What should we learn from our history of development aid? Development aid faces fundamental problems, especially in fragile states. These problems were addressed regularly by various representatives, academics, politicians and practitioners. The Samaritan Dilemma and a case study on used clothing is presented. It is clear that only having good intentions is not enough to help out the poor. The econometric analysis on aid effectiveness has repeatedly offered hope and repeatedly disappointed. There are several reasons why the aid flow increases anyway. Chauvet and Collier (2004) use a different approach and show potential for efficiency enhancements, but their analysis suffers from several caveats. Nevertheless, potential is shown and the authors indicate how to address the problem. Furthermore, the idea of social businesses is introduced as a market based mechanism which allows for feedback and accountability.
Recent initiatives, primarily pushed by private actors, give reason for hope. It is on the governments to provide the basic conditions in the fragile states and to back those initiatives which so far show the best results.
Agenda
Abbreviations
1. Introduction: Despite good intentions - criticism on development aid
2. Fragility and poverty efficiency
3. Development and Aid
3.1 The Samaritan Dilemma
3.2 Used-clothing and production - a case study
4. Econometric problems in the literature of aid effectiveness
5. The potential of efficiency improvements in fragile states
5.1 Promoting turnarounds
5.2 Sustaining turnarounds
6. Recent proposals for more effective aid in fragile states
7. Conclusion
References
Appendix
Abbreviations
illustration not visible in this excerpt
1. Introduction: Despite good intentions - criticism on development aid
According to the Organization for Economic Co-operation and Development (OECD) the total official development assistance (ODA) was in 2008 at its highest level ever (OECD, 2009). The ODA rose in real terms by 10.2% to USD 119.8 billion, representing 0.3% of the OECD's Development Assistance Committee combined gross national income. Adding up, the West already spent USD 2.3 trillion on foreign aid over the last five decades (Easterly, 2006). With respect to these numbers, expectations for improvement in the recipient countries are expected to be high. However, the effectiveness of development aid is a very controversial issue. And it becomes even more controversial with respect to so-called fragile states. A state is classified as fragile if "its country policy and institutional assessment (CPIA) score falls below a particular threshold" (Baliamoune-Lutz and McGillivray, 2008) Effectiveness basically means doing the right things in order to obtain the optimal result from an economic perspective. Hence, a good input-ouput ratio is required. A characteristic of an effective aid program is that at the margin, the expected contribution of aid to poverty reduction should be the same across all components of the program (Chauvet and Collier, 2004).
However, regarding the high expectations, the mere evidence seems to speak a different language. A World Bank evaluation report states that "[ d]espite the billions of dollars spent on development assistance each year, there is still very little known about the actual impact of projects on the poor" (Baker 2000, p. iv). Easterly (2006) goes further by highlighting that USD 568 billion spent on aid to Africa stand vis-à-vis to the typical African country which is today no richer than 40 years ago1. An even diverging trend between ODA and economic growth is illustrated in the appendix 1. Certainly, Easterly's approach here is relatively one-sided, neglecting the scenario if no aid was delivered. E.g. Collier (2007) argues that development aid led to additional economic growth of 1% among the bottom billion. Since the growth rate was predominantly at zero per cent, development aid actually "made the difference between stagnation and severe cumulative decline" (Collier 2007, p. 100). According to the World Bank Report 2009 the current year does not give a lot of hope for improvement but stresses the point that the poverty count is likely to rise, with the more fragile economies being especially at risk. In the past, several authors in the popular science literature2 firmly criticized development aid, ranging from Erler (1985) in a purely descriptive way, over Novogratz (2009), giving a variety of examples when good intentions in fact can do a lot of harm, to Moyo (2009), demanding a stop of development aid for Africa and rather giving money to Kiva3. Having caste d such a damning light on development aid, a response was much expected. From an institutional standpoint, the OECD has dedicated itself with the Paris Declaration4 and the Accra Agenda for Action5 to a higher degree of aid effectiveness (OECD), although, according to the World Bank report 2009 there is still a lack of progress towards theses targets. With respect to an academic approach, Dreher and Kilby (2009) literally revisit the impact of aid on growth and come to the conclusion that needed aid has a significant positive effect. Besides, Collier (2007) stresses the point that it was time for academics to set aside their tendency to write only for each other but to address the public on this issue6.
This paper examines the most important contributions with respect to the questionable effect of development aid effectiveness and a special focus on fragile states. Moreover, a few proposals aiming to improve aid efficiency are presented. In this sense, it is the aim of this paper to show the need and the potential of more effective work in fragile, as well as to outline schemes for improvement of the efficiency in development aid.
The rest of the paper is organized as follows: Section 2 explains the basics of fragility and aid efficiency. In section 3 the fundamental concept of development and aid is presented and two examples are given: a game theoretical approach and a case study of used-clothing. Section 4 continues by further going into detail of the effects of development aid and basically giving an overview on profound literature addressing the issue in question and pointing to severe econometric problems. Section 5 thoroughly analyses the paper by Chauvet and Collier (2004) Development Effectiveness in Fragile States: Spillovers and Turnarounds. The paper deals with a quantitative approach on the potential of development aid lying in fragile states. Section 6 briefly presents a couple of development aid solutions claiming to be more effective, above all the idea of microfinance will be addressed, while section 7 concludes.
2. Fragility and poverty efficiency
The World Bank uses a CPIA score of 3.2 or less to define a fragile state (World Bank, 2009a). For an overview of CPIA scores from 2008 see the appendix 2. But note that countries with a CPIA score below 3.2 may not exhibit fragility and there may be some aspects of fragility in countries with CPIA scores above 3.26. Fragile states account for more than one billion people, representing a sixth of the world's population. Collier (2007) calls the bottom billion "the world's biggest economic problem". Roughly half of all the worlds' infant deaths, and a third of all people surviving on less than one dollar a day live in states which "cannot or will not deliver core functions to the majority of their people, including the poor" (DFID, 2005). There is some consensus that weak institutions are the main cause for fragile states. Brinkerhoff (2007) and Vallings and Moreno-Torres (2005) summarize the basic reasons for fragility as follows:
- armed conflict (currently or in the past),
- poor governance and political instability,
- high degree of ethnic and social differences,
- economic decline,
- demographic and environmental stress,
- low levels of human development,
- fragile neighbors (see section 5).
Reynal-Querol (2009) emphasizes the point that these factors have a circulate nature. This is seen as the basis of the fragility trap concept meaning that the drivers of fragility are often self and mutually reinforcing. At this point development effectiveness comes into play. Since in case this vicious circle can be broken, aid can take away a huge burden and thus the massive costs, coming along with the state of fragility, can be avoided (see section 5).
Dollar and Collier derive a poverty efficient allocation, reconciling need and effectiveness in a way that aims for the goal to lift as many people out of poverty as possible. Poverty efficiency points to the limited resources of donors and the urgency of fragile states. Consequently it is the donor's responsibility to allocate development aid as effectively as possible. Dollar and Collier compare their results with the actual allocation and come to the conclusion that the actual allocation is far from being poverty-efficient. Putting it into numbers, 74% of the world's poor live in a respective country but the share of aid amounts only to 56%. In fact, the biggest deviation was that a lot of aid goes to middle income countries since they are more interesting from a commercial and political viewpoint (Collier 2007). According to a paper by Levin and Dollar (2005) fragile countries receive 43% less aid than it would be appropriate, taking into consideration their poverty rates. A further huge disadvantage is the high volatility level of aid with respect to fragile countries. Dollar and Collier (2002) calculate that according to their sample of countries the present allocation of aid lifts around 10 million people annually out of poverty. If the allocations were made more efficiently, productivity of aid would nearly double. Their results go hand in hand with Ban Ki-moon's initial quote. From the donor's perspective this allocation is understandable. Investments in fragile states are at very high risk and often pay back at a very late stage8. Besides, donors often are in need of visual results. The most famous benchmark at the time is the set of Millennium Development Goals (MDGs). But with respect to reaching these, fragile states lack far behind (see appendix 3). In each category (except for the access to sanitation, which is basically a sub goal), fragile states go last, implying that so far they made the least progress towards the goals9. But having in mind the definition of effectiveness, this situation bears a huge potential for higher poverty efficiency.
Before trying to calculate this potential (section 5), the paper will first have a closer look at the theoretical background of development and aid, as well as stress the point of the difficulties arising in the econometric analysis of aid effectiveness thereinafter.
3. Development and Aid
Raising the question why aid should help in theory usually brings up the following points. Very generally speaking, aid ideally finances investments, new jobs are created, and people gain access to credit. Besides, the workers' pro ductivity10 is raised and new technologies and knowhow are transferred. The opposite perspective is that aid actually encourages corruption and undermines incentives to produce or save with regard to both the public and the private sector. Hence, a culture of dependency is created. Local markets are distorted (e.g. food aid can lower farmer production). Lastly, it might lead to a devaluation of the currency which weakens the profitability of tradable goo ds11.
In order to get a good grasp of the strategically problematic situation the donors are facing, now a game theoretical approach, called the Samaritan dilemma, is presented.
3.1 The Samaritan Dilemma
The essence of the Samaritan Dilemma is that the existence of assistance may induce a demand for assistance. The mere expectation of charity can lead people to behave in ways that made them become recipients of charity. Applying the dilemma to development aid in a game theoretic context as Buchanan already did in 1972, consider a first player as the potential donor and a second player as the potential recipient country. Now the donor has got two strategies, either he helps or he does not help. The second player, instead, has the options to work or not to work. Buchanan distinguished between two sorts of games, the active and the passive Samaritan Dilemma. Both players have imperfect information, meaning it is a simultaneous game. The possible outcomes, representing ordinary utility indicators, are presented in 2x2 matrices.
illustration not visible in this excerpt
One can easily see that in the active Samaritan Dilemma the Nash-equilibrium is not to work and that help is granted anyway. It is the donor's dominant strategy to help. One might think that such an unconditional samaritan is fairly unrealistic, since he is determined to help no matter whether the recipient requires help or not. But the first scenario gains credibility considering the case of a fragile country, having very urgent basic needs. In case the recipient does not work the donor's utility decreases from 4 to 3. This indicates the donor's strong preference that the recipient takes up the incentive. Nevertheless, he ends up in a personal dilemma.
Looking at the passive Samaritan Dilemma12, now the donor prefers not to help but would still help when the recipient is in need. It turns out that there are now two Nash-equilibria. The first one is the same as in the previous case, whereas in the second equilibrium there is no aid flow but the recipient works anyway. The second solution only becomes possible, because the donor can credibly commit himself not to help. But as mentioned above, to put in practice such an attitude is, especially in the case of a fragile state extremely hard.
Considering the two dilemmas, one could ask himself whether it was better to delegate the responsibility for allocating aid budget to a less inequality-averse agent. But also the con ditionality argument comes into play here. According to the dilemma, institutions do not have the credibility of moving out in case conditions are not met. Further issues on that are considered in detail in the macro perspective part, in section 4.
Novogratz (2009) states that "sometimes our good intentions have unintended consequences". Following, at typical example of such is outlined.
3.2 Used-clothing and production — a case study
As already pointed out Africa is often seen as a story of failure, by the way of contrast, East Asia, once incorporating many fragile states, can be perceived as the international success story over the last 30 years in terms of economic growth, trade and human development13. Countries like Taiwan, Korea, Singapore, Hong Kong, and now China have something in common. In the beginning they started up with apparel production. This was followed by huge exports of textiles. And only at the last stage they went to increasingly sophisticated electronic and industrial goods as their economies had grown (Frazer, 2008). Hence, realizing that the start-up might lie in the apparel production, all measures taken by donors in fragile states, must not hinder the development of these local apparel market. Thus, it is at the heart of a case study conducted by Frazer (2008) whether there is a causal link between used-clothing imports and apparel production.
Approximately 16% of the containers with US exports bound for Africa in 1995 were filled with used-clothing. First of all, one has to say that just as food aid clearly benefits the consumers of food, used-clothing imports make the African consumers better-off by merely making available lower cost apparel. But more important is the impact on African producers. Frazer (2008) estimates the following OLS-regression:
illustration not visible in this excerpt
Since exporters of used-clothing tend to export into countries where apparel production failed, an en dogeneity problem is arising. Namely, U it might be correlated with sp it. Using an instrumental variable approach, this problem can be fixed. The idea is to find a variable which is highly correlated with U it, but at the same time uncorrelate d with the residual sp it. Frazer (2008) uses the geographical-determined fraction of total exports going to each African country as an instrument and thus follows a technique developed by Frankel and Romer (1999). As for the instrument, for example, Germany exports more to Ghana than to Mozambique, simply due to geographic reasons. So when the supply of used-clothing exports increases in Germany, more of this supply will be predicted to go to Ghana than to Mozambique. Implicitly, it is also assumed that the people donating used-clothing have until recently been largely unaware that the bulk of this clothing is exported and so were certainly not donating with conditions in Africa in their mind. The result is that over the period 1981 to 2000, used-clothing imports had a significant negative impact on the apparel production sectors in sub-Saharan African countries, being responsible for roughly 39% of the annual decline in apparel production, and roughly half of the annual decline in apparel employment.
From a micro perspective two examples of development aid were shown, both of them having originally good intentions. At first, the example of the Samaritan, who wishes to donate money, showed, that by doing so, he in fact prevents the poor from working. Secondly, there was the provision of low-cost cloth which turns out to have highly destructive effects on local apparel production. The focus is now turned to the macro perspective, highlighting the econometric problems of model selection and robustness in the most influential literature of the past.
4. Econometric problems in the literature of aid effectiveness
In the introductory part a few works having opposite opinions were already mentioned. The basic problem of the econometric analysis of development aid is a problem of causality. On the one hand, there is a downward bias. Imagining that aid systematically flows into countries that are highly fragile (e.g. due to war or natural disasters) which generate low or negative growth, this will in turn tend to generate a negative association for aid and growth. On the other hand, aid could systematically go to 'reward' countries, having done things well in the past. If this former growth persists, there is no foundation for a positive association since it lacks causation. Due to this difficulty, Bar der (2009) calls the analysis pointed to in the introductory part by Moyo and Easterly "popular folk regressions", implying that they were too simplistic14. However, even in the unilateral academic world there is not less of a controversial debate. While very early papers found a positive correlation between aid and growth (e.g. Levy, 1988), in 1994 Boone claimed that aid and growth are not correlated. So academics turned to the view that aid might have an effect on growth, but only under certain circumstances and the idea of conditional aid was born. The most influential paper related to this issue was written by Burnside and Dollar in 2000. Their regression approach is as follows:
illustration not visible in this excerpt
[...]
1 As a counter- example, the German former development minister Wieczorek-Zeul (2004) puts development aid in a different perspective by pointing out the example that in 2004 the rising oil prices led to a burden of USD 60 billion for developing countries (although they don't account for this rise in oil prices), approximately amounting this year's total ODA.
2 Popular science literature aimed at a mass market, mostly written by practitioners, is deliberately included in this paper in order to get a broader picture, from a practical and a theoretical point of view and lastly due to its importance in the public discussion. Although the output is not subject to the same level of technical control and rigor as academic publications, it is seen as useful complement.
3 Kiva is a platform connecting microfinance institutions (MFI) to social investors through the internet. Find more information at www.kiva.org. The idea of microfinance is presented in section 6.
4 " The Paris Declaration, endorsed on 2 March 2005, is an international agreement to [...] to increase efforts in harmonization, alignment and managing aid for results with a set of monitorable actions and indicators." (OECD)
5 The Accra Agenda for Action from 2008 led to the Third High Level Forum on Aid Effectiveness in Accra (OECD).
6 Collier's published book The bottom billion: Why the Poorest Countries are Failing and What Can be Done About It (2007) is exactly such an approach .
7 There is no common definition of a fragile state. E.g. the OECD sees fragile states are "unable to provide physical security, legitimate political institutions, sound economic management and social services for the benefit of its population" (OECD, 2006).
8 "Fragile states are the hardest countries in the world to help develop" (DFID 2005, p.32).
9 With respect to their situation and the facts pointed out above the result seems little surprising.
10 See the issue of the efficient wage hypothesis, meaning that undernourished individuals are less productive and then less capable of generating income.
11 For further information on this issue, see the literature to the Dutch Disease (e.g. Sachs, 2001)
12 The Passive Samaritan Dilemma is set up by simply transposing the donor's utility between the first and the third quadrant.
13 E.g. East Asia's poverty rate came down from 27.58% to 15.32% in the time from 1990-1998. Thus, the region has come close to the MDG of halving the proportion in poverty (17 years ahead of schedule). By now they already exceeded its target, largely attributable to China (World Bank 2009).
14 This probably applies rather to Moyo than to Easterly.
- Quote paper
- Sebastian Groh (Author), 2009, Effectiveness of Development Aid in Fragile States, Munich, GRIN Verlag, https://www.grin.com/document/144916
-
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X. -
Upload your own papers! Earn money and win an iPhone X.