty

The economic dimension encompasses the ability of people to earn income and to consume in order to meet basic needs as well as to have access to resources, both financial and physical. Measures of the economic dimension are those income or consumption indicators

explained in chapter 2.1.1.2. The World Bank further stresses the importance of growth and income distribution for the enhancement of economic capabilities.³⁶

The human dimension is based on the factors health, education and nutrition. All of them are core elements of well-being and ensure the participation in work and social live. Education is perceived to be the most important asset in the fight against poverty.³⁷ It can be measured by literacy rates, spending on education or school enrolment ratios.³⁸ Health measures include spending on health, access to medical treatment or drugs, life expectancy

and infant mortality. The share of undernourished people and statistics of calorie consumption, children’s weight and height measures nutrition.

The socio-cultural dimension refers to social status and dignity. Exclusion from social life and missing acceptance in a community contributes to voicelessness and depreciation. Among others, it can be due to race, religion, gender or illness (for example Aids) and might deny people access to work or education, prevent them from building up assets and personal networks and can lead to violent conflict.³⁹

The political dimension of poverty includes human rights violations and the lack of voice or power of poor people towards institutions of the state. Examples are the absence of the rule of law, arbitrary actions of public authorities, intimidation or corruption which hinders people’s access to assets and prevent them from participation. Measures can be

qualitative variables of the extent of civil and political liberties, assessments of human rights and human freedom (for example the country ratings of Freedom House) or indexes such as the ‘Corruption Perception Index’ of Transparency International.⁴⁰

The protective dimension of poverty concerns security and vulnerability – the ability to cope with risks and to respond to external shocks. Poor people are unable to mitigate risks like economic crisis, diseases, hunger or war because they lack the assets for engaging in mechanisms for risk reduction and coping.⁴¹ Measuring vulnerability and the exposure to risk requires a dynamic conception, as people move in and out of poverty: it requires observance of the variability of a household’s asset base in combination with data on formal safety nets, the functioning of markets and the economic policies that determine a household’s opportunity set and the range of activities it can pursue to manage risk.⁴²

The different dimensions are connected with each other and overlap to a certain degree: improving health and education can in turn lead to better economic prospects. Providing protection for poor people enables them to take advantage of opportunities. Increasing people’s voice and power influences the socio-cultural dimension as well as their access to assets. An agenda on poverty reduction has to design policies that tackle the problems along these dimensions.

2.1.2.3 Pervasive poverty

In table two of chapter 2.1.1.2 it was shown that since 1987 consumption poverty had not seen much reduction on aggregate and that progress has been spread unevenly across countries. If we account for the fact that poverty is a multidimensional phenomenon like it is described in the previous chapter, it can be concluded that some of the poorest countries are stuck in a poverty trap in which several mutually influential variables affect the outcome. A poverty trap exists when poverty has effects that act as causes of poverty; in

this case generalized poverty with its many components contributes to economic stagnation and vice versa.⁴³ Figure three illustrates a poverty trap at the international level.

Figure 3: An International Poverty Trap

illustration not visible in this excerpt

Source: UNCTAD (2002), p. 149.

The figure shows how international and national variables interact in a circular process that leads to the persistence of poverty. The left side of figure three illustrates which domestic aspects feed the poverty trap. Here we find connections with the elements identified in the definition of poverty. The middle shows a vicious circle of low investment, savings and productivity at the national level, which can be attributed to the economic (low income and resources) and human (low education and health) dimension of poverty at the individual level. Poor state and corporate capacities and weak political stability point at the political, protective and socio-cultural dimensions of individual poverty. External factors like environmental degradation and population growth contribute to poverty. The right side of the diagram shows how international relationships reinforce pervasive poverty at the national level: commodity structures, debt levels and aid/debt service are influential factors. In chapter four of this study, the relationship of debt and poverty will be assessed in detail.

Sustainable poverty reduction has to address all dimensions that affect poverty and pay attention to the relationships between them. An ‘orchestrated’ policy package of national and international policies is needed to shift an economy out of its low-level

equilibrium.⁴⁴ How the understanding of poverty reduction has changed over the years is documented in the next chapter.

2.2 Towards an agenda on poverty reduction

2.2.1 The World Bank’s thinking of poverty reduction in historical perspective

Since its founding, the World Bank has been committed to economic development and poverty reduction and has over the years influenced thinking and opinion on these issues. Today’s perspective on poverty reduction can be seen as a synthesis of past viewpoints and strategies. Therefore, this chapter illustrates the evolution of the World Bank’s thinking concerning poverty reduction and puts it in perspective with development theory.⁴⁵

Four eras can be identified and are discussed in turn:⁴⁶

1. The 1950s: trickle-down growth and investments;
2. The 1960s: approaching the poor;
3. The 1970s: waging the war on poverty;
4. The 1980s and 1990s: free markets and adjustment.

Today’s understanding of poverty reduction is presented afterwards in chapter 2.2.2.

2.2.1.1 The 1950s: trickle-down growth and investments

Economic theory in that era regarded poverty purely as a lack of income, thus development strategies focused on per capita income growth. It was believed that growth would ‘trickle- down’ in a virtuous cycle to all levels of society by bringing employment and other economic opportunities. Large-scale investment planning of industry and infrastructure was regarded as necessary, as it was argued in Rosenstein-Rodan’s ‘big-push’ or Nurkse’s

‘balanced growth’ theory.⁴⁷

The World Bank had been mainly in line with this theory: it focused on project implementation of capital-intensive infrastructure and government advice.⁴⁸ Topics such as

poverty, agriculture or social services were regarded as secondary; support was allocated mainly towards more advanced economies rather than to the poorest. The meaning of development at the World Bank “more often meant physical output than human betterment, economic opportunity rather than social justice.”⁴⁹ However, in this era the World Bank developed several mechanisms and capacities that allowed it to pursue an independent future course towards poverty reduction. Most importantly in this regard had

been the creation of the International Development Association (IDA), which began providing soft loans to developing economies.

2.2.1.2 The 1960s: approaching the poor

This era can be regarded as an enhancement of the World Bank’s activities towards the poorest countries. The understanding of development started to include poverty alleviation for the less developed countries – but still in the sense of raising income levels. This was mainly reflected in increased lending to developing countries and a poverty sensitive sector allocation, which was both facilitated by the work of the IDA whose total lending in this period was almost as high as the whole World Bank lending to less developed countries in

the previous era.⁵⁰ But economic considerations remained at the forefront of operations:

eligibility for IDA support was dependent on per capita income and lending was focused on ‘productive’ projects contributing to economic growth, hence the soft-loan or social- lending components of the IDA were seen with reluctance.⁵¹

In the academic world skepticism grew about the trickle-down approach pursued up until now. This was supported by the theories of Lewis and Kuznets which implied that growth went together with rising inequalities.⁵² Worsening income distribution in early stages of development was regarded as a threat to poverty reduction and as an area for intervention. The World Bank dealt with this issue by recognizing ‘underdevelopment’,

rising inequalities and the need for redistribution as important aspects, yet not taking actions until the 1970s.

2.2.1.3 The 1970s: waging the war on poverty

The starting point of this era was the appointment of Robert McNamara as president of the World Bank in 1968. His aim was to redefine the World Bank as a development agency and to give the fight against poverty high priority.⁵³ A rethinking of development experience took place and brought topics such as nutrition, health, education, employment, urbanization or water supply on the agenda. As a consequence, the World Bank adopted the basic human needs approach first formulated by the International Labor Organization (ILO), which regarded the fulfillment of basic needs of the individual as the overriding objective of a poverty reduction strategy.⁵⁴ Parallel to this, a study co-hosted by the World Bank entitled “Redistribution with Growth” contributed to the skepticism towards former growth-oriented strategies and emphasized the issues of inequalities und redistribution.⁵⁵

In accordance with this change of thinking, the operations of the World Bank altered in two ways. First, there was a large increase in projects targeted directly to the poor (called ‘poverty lending’) – mainly towards agricultural and rural development – that now accounted for more than one-quarter of total lending.⁵⁶ Second, the World Bank started to exert increasing influence on countries policies and public opinion: on the one side through discussions and negotiations with borrowers and on the other side through its authority gained by its lending operations and its research.⁵⁷

The understanding of poverty was substantially broadened throughout this decade and for the first time, the success in poverty reduction served as a measure of development efforts. Nevertheless, according to Kapur/Lewis/Webb the results were disappointing due to failures of projects regarding poverty relief and difficulties in designing adequate

policies.⁵⁸ They note that still several important policy areas had been excluded from the agenda, for example participation and empowerment, discrimination, environment and major health issues.

2.2.1.4 The 1980s and 1990s: free markets and adjustment

Economic stagnation and a series of macroeconomic shocks (debt crisis, energy crisis and global recession) detrimental to the growth and poverty relief prospects of the developing countries gave rise to the belief that large changes in economic policy were needed in response to that challenge. It was assumed that governmental interventions brought about distortions and rigidities in the market mechanism. According to a neoclassical viewpoint,

removing price distortions and ensuring free international trade and free play of market forces would have positive effects on growth and poverty reduction.⁵⁹ This was manifested in the so-called ‘Washington consensus’ that formulated a liberalization agenda that the Washington-based institutions (World Bank and IMF) should enforce in the developing countries. It proposed the following reforms: fiscal discipline, redirection of public expenditure, tax reform, interest rate liberalization, competitive exchange rate, trade

liberalization, foreign direct investment, privatization, deregulation and secure property rights.⁶⁰ The consensus had wide implications and the neoclassical agenda was theoretically backed up by studies finding that growth did not go together with rising inequalities (thus questioning Kuznets’ hypothesis) but contributed to poverty reduction.

The World Bank went in line with these policy proposals. The poor performance of projects directly targeted at the poor brought more generalized policy and market reforms into focus: at the core were programs for structural adjustment undertaken in conjunction

with the IMF which consisted of loans to developing countries disbursed on the restriction of policy adjustment according to World Bank/IMF conditions.⁶¹ Thereby, the World Bank turned against the basic needs agenda of the 1970s and returned to an understanding that stressed economic efficiency and growth as the main drivers towards poverty reduction.

Nevertheless, there have been debates about that course and about the impact of structural adjustment: at the end of the 1980s the issue of social costs of structural adjustment was raised and the discussion about the poverty agenda renewed.⁶² It was argued that economic adjustment worsened the living standards of the poor instead of lifting them out of poverty. In the WDR 1990 the World Bank acknowledged the social costs of adjustment and that growth was not sufficient for poverty reduction but had to be accompanied by investments in human resources, health or social safety nets.

Despite this concession, the criticism concerning the liberalization agenda and structural adjustment programs – raised in particular by NGO’s – remained strong throughout the 1990s. Concerning the World Bank’s history on poverty reduction Kapur/Lewis/Webb conclude:

“In practice, the Bank continues to direct most of its lending and advice to promote economic growth and to provide balance-of-payments and fiscal assistance. And, to these authors, it seems likely that whatever contribution has been made by the Bank to poverty reduction has been principally the result of that general support for economic stability and development rather than of its efforts to bring about more direct poverty alleviation. […] The Bank’s own view of those efforts to improve on trickle-down has been largely critical. […] Much of what is called “direct poverty lending” remains haunted by the question of fungibility.”⁶³

2.2.2 Today’s framework for poverty reduction

2.2.2.1 The current understanding

During the 1990s the market liberalization and adjustment agenda became more and more questioned due to negative experiences like the low growth in transition economies and shocks like the East Asian crisis. Also the progress on poverty reduction, like it is illustrated in table two, was not satisfactory. The new endogenous growth theory appeared and suggested a dynamic and path-dependent conception of economic growth that helped to explain divergences between nations and highlighted the importance of human capital in

the production process.⁶⁴ New studies emphasized the connections between health,

nutrition, wages and productivity, showed that human capital formation through education

leads to increasing social returns and pointed at the dangers of volatility and risk for poor households.⁶⁵

The current understanding of poverty reduction draws on these experiences and regards equity and security as a prerequisite for growth. Moreover, it focuses on improving poor people’s assets and capabilities. It builds on the comprehensive view of poverty like it is outlined in chapter 2.1.2. The World Bank illustrates its agenda in the WDR 2000/2001 addressing the following areas: Opportunity, Empowerment, and Security.⁶⁶

Opportunity is concerned with pro-poor growth, markets and assets referring to the economic and human dimension of poverty.⁶⁷ Economic growth is seen as highly important for development as long as its benefits are distributed to the poor. It is affected by levels of inequalities and it has to be accompanied by sound economic policies, institutions, human development and environmental concerns to be sustainable. Market friendly reforms on the macro level should be promoted in the spirit of the Washington

consensus. In order to avoid the negative consequences on poor people these reform have to be backed up by improvements on the micro level – like investments in infrastructure, lighter regulatory burdens, labor standards or access to finance – that give the poor access to markets and economic opportunities. Crucial is the expansion of poor people’s assets that enable them to seize opportunities, cope with risks and improve the standard of living. Important in this respect are human capabilities (like health, education) and ownership of physical or social assets (such as participation). The state has a central role in delivering services and providing an environment conductive to the accumulation of assets. The question of asset ownership is a core conception in the fight against poverty and already addresses questions of empowerment and security.

[...]

---

¹ http://www.worldbank.org/poverty/mission/up3.htm (18.12.2002). When I speak in this text of the World Bank I refer to the operations of the International Bank for Reconstruction and Development (IBRD) and the International Development Agency (IDA) as they are usually perceived as one entity.

² See World Bank (2000a), p. VI; OECD (2001), p. 15.

³ See OECD (2001), p. 17; BMZ (2001), p. 11 for the MDG.

⁴ The original HIPC-Initiative from 1996 was modified in 1999 and is since then referred to as the enhanced HIPC-Initiative or HIPC II. In this text I will continue to write ‘HIPC-Initiative’ meaning the current framework of HIPC II.

⁵ See Gilbert/Powell/Vines (2000), p. 49 for the roles and activities of the World Bank.

⁶ See Durth/Körner/Michaelowa (2002), p. 9-10 (own translation).

⁷ See Ingham (1995), p. 33; Wagner/Kaiser (1995), p. 6-7.

⁸ See Todaro (1997), p. 18.

⁹ See Nohlen/Nuscheler (1993a), p. 65 (own translation); also picked up at Wagner/Kaiser (1995), p. 7.

¹⁰ such questions are raised at Nohlen/Nuscheler (1993a), p. 65-67.

¹¹ See Nohlen/Nuscheler (1993b), p. 76-82 for a discussion of the explanatory power of indicators.

¹² OECD (2001), p. 42.

¹³ Since the WDR 2002, the World Bank classifies countries according to GNI per capita in line with the 1993 System of National Accounts (SNA). Up to the WDR 2002, classification of countries was made according to Gross National Product (GNP) per capita. See World Bank (2002a), p. 231.

¹⁴ See Todaro (1997), p. 27 for a summary of different classification systems.

¹⁵ The positive connection of growth and development is acknowledged in most of the literature, exemplary OECD (2001), World Bank (2000a). More critical views take Nohlen/Nuscheler (1993a), p. 67-68 or Todaro (1997), p. 12-13 and others. Recent publications stress the quality of growth, i.e. World Bank (2000b).

¹⁶ See Nohlen/Nuscheler (1993b), p. 82-85 for shortcomings of per capita income indicators.

¹⁷ See World Bank (2000a), p. 17; Chen/Ravallion (2000), p. 5; exact figures are $1.08 / $ 2.15 a day (1993 PPP) but referred to throughout as $1 / $2 a day.

¹⁸ See Chen/Ravallion (2000), p. 5-7 and UNCTAD (2002), p. 42-43 for the research on recent international poverty lines. The fist modeling of the international poverty lines can be found in Ravallion/Datt/van de Walle: “Quantifying Absolute Poverty in the Developing World”, Review of Income and Wealth 37, 345- 361, 1991.

¹⁹ See Chen/Ravallion (2000), p. 18-20. The overall rate of growth in real per capita private consumption for the low- and middle-income countries over 1990-97 was 2.6% per year.

²⁰ See World Bank (2000a), p. 17.

²¹ See Kircher (2002), p. 43.

²² See Reddy/Pogge (2002), p. 4.

²³ See Bhalla (2002), p. 64-67.

²⁴ See Reddy/Pogge (2002), p. 31 and Bhalla (2002), p. 139.

²⁵ See Kircher (2002), p. 37-38 and Durth/Körner/Michaelowa (2002), p. 15-16 for the definition of absolute and relative poverty. See World Bank (2002a), p. 245; UNCTAD (2002), p. 44 for examples of relative changes of poverty measures: the UN states that consumption levels shift with globalization and as societies become richer. The World Bank adds that for example the cost of living is higher in urban than in rural areas, thus urban and rural poverty lines should differ.

²⁶ See UNDP (2002), p. 34. See Annex 1 for components and calculation of HDI and HPI. 27 See UNDP (2002), p. 35. See Annex 1 for components and calculation of HDI and HPI. 28 See UNDP (2002), p. 149-161 for country data and ranking on HDI and HPI.

²⁷ See UNDP (2002), p. 35. See Annex 1 for components and calculation of HDI and HPI.

²⁸ See UNDP (2002), p. 149-161 for country data and ranking on HDI and HPI.

²⁹ See Todaro (1997), p. 66-68; Cypher/Dietz (1997), p.44-49.

³⁰ See Durth/Körner/Michaelowa (2002), p. 12. Criticism on the methodology of the HDI can also be found at Nohlen/Nuscheler (1993b), p. 93.

³¹ In 2002, 18 countries are not included in HDI calculations; the HPI covers only 105 countries. See UNDP (2002), p. 143, 159, 161.

³² See World Bank (2000a), p. 15; OECD (2001), p. 37-38.

³³ OECD (2001), p. 40.

³⁴ World Bank (2000a), p. 15.

³⁵ See World Bank (2000a), p. 34.

³⁶ See World Bank (2000a), p. 35.

³⁷ See OECD (2001), p. 38.

³⁸ See Durth/Körner/Michaelowa (2002), p. 16-20 and Nohlen/Nuscheler (1993b), p. 85-90 for measures of the human dimension. Figures are published in the WDR or HDR.

³⁹ See Durth/Körner/Michaelowa (2002), p. 26-28.

⁴⁰ See Durth/Körner/Michaelowa (2002), p. 20-23; Nohlen/Nuscheler (1993b), p. 97-100 for measures of the political dimension.

⁴¹ See World Bank (2000a), p. 36-37 and p. 135 ff. The lack of assets can be either on the people’s side (for example lack of income, low asset base), but also on the side of the state (missing safety nets).

⁴² See World Bank (2000a), p. 19-20. See also Coudouel/Hentschel/Wodon (2001), p. 54-61 for a technical discussion of vulnerability measurement.

⁴³ See UNCTAD (2002), p. 69-71; also Wagner/Kaiser (1995), p. 52-53.

⁴⁴ See UNCTAD (2002), p. 71.

⁴⁵ As the focus of this chapter is to illustrate the change in the understanding of poverty reduction, the economic theory underlying the strategies is not elaborated in detail.

⁴⁶ This structure follows Kapur/Lewis/Webb (1997), p. xv and 51-55. Similar at Kircher (2002), p. 20.

⁴⁷ See Cypher/Dietz (1997), p. 137-142 for a summary of both theories.

⁴⁸ See Kapur/Lewis/Webb (1997), p. 87, 114-115; also picked up at Kanbur/Vines (2000), p. 96.

⁴⁹ Kapur/Lewis/Webb (1997), p. 95

⁵⁰ See Kapur/Lewis/Webb (1997), p. 140-143. More on the IDA can be found at the same source p. 1119- 1160. Lending to poverty sensitive sectors especially comprised agriculture, but also education and health. Allocation to agriculture rose from 2% to 20% during this era. Total World Bank lending (IDA and IBRD) rose up to 9.5 billion dollar, compared with 3.9 billion dollar in the previous period.

⁵¹ Projects were seen as productive when they resulted in a direct payoff, which reflected a banking spirit inherent in the early years. Social projects (such as education or health) and program lending as it can be conducted through the IDA were (implicitly) regarded as not directly productive and thus discussed controversially. This tension is illustrated at Kapur/Lewis/Webb (1997), p. 141, 154-160, 196-202.

⁵² See Kanbur/Vines (2000), p. 89. For the Lewis model see Cypher/Dietz (1997), p. 148-157. For Kuznets hypothesis see Todaro (1997), p. 160-163 or Ingham (1995), p. 227-230.

⁵³ See Kapur/Lewis/Webb (1997), p. 215.

⁵⁴ See Kapur/Lewis/Webb (1997), p. 265-268 and Cypher/Dietz (1997), p. 577 for the World Bank and the basic needs approach. Basic needs should include material as well as non-material demands that give human beings the opportunity to live a full live. See ILO “Employment, Growth and Basic Needs”; Geneva 1976 and Streeten, Paul “First Things First: Meeting Basic Human Needs in the Developing Countries”; Oxford 1981. For a summary of the basic needs approach see Kircher (2002), p. 26-30.

⁵⁵ See Chenery, Hollis et al. “Redistribution with Growth”; London 1974. For a summary see Kircher (2002), p. 21-26.

⁵⁶ See Kapur/Lewis/Webb (1997), p. 309-321; Kanbur/Vines (2000), p. 96. The proportion of poverty lending rose from 5% in 1968-1970 to 29.5% in 1979-1981.

⁵⁷ See Kapur/Lewis/Webb (1997), p. 269-274.

⁵⁸ See Kapur/Lewis/Webb (1997), p. 321-329; picked up at Kanbur/Vines (2000), p. 97.

⁵⁹ See Kanbur/Vines (2000), p. 91; Kircher (2002), p. 30-34.

⁶⁰ See Kanbur/Vines (2000), p. 91; Williamson (2000), p. 252-253. Williamson remarks that the term ‘Washington consensus’ in its original sense referred to his proposal for policy advice for Latin American countries, but is used today in a misleading way indicating neoliberal of market-fundamental policies.

⁶¹ See Cypher/Dietz (1997), p. 578-580. According to them, structural adjustment lending accounted for approximately one third of total lending at the peak during this period. In contrast, lending directly targeted towards the poor fell down to approximately 20% (Kapur/Lewis/Webb (1997), p. 332).

⁶² See Kapur/Lewis/Webb (1997), p. 349-357. The first evaluations of structural adjustment lending were Cornia/Jolly/Stewart “Adjustment with a human face”; Oxford University Press 1987 or Mosely/Harrigan/Toye “Aid and Power: The World Bank and policy based lending”, Routledge 1991.

⁶³ Kapur/Lewis/Webb (1997), p. 378.

⁶⁴ See Cypher/Dietz (1997), p. 243-255 among others for a summary of the endogenous growth theory.

⁶⁵ See Kanbur/Vines (2000), p. 92-95

⁶⁶ See World Bank (2000a), p. 38-40.

⁶⁷ See World Bank (2000a), p. 38. The following presentation relies on the chapters on growth p.45-59, markets p. 61-76 and assets p. 77-96 from the same source. See also Kircher (2002), p. 56-71.