It takes a village to educate a child : a study on governance and foreign aid effectiveness in Sub-Saharan African education

(1)

A study on governance and foreign aid effectiveness

in Sub-Saharan African education

Master’s thesis

Andrea Rodriguez Frohwein (6055664) Master Political Science – International Relations

University of Amsterdam Supervisor: Farid Boussaid

(2)

“It takes a village to raise a child” – African proverb

(3)

Abstract

This thesis explores aid effectiveness and the role of governance in the Sub-Saharan African education sector. The results do not support the optimist thought that the more aid is given to education, the better the results in this sector. However, Tanzania’s education sector has clearly benefited from foreign aid. More evident proof has been found for the institutionalist view. The better the institutions of a country, the higher its education rates. The case study on Tanzania shows that bad governance – corruption and government ineffectiveness in

particular – impedes its progress on education. Despite this, Tanzania is one of the Sub-Saharan African countries that made most progress on education. This can be explained by strong donor support, government awareness of the importance of education and Tanzania’s strong civil society.

(4)

Thanks to:

Farid Boussaid Timothy van Roon Janneke Frohwein

Jesús Andrés Rodriguez Jimenez Rebeca Rodriguez Frohwein

Diana van Stijn Enea Kayange Erord Simae Maureen Roëll Tumsifu Mmari Florige Lyelu Oscar Mwambene Emmanuel Kayange Youngson Songa Eleopa Karrigo Hillary John Haule

Esther Kayange Wema Mbella Driver ‘Emmanuel’

(5)

“Education is a human right with immense power to transform. On its foundation rest the cornerstones of freedom, democracy and sustainable human development.”

(6)

Acronyms

CBO Community Based Organization CCM Chama Cha Mapinduzi

EFA Education for All GBS General Budget Support GDP Gross Domestic Product II Institution Index (World Bank) LGA Local Government Authority

LGRP Local Government Reform Program LHRC Legal and Human Rights Center LGA Local Government Authority IMF International Monetary Fund MDG Millennium Development Goal

MKUKUTA National Strategy for Growth and Reduction of Poverty of Tanzania MoEVT Ministry of Education and Vocational Training

MoFEA Ministry of Finance and Economic Affairs NGO Non-governmental Organization

ODA Official Development Assistance

PEDP Primary Education Development Program

PMO-RALG Prime Ministers’ Office – Regional and Local Government RS Regional Secretariat

SAP Structural Adjustment Program

SC School Committee

SD Standard Deviation

SPS Sector Program Support

TAMONGSCO Tanzania Association of Managers and Owners of Non-Government Schools and Colleges

TANU Tanganyika African National Union TSH Tanzanian Shillings

UN United Nations

UPE Universal Primary Education US $ United States Dollars

(7)

1. Introduction

In 2000, the United Nations (UN) articulated eight concrete targets for improving the life of people threatened by poverty, hunger and disease before 2015: the Millennium

Development Goals (MDGs). It aims to combat major diseases, eradicate extreme poverty, reduce child mortality and promote gender equality among others (UN MDG Report, 2014, p.17). The MDGs are an intermediate target, not a final goal: even if all goals are met, there will still be an enormous amount of extreme poverty in de world (UN MDG Report, 2014, p.17). In order to make a significant step towards reaching the MDGs, a two-pillar strategy was adopted in Monterrey (Mexico) in 2002 (IMF, 2015). This strategy includes sound

policies, good governance, larger and more effective international support and an international economic and trade environment for development.

Major achievements have been made across all MDGs: some targets were even reached ahead of the deadline. Unfortunately, some goals have not been met, including the second one. This goal is set up to “ensure that, by 2015, children everywhere, boys and girls

alike, will be able to complete a full course of primary schooling” (UN MDG Report, 2014,

p.30). Even though primary education is still not universal, considerable progress has been made in many countries, particularly regarding enrolment rates in primary schooling.

Sub-Saharan Africa scores significantly lower than all other regions in the developing world regarding equitable access to education (MDG Report Africa, 2014, p.34). At the same time, the greatest improvement on education was made in this region: the primary school enrolment rate increased by 18% between 2000 and 2012, while the population of primary school-aged children grew with 35% during this period (ibid., p.17). Although these countries performed remarkably well in terms of effort and progress, the gap between the current achievement and the 2015 target still remains large. For example, the literacy rates among youngsters between 15 and 24 years rose slightly from 67% to 70% in the period 2000-2010 and dropped again to 68% in 2012 (UNESCO Database, 2015). Moreover, the primary school completion rates decreased: 58% of the pupils starting grade one reached the last grade of primary education in 2011, while this rate was 61% in 2000 (UN MDG Report, 2014, p.18).

The UN point out several causes for failing to reach MDG 2. The lack of interest for improving education and insufficient donor aid are seen as the most important reasons (Sachs, 2013, p.2; MDG Report Africa, 2014, p.34). Another important cause is conflict: an estimated 50% of all out-of-school children of primary school age live in conflict-affected areas (ibid., p.17). Sub-Saharan Africa accounts for 44% of these children. Other reasons are late entry to

(10)

school, travelling long distances from home to school, household poverty, the combination of work and study, and opportunity costs. 1

The MDGs are both criticized and applauded by economists, political scientists and experts in the field of development all over the world. Jeffrey Sachs, one of the most important developers of the MDGs, stresses that the only obstacle of this development program is the insufficient allocation of aid (Sachs, 2013, p.2). He states that poor countries, including the tropical parts of Sub-Saharan Africa, are stuck in a ‘poverty trap’: too poor to achieve economic or any other kind of growth (Sachs et al., 2004, p.121). Only a ‘big push’ of aid increases public investments drastically and allows them to achieve the MDGs.

On the other hand, William Easterly (2003, p.34) argues that the focus should not be on the quantity of aid, but on the quality. He disapproves of Sachs’ theory by stating that the idea of foreign aid leading to economic growth is simplistic and generalizing (ibid., p.40). Not only does this idea imply that the developing world is homogenous, it is also unfair to set targets for regions that are far more underdeveloped than other parts of the world.

Supporters of the institutional approach like David Dollar and Dani Rodrik, criticize Sachs for underestimating the role of governance in the effectiveness of foreign aid. These economists state that foreign aid directed at well-governed countries is spent more effectively than foreign aid directed at poorly-governed countries. In order for countries with weak institutions to achieve the MDGs, significant improvements in governance are required. However, research has shown that institutional reform cannot be achieved with foreign aid (Burnside & Dollar, 2004, p.20). Therefore, institutionalists plead for the allocation of aid only to countries with relatively good institutions.

This thesis tests the influence of foreign aid and governance on the progress regarding

education in Sub-Saharan Africa. The case study on Tanzania explores the role of foreign aid and bad governance in the provision of education in this country. The results do not support the optimist thought that the more aid is given to education, the better the results in this sector. More evident proof has been found for the institutionalist view. The better the institutions of a country, the higher its education rates. Also, the case study shows that government

ineffectiveness and corruption impede the progress Tanzania made on education. The decentralization of the government has not proven to address this problem yet. Despite this,

1

This piece is partially retrieved from my earlier work on this topic: final paper “The more, the better? Exploring foreign aid effectiveness and the MDG on education in Sub-Saharan Africa” (written by Andrea Rodriguez Frohwein, January 2015)

(11)

Tanzania is one of the Sub-Saharan African countries that made most progress on education. This can be explained by three factors. First, Tanzania received large amounts of foreign aid. Second, the Tanzanian government is aware of the important role of education. Third,

Tanzania has a strong civil society that contributes heavily to the provision of education.

This research is relevant on a societal, scientific and political level. First and most important, ensuring that every child enjoys basic education is crucial because it is the key to inclusive growth, equity, social transformation and sustainable development (MDG Report Africa, 2014, p.4). It is essential to discover which factors contribute to a sound education sector providing every individual with reading, writing and other skills. Moreover, this research contributes to the theoretical debate on foreign aid effectiveness: it tests the two most prominent and contrasting theories in this field. Finally, it provides in-depth information on the role of governance and aid effectiveness in ‘donor darling’2 Tanzania.

This thesis can broadly be divided into six parts. First, an overview is given of the existing literature in the field of the effectiveness of foreign aid. The second part explains which theories and hypotheses are tested. Third, the methodology and operationalization of variables is explained. Fourth, the hypotheses are tested using a quantitative method. Fifth, the in-depth case study gives insights on the role of foreign aid and governance on the education sector in Tanzania. Finally, a conclusion is drawn from the results.

2

(12)

2. Literature review

The discussion on aid effectiveness is a controversial one that goes back decades (Radelet, 2006, p.3). The debate can be roughly divided into three groups: the pessimists, the optimists and the institutionalists. First, it should be emphasized that the term ‘foreign aid’ or ‘aid’ is defined as “financial flows, technical assistance, and commodities that are designed to

promote economic development and welfare as their main objective” (ibid., p.4). Only official

development assistance (ODA), aid provided by donor governments to low- and middle- income countries, is taken into account. The term ‘growth’ or ‘economic growth’ refers to the annual percentage growth rate of GDP (gross national product) at market prices based on constant local currency (World Bank, 2015). ‘Governance’ and ‘institutions’ both refer to the same phenomenon: “the traditions and institutions by which authority in a country is

exercised” (World Bank Governance Indicators, 2015). Most of the articles that are discussed,

focus on foreign aid and economic growth in developing countries all over the world, while other authors chose to focus on (Sub-Saharan) Africa. This region is being studied frequently because it is the poorest continent in the world. Sub-Saharan African countries also grow generally more slowly than other developing countries (Sachs et al., 2004, p.122), while the former is one of the biggest recipients of aid (World Bank, 2015).

2.1. The pessimist approach

Pessimists like Peter Bauer and William Easterly argue that aid has enlarged government bureaucracies, maintained bad governance and enriched the elite in poor countries (Bauer & Yamey, 1982; Easterly, 2003; Radelet, 2006, p.3). The economist Peter Bauer argues that foreign aid is not essential for economic growth: it has never been necessary for the development of any country, anywhere (Bauer & Yamey, 1982, p.303). Economic achievement depends on people’s own motivations, ways of life and on institutions and policies. Foreign aid is not the way to achieve growth because economic achievements produce assets and money, not the other way around. The rapid growth of the Asian Tigers (Hong Kong, Singapore, South Korea and Taiwan) between the 1960s and 1990s illustrates this point: these countries have been able to develop without much help from rich countries. Bauer also states that aid increases the money, patronage and power of the ruling elite of recipient governments. Developing countries become highly politicized, and people become dependent on political and administrative decisions. According to Bauer, this situation promotes conflict. In addition, aid can lead to the ‘Dutch disease’: it helps create or maintain

(13)

overvalued exchange rates, which results in higher prices and expensive labor. Local

companies cannot compete with international companies anymore, which leads to a stagnation of economic growth.

Bauer’s view is problematic for two reasons. First, determining a causal relationship between aid and growth is difficult. Assuming that foreign aid impedes growth is thus problematic. Second, Bauer’s view on foreign aid is relatively old, which makes the current relevance of his arguments questionable. For example, he does not acknowledge the

difference between good and bad governance. Bauer does not explain the effect of aid directed at countries with good institutions and low levels of corruption.

However, his point on foreign aid being mostly imposed by developed countries is fair. The Third World hardly has a say in what kind of aid they need in order to develop.

The scholar William Easterly (2003, p.23) acknowledges that aid led to noticeable changes, like the elimination of smallpox, the near elimination of river blindness, the general rise in life expectancy and fall in infant mortality (ibid., p.36). Still, he states that foreign aid does not lead to economic growth and is, most of the time, wasted. He argues that aid agencies have misspent their effort looking for an innovative way to enable growth in developing countries (ibid., p.40). Mostly, increasing the volume of aid seems to have the highest priority, while the focus should be primarily on the quality of aid (ibid., 34). The idea of one homogenous developing world that will take off as a result of an aid boost is a ‘heroic simplification’ of reality (ibid., p.40). By stating this, Easterly clearly disapproves of Sachs’ concept of a ‘big push’ of foreign aid in order to enable growth. In addition, Easterly says the MDGs are unfair to Africa. Achieving the MDGs is less likely in this continent, since it is far more

underdeveloped than other parts of the world (Easterly, 2009, p.36). This makes African successes, like making progress which is in line with or above historical or contemporary experiences in other regions, look like failures.

Easterly (2003, p.40) also states that aid agencies do not have much incentives to achieve results, because the results are mostly unobservable. This assumption contradicts his earlier point on aid leading to important changes. Supposedly, the point he is trying to make is that the causal relationship between aid and growth is difficult to determine. Easterly (2003, p.35) explains his findings as the deteriorating economic performance of Africa despite rising aid (figure 1). Still, it is important to stress that this figure shows aid and growth in Africa 30 years before the initiation of the MDGs: from 1970 to 1999. It would be interesting to explore if this trend persists from 2000 onwards.

(14)

Figure 1 Aid and growth in Africa (Easterly, 2003, p.35)

In addition to the current problematic philosophy behind aid, Easterly points out several practical problems. First, donors sometimes provide foreign aid for serving their own interests (e.g. rewarding allies, promoting donor country exports or fighting drug trafficking). In addition, the governments of poor countries often do not have motives to increase the productivity of the poor, as it might encourage political activism that threatens the political elite. Placing conditions on loans and aid is not an effective way of increasing aid

effectiveness, because agencies often provide additional future loans without taking the performance on previous loans into account. Finally, the lack of evaluation of projects is also a problem. Currently, aid organizations are reluctant to admit their failures because it could lead to bad publicity and less funding (Easterly, 2003, p.35).

In his future vision for foreign aid, Easterly (2003, p.39) pleads for a more careful look at the variety of institutions, cultures and histories of poor nations. Studying every developing country individually, critically evaluating what works and what does not, will increase the probability of creating a successful strategy (ibid., p.38). The main goal of foreign aid should be benefiting some poor people, some of the time. The focus should be more on quality instead of quantity, and aid agencies should give more priority to evaluating their projects honestly.

2.2. The optimistic approach

Optimists state that foreign aid generally has positive effects, including economic growth (Radelet, 2006, p.11). Scholars such as Jeffrey Sachs and Joseph Stiglitz argue that large

(15)

amounts of foreign aid targeted at well-governed countries lead to poverty reduction and economic growth (Sachs et al., 2004; Stiglitz, 2002; Radelet, 2006). The failure of aid is mostly due to the weaknesses of donors, not recipients. Jeffrey Sachs is one of the most prominent and controversial economists in sustainable development. The American professor is Special Advisor to UN Secretary-General Ban Ki Moon and played an important role in setting up the MDGs. The way Sachs advocates the geography approach lies at the heart of his thoughts on foreign aid. He states that poverty can be largely explained by geographic factors like proximity to fossil fuels, distance to a port and tropical diseases (McCord and Sachs, 2013, p.2,16). For example, McCord and Sachs (2013, p.16) find that the distance to a port and fossil fuel reserves explain 58% of the variation in per capita income. The variables hydroelectric production, malaria ecology, land quality and temperature increase the variation even more, up to 75%. Sachs et al. (2004, p.130) give several reasons for Sub-Saharan Africa being stuck in poverty. For example, many countries in this region have very high transport costs because they are landlocked. Furthermore, Sub-Saharan Africa has a small market size, a tropical climate and thus low-productivity agriculture and a very high disease burden, unfavorable geopolitics and a very slow diffusion of technology from abroad. Overall, the fact that Sub-Saharan Africa is the poorest region in the world can thus be explained by its

unfortunate geographic position (Sachs et al., 2004, p. 117).

In contrast with supporters of the institutional approach, Sachs argues that bad governance is not the crucial cause of poverty. Although Sachs and his colleagues (2004, p.120) acknowledge that governance is a problem, they state that Africa’s development challenges run much deeper. The scientists use the findings of this study to strengthen their position and conclude that many parts of Africa are well-governed but still stuck in poverty. After controlled for income, only one country (Angola) out of 33 receives the score ‘poor’ (high level of corruption). All other countries score ‘average’ (moderate level of corruption) or ‘good’ (low level of corruption). Still, Sachs et al. (2004) do not clarify why corruption is used as the only indicator for governance. Moreover, it is not clear where the boundaries between these categories are set. For example, the Republic of Congo scores low on control of corruption on the World Bank Institutions Index (20 on a scale of 100 in 2002) and is still given the score ‘average’ (Sachs et al., 2004, p.119).

The scholars explain that extreme poverty leads to low national saving rates, which lead to low or negative economic growth rates. Africa does not attract foreign capital, such as foreign direct investment, because of its poor infrastructure and weak human capital. Sachs et al. (2004, p.122) state that what is needed is a large amount of foreign aid which will increase

(16)

public investments drastically (ibid., p.122). They do not plead for endless flows of increased aid, but rather for increased aid as an exit strategy from the poverty trap. This ‘big push’ will produce a rapid increase in Africa’s productivity and will enable them to achieve the MDGs by 2015. These goals do not necessarily lead to economic growth, but they do address key sectors in which major improvements are needed and achievable. As Sachs et al. (2004, p. 147) put it:

“We are not arguing that the MDGs are sufficient for escaping poverty. We are arguing that

they are a sensible, internationally agreed way station on the road to eliminating extreme poverty and ending Africa's poverty trap.”

They also state that recipient countries should use aid to invest in targeted areas with the help of local non-governmental organizations (NGOs). Increased aid does not allow the recipient country to decrease its own efforts because the rates of public investment in the targeted areas are already very low. Moreover, their response to critics who fear for the Dutch disease is that it should be managed carefully. Aid should be directed towards increasing supply-side

productivity (e.g. agriculture, human capital) rather than towards an explosive increase in consumption which brings the tradable goods sector in jeopardy (ibid., p.172).

It should be noted that the general line of argumentation behind the MDGs is quite one-dimensional. Sachs et al. (2004) imply that increasing the quantity of aid drastically is a suitable strategy for every developing country in Africa. In addition, not achieving the MDGs is due to the insufficient allocation of aid. Thus, Easterly’s critique on erroneously placing more emphasis on quantity instead of quality is valid on this point. Still, Sachs and his colleagues do acknowledge that public investment programs need to be developed based on bottom-up assessments of needs rather than donor countries imposing targets on developing countries (ibid., p.168). As a result, the necessary external resources should be allocated to well-governed countries. This indicates that although the philosophy behind the MDGs seems to be mostly concerned with increasing quantity and reaching targets, it does not implore for allocating large amounts of aid randomly.

The American economist Joseph Stiglitz (2002, p.1) also argues that aid to developing countries with relatively good policy environments has significantly increased growth and reduced poverty. Even though not all the money is well spent, foreign aid generally is

(17)

find different mechanisms that bypass the government and to focus assistance on projects that directly help the poor. Even though these thoughts are in line with Sachs’ vision, Stiglitz and his colleague Michael Doyle argue that the MDGs do not represent a complete or

comprehensive vision of human development and lack a vision of equitable development (Doyle & Stiglitz, 2014, p.3). A certain degree of equality lies at the heart of economic growth and development. In contrast, extreme inequality has harmful economic (low

aggregate demand, rent-seeking, low investment in public goods) and social effects (sense of unfairness, little cohesion and mobility). Therefore, Stiglitz and Doyle propose an additional MDG: the elimination of extreme inequality at the national level in every country (ibid., p.6). Even though Stiglitz’ point on focusing projects on the poor is fair, it is a short-term solution. The scholar does not elaborate on how aid can be effective on the long-term, eventually making poor countries less dependent on aid.

2.3. The institutional approach

Over the last decade, the institutionalist approach has become prominent in the debate on foreign aid (Dollar & Levin, 2006, p.2034). Institutionalists state that foreign aid is more effective when targeted at well-governed countries. This thought is anchored in the UN Monterrey Consensus (Dollar and Levin, 2006, p.2034). Since its adaption in 2002, this consensus has become an important point of reference for international development cooperation.

Theorists like David Dollar, Dani Rodrik and Nancy Birdsall stress the importance of good political and economic institutions for economic growth (Collier & Dollar, 2002; Rodrik, Birdsall & Subramanian, 2005). The institutional view, as explained by Acemoglu and Robinson (2012, ch.4), constitutes the basis for this approach. They state that institutions are the most important factors in explaining poverty. Inclusive institutions allow and

encourage civil participation, feature secure private property, have an unbiased system of law and a provision of public services and permit the entry of new businesses (ibid., ch.3). They create inclusive markets, which are mostly accompanied by technological improvement and (high-quality) education. Moreover, inclusive institutions give citizens the chance to get education and choose their own career paths. Having the freedom to develop their talents increases their productivity and stimulates the efficiency and growth of firms. In contrast, extractive institutions extract incomes and wealth from one subset of society to benefit a different subset. In countries with extractive economic institutions, people often get little or no education, which deprives them from developing their talents and forces them to do

(18)

unskilled labor. In sum, extractive economic institutions impede the development of countries, whereas inclusive ones stimulate development.

Rodrik and Birdsall, two advocates of the institutional approach, argue that good institutions lie at the heart of economic growth and development of countries (Birdsall, 2004, p.6; Birdsall, Rodrik & Subramanian, 2005, p.137). They also state that developed countries should internalize the thought that sustainable progress is in the hands of developing countries themselves (Birdsall, Rodrik & Subramanian, 2005, p.137). In addition, Rodrik (2013, p.1) postulates that although it is hard to determine what the results of the MDGs exactly are, they certainly have raised awareness of poverty and have boosted foreign aid flows. Still, he thinks that the MDGs should be more focused on what developed countries can or should do, instead of only setting targets for developing countries. The MDGs imply that the only things that are missing in achieving them are resources and political will. Rodrik (2013, p.2) states that this assumption is false: significant improvements in governance and political institutions are required before such goals can be achieved. However, real reform cannot be bought with money. A new global compact should thus be more focused on the responsibilities of rich countries and emphasize policies beyond aid and trade that have a positive impact on poor countries’ development prospects (ibid., p.3).

Rodrik and Birdsall also argue that the failure of foreign aid is mostly due to donors, not recipients (ibid., p.152; Birdsall, 2004, p.). Birdsall (2004) has formulated the ‘seven deadly donor sins’ that could cause the failure of aid. Two examples are impatience for results and failure to evaluate. Donors should realize that building good institutions is not a short-term achievement. In contrast, aid is budgeted annually in the donor countries while nation-building takes predictable and continuous support over many years (ibid., p.7). As donors also are impatient for results, they mostly do not want to spend money on high-risk programs such as police and security assistance, because the immediate benefits are less visible than for example humanitarian assistance. Moreover, development assistance agencies hesitate to advertise the limits or failures of their projects, because they fear for the reduction of future funding (ibid., p.13). Also, aid professionals who work with the urgent needs of people in poor countries do not see costly project evaluations as a priority.

Burnside and Dollar (2000, 2004) tested the assumption that foreign aid directed at well-governed countries is spent more effectively than aid directed at badly-well-governed countries. Based on two large cross-country statistical researches examining developing countries all

(19)

over the world, they conclude that foreign aid has little or no effect on growth in countries with poor institutions and policies. In contrast, the correlation between foreign aid and good fiscal, monetary and trade policies is significant (figure 2). Still, they note that additional sources of information, such as theory, case studies and project evaluation should also be taken into account (Burnside & Dollar, 2004, p.18). Finally, the scholars state that, in the past, aid has not systematically led to the improvement of institutions and policies (ibid., p.20). Allocating aid only to countries with relatively good institutions will increase the probability that reforms are successful and politically sustainable. Still, Burnside and Dollar (2004) do not elaborate on how foreign aid targeted at badly-governed countries could be effective. Following the line of thought of Acemoglu and Robinson, these countries are not only poor but lack important basic needs such as health care and education. Citizens of these countries are thus in serious need of aid.3

Figure 2 Growth, Aid, and Policy (Burnside & Dollar, 2004, p.4)

The institutionalist approach has shaped the allocation of aid (Dollar & Levin, 2006, p.2034). Dollar and Levin (2005, p.13) found that the probability that a country receives aid is higher for countries with good institutions. They also conclude that the success of projects that are sponsored with foreign aid depends primarily on the quality of institutions in the recipient country. The scholars plead for the allocation of aid to poor countries with good institutions,

3

This piece is partially retrieved from my earlier work on this topic: final paper “The more, the better? Exploring foreign aid effectiveness and the MDG on education in Sub-Saharan Africa” (written by Andrea Rodriguez Frohwein, January 2015)

(20)

as aid resources have the greatest impact on development in these countries. Dollar and Levin (2006, p.2044) find that in the researched period (1984-1989), aid was mostly allocated to democratic countries with bad economic institutions. Aid was often not well used in these countries. From 1990 onwards, multilateral aid was mostly allocated to countries with good economic institutions. Bilateral aid did not show this correlation, which means that good governance is a less important criterion for the allocation of this type of aid. A possible explanation for this is that other factors, such as political interests and colonial past, explain the allocation of bilateral aid.

In contrast, Collier and Dollar (2002, p.1475) find that, in 1996, aid was still mostly targeted to weak policy environments and to countries that do not have severe poverty problems. They also conclude that the actual allocation of aid is significantly different from the poverty- efficient allocation. This is partly due to the encouragement of policy reform and for several political and strategic reasons (ibid., p. 1497). Allocating aid to badly-governed countries could be justified by policy improvement, but Collier and Dollar find the opposite: finance is ineffective in inducing either policy reform or growth in a bad policy environment. Collier and Dollar (2002, p.1497) also find that the current allocation of aid lifts around 10 million people per year out of poverty. This number would nearly double if aid would be allocated based on the GDP of a country. Even though the main conclusion of this study contradicts Dollar and Levin’s findings, the result that aid directed at badly-governed countries does not lead to policy reform is a valuable addition to these studies. Moreover, it confirms Rodrik’s earlier statement that reform cannot be bought with money. Still, it is interesting to explore whether foreign aid directed at Sub-Saharan Africa is allocated based on governance or based on poverty rates.

In sum, institutionalists state that good governance is crucial for economic growth. Moreover, foreign aid targeted at well-governed countries is spent more effectively than foreign aid targeted at poorly-governed countries. Aid should thus be allocated based on the quality of institutions.

The next chapter explains why the optimistic and the institutional approach are the most plausible theories. Based on assumptions derived from these two theories, hypotheses are formulated.

(21)

3. Theory

In this chapter, the hypotheses that are tested in the quantitative analysis are made explicit. First, it should be noted that the academic debate is mostly concerned with the correlation between foreign aid and economic growth, while this thesis is not. The reason for this is that the causality between these two variables is problematic. It is difficult to determine whether foreign aid leads to economic growth because there is a wide variety of intervening variables which distort the correlation. For the sake of the analysis, this research is focused on the influence of specific foreign aid flows (ODA for primary education) on specific areas (primary education rates).

3.1. Hypotheses

In order to obtain a broader view on the effectiveness of foreign aid for primary education, it is useful to explore the allocation of aid first. Based on the findings of Collier and Dollar (2002) and Dollar and Levine (2006), one can expect that the allocation of aid in the period 2000 to 2010 is not based on the GDP of a country. Moreover, the quality of institutions of a country is a plausible explanatory factor for the allocation of aid. In other words, aid is mostly directed at well-governed countries. In addition, the conclusions of both studies fit the

institutional paradigm that has been present among development organizations in the last two decades. The following hypotheses are tested:

H1: The GDP per capita of a country is not associated with the amount of ODA for primary

education that is allocated to that country.

H2: The quality of institutions in a country is associated with the amount of ODA for primary

education that is allocated to that country.

Even though this thesis is not concerned with the effect of foreign aid on economic growth, it is interesting to generally illustrate this assumption for Sub-Saharan Africa in the period 2000-2010. As figure 3 shows,the postulation that foreign aid impedes economic growth is not plausible for this region in this period. Both the net ODA received per capita and the GDP per capita have increased linearly, with one outlier in 2005 in the former*. Moreover, even if it was not ODA that led to economic growth, there is no reason to discard foreign aid as a whole: other valuable results (e.g. poverty alleviation, eradication of diseases) can be reached. Thus, it is more interesting to test Sachs’ theory, which has been prominent in setting up the MDGs.

(22)

Figure 3 Net ODA received per capita and GDP per capita PPP, in US dollars, 2000-2010 (data: World Bank, 2015)

Sub-Saharan Africa 0 5 10 15 20 25 30 35 40 45 50 2000 2002 2004 2006 2008 2010 0 500 1000 1500 2000 2500 U S

$ Total ODA received

GDP per capita PPP

* ODA was exceptionally high in 2005 due to large Paris Club debt relief operations to Nigeria (OECD, 2007).

Sachs’ assumptions on the relationship between quantity and effectiveness of foreign aid are plausible for an important reason. As explained earlier, there has been a major improvement on primary school enrolment and youth literacy rates in Sub-Saharan Africa starting the year in which the MDGs were initiated until ten years later. Figure 4 shows the way the education indicators developed in this period.

Figure 4 ODA for education received (in US billion $) and primary education rates in

Sub-Saharan Africa (in % of relevant age), 2000-2010 (data: World Bank, 2015; OECD, 2015) Sub-Saharan Africa 50 55 60 65 70 75 80 85 90 95 100 2000 2002 2004 2006 2008 2010 % o f re le v an t ag e 2 2,5 3 3,5 4 4,5 U S b il li o n

$ School enrolment primary

Youth literacy rate Primary completion rate ODA for education received

(23)

Sachs states that the fact that some MDGs, including the second one, were not achieved before 2015, is because creating a sufficient global fund on education proved impossible (Sachs, 2013, p.2). His following statement captures the line of thought on this:

“[…] creating a sufficient global fund for education proved impossible. The cause of

universal access to education turned out to be a policy orphan, unable to mobilize the same kind of donor interest as disease control did. Yes, modest aid helped millions of children attend primary schools, but because of the shortfalls, those schools often lacked basic

materials, trained teachers and even safe water. Millions of other kids remain out of school.”

Based on this philosophy, it can be assumed that countries that receive more foreign aid for education achieve more progress on education than countries that receive less foreign aid for education. The third hypothesis is:

H3: The more ODA for education a country receives, the more the progress on primary

education rates a country achieves.

Institutionalists state that governance is crucial when it comes to the development of a country (Burnside and Dollar, 2000, 2004; Dollar and Levin, 2005; Collier and Dollar, 2002). Good institutions lead to economic growth and the provision of public goods such as infrastructure, healthcare and education. In contrast, bad institutions hamper the development of a country, impede economic growth and maintain poverty. As sound policies are needed in order to achieve progress, badly-governed countries often do not invest effectively in public goods (including education). In these countries, foreign aid is more likely to enrich the ruling elite. Even Sachs, who is not a proponent of the institutional approach, acknowledges that aid directed at well-governed countries is more effective than aid directed at badly-governed countries (Sachs et al., 2004, p.120). Based on these assumptions, the following hypothesis is tested:

H4: The better the institutions of a country, the more progress on primary education rates a

(24)

4. Methodology and case selection

In this study, quantitative and qualitative methods are combined. Throughout the thesis, the focus is on primary education. The reason for this is that including more levels of education goes beyond the scope of this research. In addition, data on primary education is better available than other levels. However, the case study does allow for the inclusion of pre-primary and secondary education. When relevant, these levels are included in the results.

4.1. Quantitative study: Sub-Saharan Africa

In the first part, an overview of descriptive statistics on Sub-Saharan Africa is provided. Moreover, the hypotheses are tested using a simple regression analysis. A quantitative study allows for the discovery of correlations between variables. As a result, general conclusions on Sub-Saharan Africa as a whole can be drawn. Data for this analysis are retrieved from the UNESCO database (2015) and the World Bank database (2015).

The use of quantitative data in this study has two important limitations. First, data are missing in many developing countries in this region (UN MDG Report, 2014, p.36). Only a selection of Sub-Saharan African countries can thus be used. Second, intervening variables are not included. The reason for this is that including more variables goes beyond of the scope of this thesis and the knowledge of the author. However, one important variable is excluded from the analysis: war. In the quantitative chapter (paragraph 5.1.), more detailed information on this is given.

4.2. Case study: Tanzania

The qualitative part of the study consists of a case study. Due to the size and scope of this research, only one country will be explored in-depth: Tanzania. This has the important disadvantage that a comparison between two or more countries is not possible. Still, the careful selection of the case allows for a valuable addition to the statistical part of the research. Moreover, it tries to compensate for the weaknesses of the quantitative analysis.

Tanzania is an interesting case for several reasons. First, the country has made above average progress on education in comparison to other Sub-Saharan African countries (UN MDG Tanzania, 2010, p.15). The case study explains how Tanzania has been successful at increasing its education rates and how the country has overcome its obstacles. Second, Tanzania is the second largest aid recipient of the region. It is explored if foreign aid is used effectively in this country and how Tanzania and its donors face challenges in this regard.

(25)

Third, Tanzania is not well-governed. The case study explains what the consequences are of

bad governance for the effectiveness of foreign aid, and for the education sector.

For the case study, research was conducted during a field trip to Tanzania. Interviews with an Education Specialist of UNESCO Tanzania, a Development Consultant of NGO eMJee and a Program Officer of HakiElimu were conducted in Dar es Salaam. These interviewees are selected based on their expertise in the field of development and/or

education. Moreover, five schools were visited in Mbeya and Sumbawanga. At these schools, directors, head teachers and teachers from pre-primary, primary and secondary schools were interviewed. One school (Mwangaza secondary government school) was contacted trough e-mail. A short description of each school is provided in tabulation 1.

The combination of professionals in the field of education and development and people working in the field of education allows for an interesting comparative analysis.

All interviews have been recorded. Afterwards, they were transcribed and divided into topics. Moreover, government reports and researches conducted by NGOs (HakiElimu, eMJee and LHRC) are used for the analysis.

Tabulation 1 Basic data on visited schools in Tanzania (amounts are in Tsh4)

Name Level Type Place Pupil/

teacher ratio School fee (/year/ pupil) Average teacher salary (/month) Mbalizi Primary Private Mbeya 31:1 900,000 700,000 Jifunzeni Secondary Private Mbeya 25:1 700,000 700,000 Iyera Secondary Public Mbeya 37:1 20,000 500,000 African Rainbow Secondary Private/ Church supported Sumbawanga ? 500,000 300,000 FF4Future Pre-primary NGO/ Church supported Sumbawanga 35:1 5,000 75,000 (+ part of school fees) Mwangaza Secondary Public Sumbawanga 28:1 20,000 500,000

Mbeya is a city located in the south of Tanzania and is part of the Mbeya region. The city has approximately 385.000 inhabitants (City Population, 2012). Sumbawanga is a small city located in the Rukwa region in southwest Tanzania. Its population was 209.000 in 2012 (City

4

(26)

Population, 2012). The tabulation shows that the two public schools have school fees of Tsh 20,000 ($10) a year per pupil and an average teacher salary of Tsh 500,000 ($250) a month. However, the difference between the pupil/teacher ratio at these schools is large (37:1 as opposed to 28:1). The two private schools differ slightly regarding the pupil/teacher ratio, school fees and teacher salaries. This can be explained by the difference in level. Still, it is remarkable that the pupil/teacher ratio of Mbalizi is worse, while the school fees are higher. The church supported schools deviate heavily from the other schools in terms of teacher salary. At these schools, teachers get paid below the government minimum of Tsh 500,000 (secondary school) and Tsh 200,000 (pre-primary school).

4.3. Operational definitions

The variables that are used in this thesis are GDP per capita, ODA for primary education, progress on education rates and governance. Primary education rates is an additional variable that is needed to determine the progress on education rates. All amounts in the quantitative study are in US dollars ($) and all amounts in the qualitative study are both in Tanzanian Shillings (Tsh) and in US dollars ($).

GDP per capita is the gross domestic product divided by the population of a country in a given year (World Bank, 2015). GDP is the sum of gross value added by all resident

producers in the economy plus any product taxes and minus any subsidies not included in the value of the products (ibid.). As this number fluctuates moderately, the variable is measured as the average of de GDP per capita PPP5 in the years 2000, 2005 and 2010. ODA received for primary education is the total amount of foreign aid that a country receives for basic education per primary school-aged child. As this amount fluctuates heavily, the average of the years with available data in the period 2000-2010 is used. In order to make a valid comparison between countries regardless of the population size, the amount per primary school-aged child is used instead of the total amount.

The variable ‘primary education rates’ consists of three elements: 1) net enrolment ratio in primary education, 2) percentage of children that complete primary school, and 3) youth literacy rate. The first element is measured as a percentage of the population of official primary education age, the second is expressed as the percentage of children of the relevant age, and the last is expressed as the percentage of people aged 15-24 years (UN MDG Report, 2014, p.6). It is important to not only measure if children are enrolled in a primary school,

5

GDP per capita PPP (purchasing power parity) reflects the relative value of different currencies. The GDP per capita is converted to international dollars using purchasing power rates.

(27)

because this does not indicate whether they stay in school. The completion rates give valuable additional information on this. The youth literacy rate says something about the quality of education: every child should be able to read when it has completed primary school. Progress on primary enrolment rates is the primary enrolment rate in 2010 minus the primary

enrolment rate in 2000. Progress on primary completion rates is the primary completion rate in 2010 minus the primary completion rate in 2000. Progress on youth literacy rates is the youth literacy rate in 2010 minus the youth literacy rate in 2000.

In order to measure whether a country has good or bad institutions (governance), the World Bank Institution Index, developed by Kaufmann, Kraay and Zoido-Lobatón (1999), is used. This index is based on six indicators: 1. Voice and accountability (degree of democracy, freedom of expression, freedom of association and free media), 2. Political stability and absence of violence/terrorism (likelihood of destabilization of the government, degree of violence and terrorism), 3. Government effectiveness (quality of public and civil services, policy formulation and implementation, and credibility of the government’s commitment to such policies), 4. Regulatory quality (ability of the government to formulate and implement sound policies and regulations), 5. Rule of law (confidence and obedience to rules society, the quality of property rights, the police and the courts, and the likelihood of crime and violence), and 6. Control of corruption (the extent to which public power is exercised for private gain and the interests of the state by elites and private interests). These measures are based on 100 different variables obtained from 31 data sources which are retrieved from survey

respondents, non-governmental organizations, commercial business information providers and public sector organizations (ibid., p.2). Each country receives a score between 0 and 100 on each of the six measures in a given year. The Institutions Index score (after this: II score) of a country is the average of the six measures. As this rate does not fluctuate heavily, the average II score of a country is measured based on the scores in 2000 and 2010. If this score is lower than 33, the country is badly-governed. A score between 33 and 64 receives the label

‘average’. If the II score is higher than 64, the country is well-governed. Three indicators are also analyzed separately: government effectiveness, regulatory quality and control of

corruption. The first is the most important one for this research because education is a public service. Thus, the quality of policies in the field of education and the way they are

implemented are crucial for the quality and achievements in education. Regulatory quality is also relevant, since this indicator is closely related to the former. It is also important how a country deals with corruption, as this heavily influences government budgeting.

(28)

5. Quantitative analysis on Sub-Saharan Africa

In this section, results based on quantitative data on foreign aid for education, (progress on) education rates and quality of institutions are presented. The first paragraph elaborates on section 4.1. in the previous chapter: the method. The second section gives an overview of the descriptive statistics of the four variables that are used in the analysis. In paragraph 5.3., the hypotheses are tested.

5.1. Method

In order to increase the reliability and accuracy of the analysis, Sub-Saharan African countries in war during the period are left out the analysis. The reason for this is that conflict impedes the development of countries, which also influences its education rates. Institutions in

conflict-affected areas are weak and families often have to flee, which makes it impossible for the government to make progress regarding education. According to the Conflict Barometer, political conflict is “[…]positional difference between at least two assertive and directly

involved actors regarding values relevant to a society (the conflict items) which is carried out using observable and interrelated conflict measures that lie outside established regulatory procedures and threaten core state functions, the international order, or hold the prospect of doing so” (Heidelberg Institute for International Conflict Research, 2011).

Based on its intensity, there are five levels of conflict: dispute, non-violent crises, violent crises, limited war and war. Countries in war in the period 2000-2010 are the Democratic Republic of Congo, Sudan, Somalia, Chad, Burundi, Central African Republic, Liberia and Côte d’Ivoire (ibid., 2001-2011). It should be noted that Mali, Guinea-Bissau, Senegal, Angola, Ethiopia, Nigeria, Uganda, Kenya and Rwanda experienced limited war in this period, but are however included in the analysis. This might influence the results.

Moreover, Djibouti, Gabon and Sierra Leone are left out of the analysis because data for these countries are missing. As a result, 35 out of 46 Sub-Saharan African countries are analyzed. As the selection of countries is different from the ones used by the UN, the results of this analysis are not completely consistent with the rates reported by this organization.

The data for 4 countries on progress on primary education rates are not complete. This should however not influence the analysis. The time frame that is used in this research is from 2000 (year of initiation MDGs) to 2010 (most recent year in which data are available). If data for one of these years is unavailable, a deviation of one year from 2000 and a deviation of three years from 2010 is allowed. As the received ODA for primary education per country is

(29)

not available for every year, the years 2006, 2007 and 2008 could not be included in the analysis. The average of the years 2000-2005 and 2009-2010 is used.

5.2. Descriptive statistics

In this paragraph, an overview is provided of the descriptive statistics of the four variables that are used in the analysis.

5.2.1. GDP per capita and ODA for education in 2010

The average GDP per capita 2000-2010 in Sub-Saharan Africa varies from $694 to $26,534 per year. Out of all 35 countries, the GDP per capita for 6 countries is lower than $1000, for 19 countries between $1000 and $3000, and ten countries higher than $3000 (figure 4). The three poorest countries are Mozambique ($694), Ethiopia ($748) and Malawi ($767). The three richest countries are Equatorial Guinea ($26534), Seychelles ($18654) and Mauritius ($12866), followed by Botswana ($11209), South Africa ($10392) and Namibia ($7130).

Figure 4 GDP per capita PPP in Sub-Saharan Africa (average 2000-2010), grouped

GDP per capita PPP in Sub-Saharan Africa (average 2000-2010) 0 2 4 6 8 10 12 14 <1000 1000-2000 2000-3000 3000-8000 >8000 US $ /year fr e q u e n c y

Sub-Saharan Africa as a whole received $27 on average (SD=22.5) ODA for primary education per year per primary school-aged child in the period 2000-2010 (figure 5).

Seychelles ($89,83), Cape Verde ($73.17) and Equatorial Guinea ($60.33) are three outliers. This can be explained by the fact that they all have small populations. The three outliers are

(30)

followed by Zambia ($41.33), Sao Tomé and Principe ($40.33) and Eritrea ($39.50). In contrast, Nigeria ($1.83), Zimbabwe ($5.83) and South Africa ($7.17) received the least ODA for primary education. In the case of Nigeria and Zimbabwe, this can be explained by its high level of corruption (control of corruption score is 10.55 and 9.94 respectively). In the case of South Africa, it may be explained by the strong GDP growth it experienced in this period. Donors may not see the purpose of giving aid to a country which is considered as an emerging market.

Figure 5 ODA for education received in Sub-Saharan Africa (average 2000-2010), grouped

ODA for education received in Sub-Saharan Africa (average 2000-2010) 0 2 4 6 8 10 12 <10 10~20 20-30 30-50 <50

US $ / prim ary school aged child

fr e q u e n c y 5.2.2. Quality of institutions

The researched Sub-Saharan African countries score 35.7 on average (SD=17.8) on the World Bank Institutions Index in the period 2000 to 2010 (figure 6). Almost half of the countries (15 out of 35) score below 33, which means that they are badly-governed. The other half (14 out of 35) have an II score between 33 and 66, which means that they have average institutions. Only three countries (Mauritius, Botswana and Cape Verde) score above 66 points. Angola (9.6), Zimbabwe (9.75) and Equatorial Guinea (11.53) are the worst-governed countries of Sub-Saharan Africa. These countries score especially low on control of corruption, rule of law and government effectiveness. Equatorial Guinea also scores extremely low on voice and accountability. These findings contradict Sachs’ statement that most parts of Africa are well-governed. While he states that only Angola is poorly-governed and all other countries are

(31)

averagely or well-governed, the findings of this research are different. This also has

consequences for the statement made by Sachs and the institutionalists that only aid directed at well-governed countries is effective. It implies that aid to only a few Sub-Saharan African countries is effective. This makes it even more important to explore how aid targeted at badly-governed countries could be effective.

Figure 6 Quality of institutions in Sub-Saharan Africa (average 2000-2010), grouped

Quality of institutions in Sub-Saharan Africa (average 2000-2010) 0 2 4 6 8 10 12 14 16

poor institutions average institutions good institutions

fr e q u e n c y

5.2.3. Primary education rates in 2010

The average primary school enrolment rate in Sub-Saharan Africa is 83.1% (SD=14.8) in 2010 (figure 7). Out of 35 Sub-Saharan African countries, 12 have enrolment rates above 90% and only three below 60% in 2010. Mauritius (99.8%), Sao Tomé and Principe (98.7%) and Seychelles (98.6%) have the highest enrolment rates, whereas Eritrea (33%), Equatorial Guinea (56.6%) and Niger (57%) have the lowest. The second is very remarkable, since this country has the highest GDP per capita in Sub-Saharan Africa and also receives high amounts of ODA for primary education. The low enrolment rate may be explained by the II score of this country (11.53). Especially the average II scores on voice and accountability (4),

government effectiveness (2.42), regulatory quality (6.30) and control of corruption (1.69) in the period 2000-2010 are extremely low.

The average primary school completion rate in Sub-Saharan Africa is 73.4% (SD=19.9) in 2010 (figure 7). Nine countries have completion rates below 60%, while only six score higher

(32)

than 90%. The lowest completion rates are present in Eritrea (39%), Niger (40.9%) and Angola (46.8%), whereas Cape Verde (100%), Botswana (97%) and Ghana (92%) have the highest. Three countries (Mauritius, Seychelles and Zambia) have primary school completion rates above 100%, which means that some children are over-aged when they complete

primary school. This is due to the repetition of one or more courses before graduating. The average youth literacy rate in Sub-Saharan Africa is 76.5% (SD=19.11) in 2010 (figure 7). This percentage is lower than 60% in 5 countries, and higher than 90% in 11 countries. Niger (24%), Burkina Faso (40%) and Ethiopia (44%) have the lowest youth literacy rates, whereas Seychelles (99%), South Africa (98.6%) and Cape Verde (98%) have the highest.

Figure 7 Primary school education rates in Sub-Saharan Africa (2010), grouped

Primary school education rates in Sub-Saharan Africa (2010)

0 2 4 6 8 10 12 14 <60 60-70 70-80 80-90 <90 % of relevant age fr e q u e n c y Enrolment Com pletion Youth literacy

Overall, the data indicate that in most relatively peaceful Sub-Saharan African countries, primary education is still not universal. Large numbers of children do not complete primary school and cannot read and write. In many countries, less than 60% of the children makes it to the last grade of primary education. Still, the vast majority of the countries have enrolment rates above 90% and more than 80% of the youth between 15 and 24 can read and write in half of the countries. Completing primary school is the biggest challenge in Sub-Saharan Africa, since almost half of the countries have completion rates below 70%.

(33)

5.2.4. Progress on education

Sub-Saharan African countries made 12.3% progress on average (SD=15.3) on primary school enrolment rates in the period 2000-2010 (figure 8). Nevertheless, 9 countries out of 33 countries (data for 2 countries were unavailable) decreased. Equatorial Guinea (-11.75%), South Africa (-9.39) and Eritrea (-6%) performed the worst. This rate stagnated in one country: Mauritius. The countries that performed best on enrolment are Ethiopia (+40.60%), Angola (+40.52%) and Tanzania (+38.58%).

The primary school completion rates increased 17.3% on average (SD=17.3) in the 34 Sub-Saharan African countries with available data (figure 8). This rate deteriorated in 6 countries. Namibia (-9.94%), Zimbabwe (-8%) and Cape Verde (-5.43%) experienced the biggest drop. It should be noted that the drop in Cape Verde should be interpreted as progress, because this country went from 105.43% in 2000 to 100% in 2010. Ethiopia (+49.53%), Mozambique (+44.2%) and Tanzania (+39.94%) made the most progress on enrolment rates. Namibia and Zimbabwe both had high primary school completion rates in 2000 (91.2% and 97% respectively), which means that they still had relatively high primary school completion rates in 2010 (81.2% and 89% respectively).

Based on 34 countries with data, Sub-Saharan Africa made 4.2% progress on average (SD=4.2) regarding youth literacy rates (figure 8). Eight countries experienced a drop in youth literacy rates, whereas this rate stagnated in Sao Tomé and Principe. The countries that performed the worst are Lesotho (-7.93%), Madagascar (-6.24%) and Ethiopia (-6%). In contrast, Mali (+20.30%), Mozambique (+19%) and Ghana (+15.05%) made most progress on youth literacy rates. It is remarkable that the youth literacy rates dropped in Ethiopia, while this country has made the most progress on primary school enrolment and completion rates. Supposedly, its education system has not been able to cope with the large numbers of new students, which affects the quality of education. Still, the increase in the primary completion rates indicates that children do finish school and thus should be able to read and write at age 15 or older. A possible explanation for this is that the enrolled children make it to the last grade of primary school without passing the required exams.

Overall, Mozambique (+32.27%), Ethiopia (+28.04%) and Tanzania (+25.69%) made most progress on primary education rates. It should be noted that although these countries have not received excessive amounts of ODA for education ($32.17, $8.33 and $16.17 per primary school-aged child on average), they are known as ‘donor darlings’. This means that they are the three largest recipients of foreign aid in Sub-Saharan Africa (Policy Management Brief, 2006, p.8). Nevertheless, the fact that they receive large amounts of aid might be

(34)

related to the (changed) quality of institutions in these countries. For example, the political stability in Mozambique went up (from 38.9 in 2000 to 57.1 in 2010) and Ethiopia’s government effectiveness increased significantly (from 15.6 in 2000 to 42.1 in 2010). In Tanzania, control of corruption rose as well (from 14.2 in 2000 to 34.8 in 2010) but government effectiveness decreased (from 33.5 in 2000 to 39.7 in 2010). The case study elaborates on this country and gives an in-depth explanation of its progress on education and the way it is related to governance.

Namibia (-3.16%), Cape Verde (-3.05%) and Zimbabwe (-2.17%) made the least progress: the primary education rates in these countries even dropped slightly. It should be noted that the education rates in these countries are still relatively high (most above 90%), which means that a slight drop is less problematic.

Figure 8 Progress on primary education rates in Sub-Saharan Africa (2000-2010), grouped

Progress on primary education rates in Sub-Saharan Africa (2000-2010) 0 2 4 6 8 10 12 14 16 18 decline/no progress 1-10% 10-20% 20-30% <30% fr e q u e n c y Enrolment Completion Youth literacy 5.2.5. Conclusion

In sum, most relatively peaceful countries in Sub-Saharan Africa are averagely or badly-governed, whereas only three countries are well-governed. Moreover, in most countries, primary education is still not universal. Mozambique, Ethiopia and Tanzania made most progress on primary education rates, whereas Namibia, Cape Verde and Zimbabwe made no progress.

(35)

5.3. Regression analysis

In this section, the hypotheses are tested using a simple linear regression analysis. Unfortunately, multiple regression analyses could not be executed due to the lack of knowledge of the writer.

5.3.1. Aid allocation

The regression analysis shows that the GDP per capita of a country is not associated with the amount of ODA per primary school-aged child that is allocated to that country (F=0,650; df=32; p=n.s.; R²=0,14). This finding confirms the first hypothesis: it is not the case that the poorest countries receive the most ODA for education. This result is in line with those of Collier and Dollar (2002, p.1475), who found that the allocation of aid is not poverty-efficient.

In contrast, the quality of institutions in a country is associated with the amount of ODA per primary school-aged child that is allocated to that country (F=11,723; df=32; p<0,01; R²=0,27). As figure 9 shows, this correlation is positive: the better governed a country is, the more ODA for education a country receives. The data fit the regression model quite well: 27% of the variation in the amount of ODA for primary education allocated to a country can be explained by the variation in the quality of institutions (II score). This finding is in line with the second hypothesis and with the findings of Collier and Dollar (2002) and Dollar and Levine (2006) that governance is an important factor in the allocation of aid.

Figure 9 The correlation between the quality of institutions (IIscoretotal) and the allocated

(36)

5.3.2. Quantity of aid and progress on education

The analysis also shows that there is no correlation between the amount of ODA for education a country receives and the progress on primary school enrolment rates (F=3,509; df=30; p=n.s.; R²=0,11), primary school completion rates (F=0,77; df=31; p=n.s.; R²=0,003) or youth literacy rates (F=1,14; df=31; p=n.s.; R²=0,04). The third hypothesis can thus not be

confirmed. The analysis has also been executed leaving out badly-governed countries (with II score lower than 33), but led to the same results. The amount of ODA allocated to a country, even when averagely or well-governed, does not predict the progress on education rates.

This means that no support has been found for Sachs’ view on foreign aid: there is no evidence that the more aid for education a country receives, the better the results in that sector. Supposedly, assuming that the higher the amount of received foreign aid, the better the results, is to short-sighted. Some countries may already have relatively high education rates, and are investing in other fields of education which are not reflected by the three education indicators that have been used. It should also be noted that the three countries that receive most foreign aid in total (Mozambique, Ethiopia and Tanzania), also have made the most progress on education rates. This money may be invested in other sectors which benefit education indirectly. For example, if countries invest in infrastructure, children have the ability to travel to school because there is a proper road that takes them there safely. Then, the likelihood that children go to school and/or complete it is much higher.

5.3.3. Governance

The fourth hypothesis tests whether the better the institutions of a country are, the better the results on education. The data indicate that the quality of institutions is not associated with the progress on primary school enrolment rates (F=3,739; df=32; p=n.s.; R²=0,11), primary school completion rates (F=2,513; df=31; p=n.s.; R²=0,077) or youth literacy rates (F=0,030; df=31; p=n.s.; R²=0,001). This is also the case for every separate indicator on the Institutions Index. The assumption that the better the institutions of a country are, the better the results on education, is not supported. This finding might be explained by the fact that some countries with relatively good institutions already had high education rates. For these countries, making less progress is not necessarily negative.

However, a correlation has been found between the quality of institutions and primary school enrolment and completion rates in the same year. Thus, the II score in 2000 is

associated with the primary school enrolment in 2000 (F=10,045; df=30; p<0.01;

(37)

p<0.01; R²=0,25)(figure 11). Likewise, the II score in 2010 is associated with the primary school enrolment in 2010 (F=6,2; df=34; p<0.05; R²=0,158)(figure 10) and with the primary school completion rate in 2010 (F=10,764; df=32; p<0.01; R²=0,258)(figure 11). The

regression model fits best with the data on quality of institutions and the enrolment rates in 2000 and completion rates in both years. More specifically, 25.7% (2000) and 15.8% (2010) of the variation in primary school enrolment rates can be explained by the variation in the II scores. Moreover, 25% (2000) and 25.8% (2010) of the variation in primary school

completion rates can be explained by the variation in the II scores. In other words, the better the quality of institutions of a country, the higher the primary school enrolment and

completion rates. This finding is in line with the institutionalist approach that governance is related to the public services that are provided to citizens (Acemoglu & Robinson, 2012).

Figure 10 The correlation between quality of institutions (IIscore2000 and IIscore2010) and

primary school enrolment rates in 2000 and 2010 (enrolment2000 and enrolment2010)

Figure 11 The correlation between quality of institutions (IIscore2000 and IIscore2010) and