! !"#$%&'()%*( "+,-".$/( ( A comparative analysis of Sub Sah aran Africa, South Asia, Southe ast Asia Alexis M. Rossetti [Senior Honors Thesis] Advised by Dr. Benjamin Smith Spring 2018
! Abstract This thesis examines the outcomes of foreign aid in South and South e ast Asia during the 1960's, and in Sub Saharan Africa in the 1990's. The project seeks to determine what caused the divergence in outcomes of the two regions given that they were both a focus of the international aid community for a similar period of time. I fi nd that aid to South and Southe ast Asia was better explain ed by the common i ndicators of aid effectiveness than in Sub Saharan Africa. This suggests that the way aid institutions and policy makers evaluate aid does not fit the Sub Saharan model. In the single case study of Nigeria, I found that aid had no significant effect on growth during the time period of the study, and historically has had a negative effe ct by prolonging the civil war. This project is a first step in closing the growing paradox of aid research, and its findings support the claim that foreign aid is not having the desired effect in Sub Saharan Africa.
! # Introduction This thesis is designed to b egin to asses s a growing gap between foreign aid research that results in positive and negative aid result s It set out to answer the question of why aid outcomes in South and Southeast Asia appear to be positive while aid outcomes in Sub Saharan Africa seem to be negative. Outside of the quantitative findings, I also consider possible explanations for the divergence in the findin gs of aid researchers by includi ng considerations of the methodol o g ies of past research. This project attemp t s to begin to close this gap by including the classic indicators from both macroeconomic and microeconomic studies of aid. The paper proceeds with an introduction to a id history, aid research, development outlines for the regions in the study, the results of the study, and concludes with a discussion about directions for future research This project use s a quantitative comparative study of the aid era s 1960 1970 in So uth and Southe ast Asia, and 1990 2000 in Sub Saharan Africa in order to investigate the divergences that may explain the current differences in Sub Saharan Africa economies and those seen in South and Southe ast Asia 1 While the purpose is not to assume that aid was the causal mechanism in the success of South and Southe ast Asia, but the fact r emains that the outcomes of these regions who were the central focus es of international aid policy have been different. Although South an d Southe ast Asia have not yet reached full development, on aggregate they are improving, and some individual states in these regions have significant positive growth rates. Although growth !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! 1 I chose these regions because using USAID Foreign Aid Explorer, I found that both were the regional focus of US foreign aid and aid policies during the respective time periods of this study
! $ is slowly improving in Sub Saharan Africa, it remains the poorest region in the world. 2 This could indicate that aid in Sub Saharan Africa may have a malignant effect. 3 Sub Saharan Africa as a region has high net inflows of offic ial aid and yet their economies do not show signs of following the same pattern of success as HPAEs 4 The vast amounts of aid now being received by Sub Saharan Africa have not generally resulted in i ncreased development overall. I set out to determine what it is about aid to Africa that has seemingly produced such different results as seen in the other regions I consider internal, external, and historic al indicator s as possible causes for the divergence in outcomes. The frame of this project is designed to suggest initial points of diverge nce between the aid eras and their consequent outcomes in their respective regions. In order to create a comprehensive comparative study, I will dedicate a large amount of space to contextualizing each region in the respective time period bein g studied. Taking notes from Capitalist Development and Democracy ( Rueschemeyer, Stephens, and Stephens 1992 ) part of the methodology for this project is comparative historical analysis This method originally was used to approach cross national case studies over long periods of time in order to give individual factors their due in terms of the context th at t hey were being analyzed in. In this study I use a scaled back ver s ion of this method to analyze an individual case in order both !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! 2 Moyo, Dambisa. Dead aid: Why aid is not working and how ther e is a better way for Africa Macmillan, 2009. 3 Moyo, Dambisa. 4 Booth, Anne. "Colonial revenue policies and the impact of the transition to independence in South East Asia." Bijdragen tot de taal land en volkenkunde/Journal of the Humanities and Social Sciences of Southeast Asia 169, no. 1 (2013): 37 67. HPAE's: defined in the paper as high performing Asian economies
! % contextualize the quant itative findings of that case and also test the findings against the findings of the large N study. The quantitative portions of the paper draw from aid data collected from the World Bank and Aid Data sources I used two different sources in order to build the most complete data set possible. World Bank data begins in 1960, however that data alone had a significant number of gaps. This was especially problematic in the Asia cases. The Aid Data datasets begin a s early as 1950, and this enabled me to fill in some of the gaps. Having the most complete dataset possible was crucial to the project. There is an obvious disparity in number of cases between the regions, and so it was critical t hat every case possible be included. Although combing datasets was not the most ideal option, due to the risk of indicators being evaluated differently between the sets, I was careful t o use indicators that were measured in the s ame units to avoid these pot ential risks. There i s both an aggregate regional comparison as well as a focused study on aid in Nigeria. I use STATA software to establish initial correlations between developm ent indicators across economic and health sectors. Then I utilize multiple regression analysis to study the effects of multiple independent variables on the outcome of the central independent variable: growth in GDP. I use the method of comparative historical analysis in addition to the statistical findings to add constructive context to the quantitative analysis. This process helps to explain the potential gaps in the data. A Brief Introduction to Foreign Aid Foreign aid is the assistance given from one country to another, most commonly in the form of cap ital, but also can include food aid and technical assistance. Aid policy since World
! & War I is broken up into seven periods by Dambisa Moyo in Dead Aid. While there are some instances of foreign aid occurring before thi s time period, for the scope of this paper they are not relevant. The periods I will focus on include the Bretton Woods era, the Marshall Plan, industrialization, poverty eradication, stabilization, defense of demo cracy, and democracy building In the aftermath of the Great Depress ion, the Bretton Woods Conference of 1944 was organized in order to discuss and formulate a restructuring of the international system. The Bretton Woods system fostered the first large scale transfer of aid from government to government. 5 Out of this conference also came the biggest institutions involved in foreign aid today, the World Bank and the International Monetary Fund (further known as the IMF). These institutions were designed to be complements to one another. The World Bank was to manage capital investments in the regions of reconstruction, and the IMF was to manage global finance. However as noted by Moyo, both of these instit utions would evolve to have central focuses in development, even though their original purpose was for re construction. In the post World War II era, the Marshall Plan was c reate d out of the structure of the Bretton Woods system. The Marshall Plan included an approximate thirteen billion US dollar aid package disbursed over a five year period to Great Britain, France, Italy, and Germany. 6 Reconstruction in Western Europe is often cited as an example of aid success. 7 The success of the Marshall Plan is often anec dot al ly supported by the present success of those Western European powers. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! 5 Moyo 2009 6 Moyo 2009 De Long, J. Bradford, and Barry Eichengreen. The Marshall Plan: History's most successful structural adjustment program No. w3899. National Bureau of Economic Research, 1991.
! There is a body of dissenting opinions related to the success of the Marshall Plan that should be taken into consideration. Nowak points out that, "among economists, there is no clear o pinion of its [the Marshall Plan] impact on the output of growth in the recipient countries." 8 Despite the lack of clearly demonstrated causality, "the plan was quickly perceived as the example of development aid effectiveness and became a justification of foreign assistance for decades...becom[ing] a model of development in the underdeveloped countries." 9 The perceived success of the Marshall Plan as a model for development policy takes for granted the fact that the Marshall Plan was a reconstruction pa ckage and not a means of development Development policy usually is focused on building up countries that had not industrialized in the first place. The fact that development policy built on this foundation had such a large assumption at its core is indicative of future problems the aid industry would encounter. In the context of Asia, the Marshall Plan was branded under a program called The Colombo Plan for Cooperative Economic Development in South and Southeast Asia' in 1950. 10 The plan polit ically excluded countries based in part on the goal to build up developing countries towards democracy in hopes of staving off the spread of communism. 11 Under this plan, donor funding was focused on industrial projects. The 1960's brought in an era of in frastructure development focused aid; essentially the long game version of the 50's Marshall Plan. The general thought of the time in terms of aid to Africa was that private sectors wouldn't fund such endeavors as railways, and so aid could fill in these g aps for longer term payoffs. 12 With rising numbers of countries participating in aid !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! 8 Nowak, Wioletta. "The evolution of development assistance." Journal of US China Public Administration 11, no. 5 (2014): 454 462. ( Nowak 2014, 455 10 Nowak 2014 11 Nowak 2014 12 Moyo 2009
! ) giving, the Development Assistance Committee (DAC) was formed in 1960. 13 Overall, the aid industry was growing into its own world. Established institutions such as the IM F and World Bank, state led groups such as USAID, multilateral groups such as the DAC, have now been gaining funding and with that funding, influence. The aid industry of the 1970's turned away from industrialization and towards poverty er adication which has remained a common thread throughout the remaining decades. The World Bank, under the leadership of Robert McNamara, redirected funds away from the infrastructure investment focus and towards alleviating poverty. 14 During this decade, food aid ma de up for approximately 20% of official development a s sistance (further known as ODA) 15 The basic negative effect of food aid is that i t is easy to steal, and also risks harm ing the local economy more than it helps. Essentially an influx of foreign goods in an under developed economy crushes whatever viable producers are present in the economy. Especially in relation to food aid, because most under developing countries have agrarian based economies. 16 These negative consequences will be reviewed in depth in the Nigeria spotlight case, but for now it is important to note the food aid is often a large proportion of aid packages. The 1979 oil crisis created a domino effect of struggling economies across the wo rld, and the resulting economic collapse led to about a decade of recovering debt payments instead of development. 17 As the price of oil sky rocketed, the under developed countries were hit twice, first by high oil prices, and second when their adjustable interest rates increased exponentially. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! 13 Nowak 2014 14 Moyo 2009 *% Nowak 2014, 458 16 Nunn, Nathan, and Nancy Qian. "Aiding Conflict: The Impact of U.S. Food Aid on Civil War." NBER, www.nber.org/papers/w17794 17 Nowak 2014
! ( However, similar to the big bank collapses in the U S 2008 financial crisis, the world could not afford to let these countries default on their loans. "If emerging nations were allowed to default unchecked, this would have led to a complete collapse of the international financial structure." 18 Conveniently, for the donors, this era saw a "begin[ning] to support the market approach to development." 19 The economic adjustments were termed "Structural Adjustment Pr ograms". The effects of these adjustments have provided mixed results. "Aiming at resuming economic growth and promoting export oriented development, [Structural Adjustment Programs] were implemented in almost all African countries" 20 However, Maswana notes that these programs have been disappointing at achieving these goals and that Africa's share of exports has actually declined. Meanwhile, it is increasingly apparent that orthodox adjustment policies applied on a massive scale in Africa in the 198 0's are not contributing to [these] long term objectives." 21 The long term objectives referenced here were the from the Consensus of Long Term Development Objectives th at were established by the Lagos Plan of Action in 1980. A central debate within the aid industry itself is whether or not aid is best given in a conce ssional format or not. Currentl y concessional modes are the more popular course of action, however as mentioned above and as will be reviewed again below, the enforcement of government poli cy change in order to receive aid has not proven effective in implementation of those policies, and further is argued to be detrimental to the strength of the state itself. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! 18 Moyo 2009, 18 19 Nowak 2014, 459 20 Maswana, Jean Claude. "Economic Development Patterns and Outcomes in Africa and Asia." (2006). "* "Overview of an Alternative Long Term Development Strategy." Africa's Recovery in the 1990s: From Stagnation and Adjustment to Human Develop ment by Giovanni Andrea Cornia et al., St. Martin's Press, 1995, pp. 159 190.
! *+ In the Sub Saharan African context, it was the 199 0's that the aid community focu sed on governance (or lack thereof) as the main factor t hat was impeding Africa's development. The new decade brought a new appr oach fro m aid donors, "[m]uch of the official flow of aid was on a concessional basis" they began to invest in and attempt to engender democracy through aid giving via loans with a string of conditions attached. 22 The institutions of the IMF and the World Bank supported this shift by creating new programs, "[t]he main IMF programmes were the Structural Adjustment Facility (1986) and the Enhanced Structural Adjustment Facility (1987)" 23 The international organizations were also simultaneously continuing to develop their definitions of rights, and specifically what role these rights played in developing countries. In 1948, the United Nations General Assembly adopted the Universal Declaration of Human Rights. This document broadened previous understandings and definitions of human rights to include the inherent dignity of all members of the human family'. 24 The 1997 Maast r icht Guidelines on Violations of Rights established a responsibility on account of the state to provide paths to development. "The international human rights law assumes that international cooperation will promote human rights, especially the economic, social, and cultural rights including the right to development." 25 The idea of development as a human right establishes precedence for international actors such as the UN to intervene in cases where states are failing to provide such opportunities. However, the balance of power is important to maintain. Barbara Kalima, the coordinator for AFRODAD, notes that any practice whos e result is a limiting of a !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! "" Nowak 2014, 460 "# Nowak 2004, 460 24 Kalima Barbara. Reality of Aid: Africa Edition 2003 04 AFRODAD, 2004. 25 Kalima 2004, 14
! ** perceived state role, will cause tension in both national and inter national relations 26 This is counter productive to development goals because "[t]hese tensions undermine sustainable development, peace and security, which are the basis for good governance." 27 In the effort to improve governance through the implementing of new aid programs, the preference f or free market tenets, thought to be ensured by democratic processes, became pronounced. Institutions practiced this by rewarding countries for policy reforms modeled off of western models, specifically that, "[r]ecipi ent countries had to implement market reforms based on the assumptions of the Washington Consensus." 28 The negative results of aid conditionality are two fold. Collier points out that aid conditionality has two major issu es, psychological and economic He explains that conditional loans not only increased the recipient's unwillingness to make reforms that would have been beneficial, but also these reforms damaged the accountability of the national governments. The attempt of the aid community to influe nce the policy of the nations they are trying to help has largely led to animosity on the account of the recipients and a failure to implement those policies in any meaningful way. Literature Review This literature review will work to establish the maj or theories and methods of aid research to date. The findings of the major works of aid effectiveness generally fall into three categories. The first group finds aid to have overall positive outcomes. The second finds that !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! 26 AFRODAD: African Forum and Network for Debt and Development. Publisher of the Reality of Aid in Africa series that works to establish a truly African perspective of the issues' by consulting with a variety of country leaders, policy makers, and researche rs in Kenya, Tanzania, Zambia, and Zimbabwe. 27 Kalima 2004, 14 28 Nowak 2014, 459
! *" generally aid is ineffective, and at its worst is detrimental to development. The third group is the largest, and it find s aid to be effective under certain circumstances. While the third group may appear to have taken a middl e road instead of finding a clearly defined relationship, it probably is the most accurate representation of real life applications of aid. In addition, I will include considerations for the methods as they relate to the findings The section will proceed by reviewing the major findin gs from each of these groups f ollowed by their respective methodologies in ord er to set the foundation for the project. Positive Aid Effects The most important finding in terms of aid being effective is demonstrating aid to have a positive effect on GDP. Hansen and Tarp found tha t aid does increase the growth rate, which they measure in real GDP per capita, and that this is not conditional upon a good' policy environment. 29 Less optimistic but still positive findings from the paper Politics and the Effectiveness of Foreign Aid by Peter Boone found that aid does not effectively increase investment, but it does increase the size of government, and found again that this effect does not vary between liberal and repressive regimes. 30 The linked result between both of these papers i s that their findings of the effects of aid were consistent across var ious political regime types. In addition to examining effects between regime types, it has also been a point o f aid researchers to examine the effect of aid on government services. One of the se such effects is the tax revenue system !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! "( Hansen, Henrik, and Finn Tarp. "Aid effectiveness disputed." Foreign aid and development: Lessons learnt and directions for the future (2000): 103 128. 30 Boone, Peter. "Politics and the effectiveness of foreign aid." European economic review 40, no. 2 (1996): 289 329.
! *# The finding that aid has no negative impacts on tax revenues is a positive effect for the governments of developing co untries ; it means that they are able to collect taxes that in turn fund the government itself In the research regarding tax effort in relation to aid type, many have studied whether grants or loans have different effects As cited in Clist and Morri ssey's paper, Gupta et al (2004) found that grants negatively correlated with tax revenues while loans are positively correlated. 31 The logic behind this is that when aid is given in a loan format, it is more likely to be well managed by governments who know that repaym ent is on the horizon. 32 Later research on the topic of grant versus loan affects conducted by Clist and Morrissey found that the association between grants and tax revenue became positive after the mid 1980's. 32 Contrary to this finding concessional loans become more prevalent at this time. They go on to argue that grants are preferable to loans since they do not raise the debt burden of developing countries, and according to them do not hinder tax revenues. The article published by Camilia Minoiu and S anjay Reddy find that "developmental aid has a positive, large, and robust effect on growth, while non developmental aid is mostly growth neutral and occasionally negatively associated with growth." 33 Individual case studies are also likely to find positive effects of aid. In the case of Botswana, which became independent in 1966 as one of the poorest countries in the world, is currently a middle income country that sustained one of the world's highest growth rates. 34 While much of this growth can be accredited to !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! #* Clist, Paul, and Oliver Morrissey. "Aid and tax revenue: signs of a positive effect since the 1980s." Journal of International Development 23, no. 2 (2011): 165 180. 32 Pivovarsky, Alexander, Benedict J. Clements, Sanjeev Gupta, and Erwin Tiongson. Foreign aid and revenue response: does the composition of aid matter? No. 3 176. International Monetary Fund, 2003. 32 Clist, Morrissey 2011, 162 33 Minoiu, Camelia, and Sanjay G. Reddy. "Development aid and economic growth: A positive long run relation." The Quarterly Review of Economics and Finance 50, no. 1 (2010): 27 39. 34 Pivovarsky, Clements, Gupta, Tiongson 2003 16
! *$ natural resource reserves, specifically diamonds, international aid was a crucial resource that the government used strategically to develop physical and social infrastructure " 35 In conclusion, positive aid effects have been found on the macroeconomic level as well the microeconomic As time has gone on, these studies can be done again with new data, and thei r results are sometimes confirmed and sometimes refuted. The aid industry has developed as time has gone on and so it is important to continue to build upon and expand the research that has already been done. Aid has become a policy tool of both states, institutions, and the non governmental sector, so its ramifications have only become more important to under stand. Negative Aid Effects The negative result found when looking at aid effectiveness is an especially distre ssing finding. Often these studie s find at least occasional evidence of when aid has been detrimental to growth in developing countries. The threat that IFI's and international aid institutions suffer from the same western dominated policy prescriptions, as seen in UN and other international doctrines, is a credible one. The following s ection is a review of studies who have found various negat ive effects of foreign aid. Doucouliagos and Paldam have done multiple studies to determine the effectiveness of aid. In their 2008 study, they found that while the total average of their meta data analysis of existing aid effectiveness literature was po sitive, that this result ultimately was insignificant and !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! #% Maipose, Gervase, Gloria Somolekae and Timothy Johnston. "Effective Aid Management." Foreign Aid in Africa (1997): 16.
! *% falling. 36 In an update to this paper, the authors revisit whether or not the literature has finally overcome the aid ineffectiveness result', and found their original statement still held. 37 They do stipulate that "some aid components may have a positive effect on growth" but also regard those results as unconfirm ed and in need of additional research 38 Dambisa Moyo published Dead Aid w here she found that th e effect of aid to be overwhelmingly negative. The simplest summary of her argument, Aid has helped make the poor poorer, and growth slower." 39 To clarify the findings from Doucouliagos and Paldam 2010, which uses an aggregate data set of data effectiveness research, "Aid in effectiveness is obvious from just simple raw averages of the reported averagesas the number of estimates has increased, the partial correlation of aid and growth keeps declining." 40 In addition to those aggregates, Moyo finds that "over the past thirty y ears, the most aid dependent countries have exhibited growth rates averaging minus 0.2% per annum." 41 Conditional Effects Beyond the works that claim to find an overall effect of aid, some researchers have found aid to be effective under particular conditions. This result is perhaps the most common in aid research, not because it is the sole finding of the majority of projects but because most projects !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! 36 Doucouliagos, Hristos, and Martin Paldam. "Aid effectiveness on growth: A meta study." European journal of political economy 24, no. 1 (2008): 1 24. #' Doucouliagos, Paldam 2011 #) ,-./-.0123-45!620728!"+**5!* 39 Moyo 2009, xix 40 Doucouliagos, Paldam 2011, 8 $* Moyo 2008, 46
! *& find some exception to their overall finding. I will review the split on methodologies in a later section Burnside and Dollar 2000 found a conditional positive effect of aid on growth. The positive relationship is correlated to developing countries who have good fiscal, monetary, and trade policies. 42 The paper advocates for more concessional loan giving, claiming that [t] o the extent that international capital markets are imperfect, foreign aid can have an important impact on a poor country." 43 Overall, Burnside and Dollar found aid to have an important effect, which could be positive given the right policy environment. An update to the Burnside and Dollar paper published in 2003 refuted their findings on account of new data. New Data, New Doubts update s the original data set with newly available data, as well as extending the time frame considered from 1993 to 1997. 44 With the new data the authors do not find the conclusions of Burnside and Dollar to hold. I include the Burnside and Dollar results to b oth demonstrate the evolution of findings in aid effectiveness and also to demonstrate the effects these findings on the aid industry. The Burnside and Dollar result provides a role and strategy for foreign aid." 45 Paul Collier also advocates for a conditional setting for aid to be effective. In his theory, he stipulated that t iming also is an important compo nent to making aid as effective as possible. His work centers aid as helpful in the immediate aftermath of a turnaround. However that comes !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! $" Burnside and Dollar Burnside, Craig, and David Dollar. "Aid, policies, and growth." American economic review (2000): 847 868. $# Burnside. Dollar 200, 847 44 Easterly, William, Ross Levine, and David Roodman. New data, new doubts: A comment on Burnside and Dollar's" aid, policies, and growth"(2000) No. w9846. National Bureau of Economic Research, 2003. $% Easterly, Levine, Roodman 2003 1
! *' with a few catches. One such example is that "[m] oney early in reform is counter productive." 46 His criticism of the aid industry is not that it doesn't do enough, but that it does not do enough at the right times. He also advocates for increasing overhead costs of aid agencies in failing states 47 For obvious reasons this theme is not popular among the NGO sector, as most gain public funding by advertising a high percentage of donations going towards whichever aid t hey intend to provide. This can only be achieved through low overhead costs. I will assess this hypothesis from a mostly quantitative perspective. I have built a dataset tailored to the tim e and countries in the study. By combining datasets from Aid Data 48 and World Bank 49 I was able to build a more complete set of data to work from tha n I would have using onl y one source. This is crucial to any quantitative investigation as these data points are repr esentations of real world case s. So, by having more data, I can be more confid ent in the relationships that I find. There were concerns about whether combining t wo different sources would risk inconsistencies in the raw data. In order to avoid this, I was sure to use only indicators that shared common units, for example current U.S. dollars The Aid Data used was an !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! $& Collier, Paul. The bottom billion: Why the poorest countries are failing and what can be done about it Oxford University Press, USA, 2008. $' Collier 2008, 118 48 Edgell, Amanda B. (2017). "Foreign aid, democracy, and gender quota laws." Democratization 26 (6): 1103 1141. DOI: 10.1080/13510347.2016.1278209. 49 World Bank. 2017. World Development Indicators. Washington, D.C.: The World Bank. http://data.worldbank.org/data catalog/world development indicators Hypothesis I. Foreign aid had a malignant effect in Sub Saharan Africa that was not present in South or Southeast Asia. II.
! *) aggregated set; where relevant the researche r s u sed an Aid Data currency deflator in order to convert the numbers to current US dollars. Theories of African Underdevelopment Sub Saharan Africa has become a popular testing ground for various questions on development The major reason for this is it remains the only region in the world that has yet to show a general and consistent trend towards improvement. Many theories have circulated ranging from colonial pasts, geographic differences, and cultural differences. I will briefly review the main findings o f such arguments while outlining the major developments in the region 's history in relation to development. Daron Acemoglu 50 creates a very compelling theory that posits the type of modern institutions as a product of their colonial pasts. A point to note is that all the regions included in this study have colonial pasts, and furthermore share the British, the Dutch, and French as central colonizing powers. His theory is not entirely divorced from the geographic theory, but he positions ge ograph y as an antecedent variable to the independent variable of institution type. He divides imperial conquests into two types, where people wanted to settl e and where they wanted to take resources from. C olonizers evaluated the value of a place as a settlement based on survival rates. Since malaria was a prominent killer of the time it was preferable to extract the natural resources present from areas with a high incidence of malaria. These cases established extractive institutions, which today translate into poor performing states. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! 50 Acemoglu, Daron. "Root causes." Finance & Development 4 0, no. 2 (2003): 27 43.
! *( The geography theory is what one would imagine it to be. Essentially the theory is a version of natural resource c urse, but with a negative onset. A vers ion of the ar gument is advanced by Jared Diamonds in Guns, Germs and Steel (1997), where the geographical determinants and raw materials are credited for different economic development outcomes. 51 Ano ther version of this theory is known as natural resource curse. The idea is that natural resource wealth encourages poor government management and can facilitate Dutch Disease. Both concepts are central to Paul Collier's book, The Bottom Billion 52 He outlines four traps that according to him a re either individually or contingently trapping the bottom billion from developing. The resource trap is an extension of the economic theory of Dutch Disease 53 and is essentially that natural resources in high amounts lead to Dutch disease and are at risk for price volatility (oil is a prime example of this) and that "[t]hey inhibit growth even if a country's politics are reasonable." (Collier, 50) However there are cases where natural resources have benefited the country A good example of this is Botswa na, where excellent policy management allowed for benefits from the diamond trade to benefit the country. 54 Natural resources have not conclusively been found to have a casual negative r elationship to growth in Southe ast Asia nor Sub Saharan Africa. Cases such as Indonesia and Botswana have both overall benefitted from natural resource incomes !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! %* "#$%&'()!*$+,(-!./0'1)!2,+%1)!$'(!13,,4-.!56778952 Collier 2008 53 Dutch disease was named after the effect of oil on the Dutch economy. The idea is that as one export becomes ultra competitive, it crowds out other export markets in the economy. This over the long run is hurtful to growth. 54 Carlsson, Jerker, Gloria Somo lekae, and Nicolas Van de Walle, eds. Foreign aid in Africa: Learning from country experiences Nordic Africa Institute, 1997.
! "+ Another aspect of geography worth mentioning is coastal access. The general theory behind its importance is that easier acce ss to the coast equals easier access to trade. Since the ability to trade is closely tied to economic growth, coastal access becomes an important perquisite for some. The case of South Asia, specifically the experience of India with growth as a function of the service industry suggests that while this theory may have been important in the early development of Southe ast Asia, it was not a large contributing factor to South Asian development and is perhaps not necessary to Sub Saha ran African developme nt. Still, just because it is not necessary does not mean it would not be helpful. The time at which African economies were able to enter competitively into the global market as opposed to Asian economies has also been cited as a reas on for under development in Africa. Collier claims that they missed the boats' in terms of producing exports and bringing in agglomerations as was seen in Asia. The basic process of agglomerations is that once one company goes to a new region and is suc cessful, other companies will soon follow. This increases the jobs, infrastructure, and potentially wealth to the new region. This process can be seen in the textile industry which is heavily concentrated in South Asia. However, the untraditional success of South Asia proves that "[t]here is a new boat that development latecomers can take." 55 For this study, I question why aid has not b een able to bring Africa to an entrance point to the international market Initial research points us to the idea that aid itself has played a limiting role in Sub Saharan African countries. The modern aid industry heavily favors free mark et principles. I n Asia, some economies have defied the free market !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! 55 Ghanz, Kharas 2008 xvii
! "* principle by succeeding under protection ist policies that were conve ntionally thought of in the western sphere as stifling to growth. Theories of South and South e ast Asian Development South e ast Asia boasts some of the strongest growing economies in the world. Aside from obvious geographic differences (more coastal access), the region also differs from Africa in its economic policies. By the early twentieth century, a very different view of fiscal functions of the state was taking hold, a view based on the idea of a strong central state taking responsibility for bot h revenue raising and expenditures." 56 This idea of a strong central state in charge of fiscal policies is something discouraged by the modern aid industry in Africa. Interference by world powers in national policy is not unknown in Southe ast Asia. "A number of treaties were signed with major powers in the latter part of the nineteenth century which obliged governments to keep tariffs low and permit greater trade" 57 Thailand was the only country to remain free of direct colonial rule but at a "cost of surrendering considerable autonomy in economic policy." After independence between the 1950's 1960's, Asian countries diverged from the previously similar African track of development. "[T]he Asian countries embarked on a very successful human capital f ormation programs and, in addition, developed and implemented pro !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! %& Booth, Anne. "Colonial revenue policies and the impact of the transition to independence in South East Asia." Bijdragen tot de taal land en volkenkunde/Journal of the Humanities and Social Sciences of Southeast Asia 169, no. 1 (2013): 37 67. 57 Booth 2013 38
! "" growth policies that resulted in tremendous improvement in living standards across the region." 58 This finding would point us to the question of governance being a key factor in the outcome of aid in each region. However, external variables continue to play a role. The debt crisis of late 1970's prompted private sector lending to slow. 59 Development economists posit that a country must achieve rapid agricultural growth in order to reach rapid non agricultural growth. 60 This is supported by many example s throughout history where a country moves into industrialization, the agricultural sector is largely replaced or heavily supplemented by the industry sector. The H PAE ( defined as High Perfo rming Asian Economies) economies have experienced larger growth rates in their industry, while most growth in Sub Saharan Africa has been in agriculture. 61 Although agriculture still plays a large role in Asian economies, especially in rural communities, t he expansion of industry is what is driving their economic growth. South Asia' s experience offers another path to development that challenges this theory. 62 Their development path points out some interesting observations that would suggest technology has prov ided another development pathway Here, coastal access, while it may have been relevant to development in the past, is no longer a necessary means to industrialization and thereby development. The reason technology has opened a new pathway is that the service industry relies less on infrastructure like ports and roads, and more heavily on access to electricity (Ghani Kharas, !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! %) 924:2;2!"++&5!# 59 Maswana 2006 60 Maswana 2006 &* 924:2;2 "++& 62 Ghani, Ejaz, and Homi Kharas "The service revolution in South Asia." SERVICE REVOLUTION (2010):
! "# xvi). Although aid to Africa went through an infrastructure focus period in the 60's, the region suffers from lack of regional maintenance of these projects. In this case, globalization offers a solution. The increasingly integrated world has made technological advances that have broken down the barriers of time and space. Tradability, which is named as one of the contributing fa ctors to the Service Revolut ion in South Asia, is defined by the fact that modern services face fewer government barriers such as border con trol and tariffs. While Southe ast Asian development was led by capitalizing on the manufacturing industry, South As ia has capt ured the service industry 63 Southe ast Asian development has had cases of good governance, and a more favorable entrance time into the international markets to distinguish its economic development from that of Africa. South Asia has followed a modern track of development that essentially uses technology as a comparative advantage. What remains to be seen is what role aid played i n this development. Because the region was the focus of aid under a different aid era than Africa it remains a question as to whether South and Southe ast A sia reacted differently to foreign aid intrinsically, or perhaps the method of foreign aid of the time period caused different outcomes The answer to this question could not be more cr itica l. Today, a number of Sub S aharan African have economies whose net inflows of aid exceed the fifty percent mark of their total GDP 's. If it is the case that aid is impeding African development, then it can be said it is also impeding African economic i ndependence. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! Ghani, Kharas 2010, 2
! "$ Method Review Researchers have looked to use both economic and social indicators in order to evaluate the effect of aid. However, most projects focus on one side or the other. For example, large meta data studies often use long run analysis of GDP growth to de scribe relationships between aid and growth. Microeconomic studies often include more indicators from the social sector, for example how many new public schools were opened. The micro macro paradox is an established conflict in aid effectiveness research. In an article published in 1987, Paul Mosley notes that the microeconomic data from various aid projects were all encouraging. However, even then, "from regressions of aid on growth across a cross section of developing count ries, are discouraging." 64 In the later meta analyses performed by Doucouliagos and Paldam suggest the few positive averages they found were related to the micro macro paradox where "half of all aid projects wor k, and few harm the recipients, but still th e aggregate has no effect." 65 Doucouliagos and Paldam offer a solution to the micro macro paradox that advocates for researchers using disaggregated studies rather than aggregates in order to find a more accurate way of describing the relationship between d evelopment and aid. 66 This finding in addition to the comparative historical analysis method used in Capitalist Development and Democracy ( Rueschemeyer, Stephens, and Stephens 1992 ), were the most informative for the method I chose for this project. While the aggregate studies do provide insight into the averages across time and !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! &$ Mosley, Paul. "Aid effectiveness: The Micro Macro Paradox." Ids Bulletin 17, no. 2 (1986): 22 27. &% Doucouliagos, Paldam 2011, 5 && Doucouliagos, Paldam 2011, 9
! "% space that would be impossible to cover without using an aggregate method, they do seem to be missing and or misconstruing some of the results that are present in the microeconomic studies. In addition to overall method, there are a variety of ways that researchers have to sought to account for the micro macro paradox by choosing to define variables in different ways that are meant to capture their truer' behavior. For the purpose s of space I cannot review all of them in this paper. I have included some that can serve as examples of how this practice has been used in aid effectiveness research and what the implications of these are on aid research overall In Moyo's analysis of aid, she does not distinguish between grants and loans (a distinction made in the Clint and Morissey papers) on the grounds, "If a large share of foreign loans are provided on highly concessional terms, loans are frequently forgiven, policymak ers in poor countries may come to view them as roughly equivalent to grants, and as such the distinction between [aid] loans and grants i s practically irrelevant." 67 This is an example of how the definition of a given variable in a study can vary between studies, and thus could lead to conflicting results when used in an aggregate study In theory, grants and loans w ould be expected to be treated d ifferently by recipient governments. Treating the variables under the assumption the effects should be diffe rent can lead to a variety of biases in results. This could indicate why there has been such varied findings in those who study aid outcomes. Macroeconomic studies focus on hard numbers, like a country's change in GDP between a set of given years Microec onomic studies are more likely to us e less overarching indicators, for example the success or efficiency of a newly established health care center. These types of indicators are difficult to quantify and even more difficult to standardize. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! &' 9-<-!"++(5!)
! "& Thus it is diff icult to compare them across meta data studies because each study could have a different measure or interpretation of a certain outcome. The paper by Minoui and Reddy (2010) offers an explanation to the micro macro paradox found in aid research. The authors note that macroeconomic studies seek to investigate the total impact of aid, which includes the non developmental portion of aid that t heir project distin guished from developmental aid, whereas microeconomic studies use projects wit h plausible development impact. Thi s could indicate that the variable s chosen in macroeconomic studies which find negative results between aid and growth have not found indicato rs that are true to the performance of aid. Or it could mean there is a selection bias among researchers o f specific development projects in which the plausible development impact' is already favorable. The question of which approach, macro or micro, pro vides the most accurate big picture of aid largely remains in question. The methods for this study seek to mirror the major pillars of aid effectiveness r esearch both the micro and macro studies I include regression analyses for both regions and time periods to determine their respective experiences with aid during the given time frames. Second, I use a single case study to test the findings of the large N study against an individual state's exp erience with aid. Essentially the single c ase study serves as both a fact check to the findings of the large N study as well as a manageable to include more historical context to the quantitative findings I n order to include aid policy as an independent variable, I had to establish that there was a significant connection between the aid policies of the literature review (focused on U.S. led aid policy) and the aggregate variables used in the dataset (total aid commitments). Second, I had to
! "' find evidenc e that this connection transferred to institutions such as the World Bank and IMF in order to draw conclusions from the data related to aid policy. To make the connection between U.S. aid commitments and total aid I used correlations between the variables to test how closely related the two were. The correlation ( r) between the total aid variable (ad_aid) and the variable for total U.S. (ad_us_aid) is around ( .5 ) This means that the R^ 2 is approximately ( .25 ) which equates to U.S. aid explaining about 25% of the variation in total aid commitments. Generally, an r value of .3 is considered a strong relationship, so the relationship here can be considered very strong. Thi s confirms that U.S. aid commitments ar e a strong proxy for overall aid To confirm that this correlation could be considered to approximate the policy of institutions as well I used research from the literature review that focused on the behaviors of aid giving institutions themselves. They conclusively note that the largest donors to aid institutions control aid policy. 68 "International organizations like the World Bank refl ect and support the interests of their most influential members; in other words, they promote whatever orthodoxy prevails in the industrialized west." 69 In conclusion, I have established that U.S. aid made up a significant amount of the total aid commi t men ts during the periods included in my study and that the aid policy of institutions is controlled, or even strongly influenced by the largest donors. Ther ef ore, I can use U.S. aid policy as a proxy for international aid policy in this study. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! &) Siddiqui, Rukhsana A., ed. Subsaharan Africa in the 1990s: Challenges to democracy and development Greenwood Publishing Group, 1997. &( :#((#;0# 6778 )!6<=
! ") The project uses aid policy between th e two decades of the study as an independent variable. As was reviewed in the literature review, the aid eras were d istinct in their policy focus. By establishing that U.S. aid does correlate significantly with total aid commitme nts and considering that the highest donors control aid policy, I can comfortably use U.S. aid policy as a proxy for overall aid trends. While it would be interesting to consider differences in aid policy between donors and their subsequent outcomes for t he interests of this project it was the overall trend in aid policy that was the independent variable of interest in the study Results The figure above shows two scatter plots for both regions considered in the study. The most important takeaway from this is that there are far less observations for the Asia cases than Sub Saharan Africa. This is in part due to the fact that the data from that time period, 1960 1970,
! "( is less complete. In addition, there are about thirty eight more cases from the Sub Saharan Africa region. This is a concern mentioned in the concluding section, but for the concerns of the results I included all of the cases possible in order to cast the biggest net in terms of identifying relationships. What these graphs are able to suggest is that both regions shared a common spread of GDP growth in percentage, and that Sub Saharan Africa, for the cases considered, received significantly more aid than did any of the cases from the Asia regions. This supports the finding from the literature that overall as a region Sub Saharan Africa has taken on more aid than did the Asia region. The regression line, which tells us what the average relationship is (either positive or negative) demonstrates the straigh t line that most closely averages all the data points. From my data, the regression line shows a negative relationship between total amounts of aid in the Asia cases. Here is where it is most important to note the lack of cases in comparison to Sub Saharan Africa. As the regression line is generated from finding a closest common line between data, it is heavily influenced by outliers, and thus most accurate with the maximum c ases. The relationship for the Sub Saharan Africa region shows a mild positive relationship. The straightforward interpretation of this is that the aggregate effect of aid on the region has been positive. The more complex interpretation is that althou gh it is positive, the slope is moderate, and considering there are so many more cases I would expect t o see a steeper slope to infer a significant relationship.
! #+ I used bivariate correlation to test how closely the data was to the regression line. For a very strong relationship, meaning the regression line represented the data very well, I would expect to see a correlation, further stated as r as close to 1.00 or 1.00 as possible. Generally, an | r|= .3 is considered a moderately strong relationship. The correlation for the Asia cases (represented by asia=0), show a correlation between total aid and GDP growth of 0.148. This does support the regression line in that it shows the relationship as negative, but the r value is insignificant. This means that the regression line is not a good measure for the Asia cases The Africa correlation (represented by asia=1) also shows an insignificant r measure. The value |r|= 0.09, although positive is extremely low. These tables also solidify the disparity between observations. The Sub Saharan cases have 417 more observations than the Asian cases. Since the number of observations is so high in this case, I can conclude from this data there is not a significant relationship between total aid and GDP growth in Sub Saharan Africa in the 1990's. Although the bivariate regressions and correlations did not yield significant results, that does not mean there are not signific ant relationships in the data. As suggested in the literature review, in order to capture the full effect of aid, more than one variable is needed. In order to test
! #* for this, I use multi variable regression analysis. This test allows me to use more than o ne independent variable at a time and measure the effects of those on a single dependent variable. In the multiple regression analys e s I used the macroeconomic indicator of growth in GDP (GDP_g) as the dependent variable of interest By using the growth variable and not GDP or GDP per capita I can proxy the growth rates of each country controlling for population and size. I used the following set of independent variables: total aid, foreign direct investment, infant mortality and life expectancy for both regions in separate regression analyses. The 1960 1970 South and Southeast Asian data specified by the binary variable Asia equal to zero (Africa coded as Asia=1), yielded a moderate predictive power of the independent variable s given above on growth I found F (4,62 ) = 2. 46 and i t did not yield a statistically signif icant corresponding P score. The R ^ 2 is equal to 0.1 4 which corresponds to a moderate strength of fit What thi s boils down to is that while the F ratio d id not reach statistical significance, the R ^ 2 did. So the regression model fits the results moderately well. This can be interpreted as the independent variabl es of this model (total aid, foreign direct investment, infant mortality life expectancy ) explain the dependent variable of growth moderately well. While the significance could be higher, it means that this model fits about 8 % of the of the variance in GDP growth. In the same analysis for the Sub Saharan Africa region, the moderately strong predictive power of those indicators drops off. Using the same independent growth variable and same set of independent variables, I found F (5,452 )= 4.14, p<.001 and an R^ 2 =.04 13 So, this regression model explains only 4% of the change in GDP growth, which is half of the variation explained by the same indicators in the Asia model. The F ratio, which measures the mean square for the model to the mean square of the residual is statistically significant in this regression Although the p value
! #" is highly significant, the R^ 2 value fell below 0.1 What this means is that there is something not captured in the regression analysis that is driving development in the region. Wh at these regression analyse s suggest is that for the time periods considered, the independent variable set tested (total aid, foreign direct investment, infant mortality life expectancy ) explains more of the GDP growth in the South and Southeast Asian cases, than in the Sub Saharan Africa cases. What this means to the project is that the common indicators for aid research suit the Asia cases bet t er than Sub Saharan Africa cases in relation to explaining growth. In the next section I will use similar te sts to examine the case of Nigeria in depth. From the large N portion of the study, I can conclude that total aid did not have a significant impact on GDP growth of any kind on Sub Saharan Africa in the 1990's. I am not able to judge the same relationship with equal weight in the Asian cases. However, since the number of observations is above the large N standard of a sample size of 30, I can say that my data did not yield a significant relationship between aid and GDP growth in this case either. Nigeria: In Depth Analysis I chose Nigeria as a spot light case because it has a unique blend of common factors and uniqueness form its neighbors My initial research also indicated that Nigeria had had a particularly problematic relationship with aid. A s previously mentioned, the micro macro par adox in aid research describes a split in the findings of aid outcomes In the case of Nigeria foreign aid is credited with prolonging their Civil War 70 I included t his section to test !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! '+ Nunn, Nathan, and Nancy Qian. "Aiding Conflict: The Impact of U.S. Food Aid on Civil War." NBER, www.nber.org/papers/w17794.
! ## whether or not my findings for the large N stud y hold in this case, and if not, where is it that they do not My hope is that this section in addition to testing the results of the large N study, will also demonstrate a clear example of how foreign aid could be harmful The broadest hope for this project is that by improving the way we test aid effectiveness, we can improve upon aid policy. This of course rests on the assumption that the aid industry would respond accordingly to such findings, but that topic is for anoth er paper. The results here, if different from the findings in the large N studies, c ould support the idea that the current research indicators do not paint an accurate picture of the effects of aid S inc e I am only using one case here, the results are n ot necessarily transferrable to the region. But it will provide a more detailed account of the interaction of Nigeria over time with foreign aid. More importantly, this is an important way to include historical analysis in order to build a more complete p icture of one case and by doing so demonstrate some of the details that may be missed with a large N study. The section will proceed with a brief historical overview, first of the country overall and second of the Biafran Civil War As was reviewed in the aid history outline, the aid policy of the 1990's was geared toward engendering democracy and was most often given in concessional formats because of this. I focus on this event specifically as it was cited in numerous papers as an example of how foreign aid can be harmful. I then utilize the same correlation and multiple regression analysis used in the large N study to see what the macro economic experience was during the time period of the study.
! #$ Brief Historical Overview Nigeri a is located on the Western coast of Africa and has the largest pop ulation of any African nation. After World War II, a series of constitutional reforms allowed Nigeria to have greater autonomy from the ruling British colonial forces and empowered Nigeri an leaders. Although formal independence occurred in 1960, the country cycled between coups and military heads of state until a peaceful transition of government in 1999. In 1970, the Biafran War led to one of the most controversial humanitarian crisis o f the post World War II era. Britain was the imperial ruling power of the region that would become Nigeria. In the earliest part of the twentieth century, the region was divided between North and South and was ruled as two separate protectorates. This style of imperial power promoted ethnic tensions through their structuring of regional organizations. The two regions were combined into a sin gle colonial government in 1914; but they maintained their regional ruling offices. "In spite of Nigeria's common colonial experience, the record also emphasized the local differences in administrative practices, if not in policies, going right back to the early years of this century." 71 When the regional ruling was dissolved in 1966, the event "helped to set off violent reactions in the north against southerners who had settled [there] contributing to the outbreak of civil war." 72 The ethnic tensions in Nigeria arose from disputes that crossed ethnic and wealth boundaries. While the Biafra conflict resulted in ci vil war, there were credible threats of !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! 71 Heraclides, Alexis. "Secessionist minorities and external involvement." International Organization 44, no. 3 (1990): 341 378. 72 Helen Chapin Metz, ed. Nigeria: A Country Study Washington: GPO for the Library of Congress, 1991.
! #% secession since the 1950's that mostly centered around regional representation in the Central Legislature. 73 The Southe ast region of the country encompasses the Niger Delta, which is a petroleum rich region. This ga ve the region an increased possibility of gaining independence. According to the Library of Congress, political parties fell harshly along geographic regions. Additionally, before the civil war there was a shortage of government jobs. Ethnic rivalry for these limited positions further stressed the national fabric of Nigeria The Biafran Civil War In the aid research completed for the literature review the Biafran Civil War was the most commonly cited event for poor aid outcomes. Both regions included in this study have a history of colonialism followed by cycles of civil conflict. There are probably many individual cases of conflict where aid played a measurable role that would be useful to future aid project s For now, I will use this particular case in order to combine more of the historical record with the quantitative study as well as a test case f or the method used in the large N portion of the project. In 1967 a secessionist movement led to the outbreak of civil war in the country that introduced an influ x of foreign aid in the country. The former Eastern region of the country seceded and sought to set the precedent for ethnic minorities self determination within the region of Nigeria. The movement was led mostly by the Ibo minority, under milita ry leader Odumegwu Ojukwu. The Biafran state largely appealed to the international community on the grounds of self determination. This event speaks to the internal post independence struggles of Nigeria's reconciling both their colonial past as well as ethnic diversity. This was not an uncommon !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! 73 Heraclides, Alexis 1990
! #& experience in Sub Saharan countries after decolonization, but in the Biafra case foreign aid took a particularly critical role. The Biafra conflict began in 1967 and concluded in 1970. In short, foreign aid sus tained the Biafran conflict longer than what would have been possible had the region received no foreign aid throughout the conflict. In a study examining US Food Aid and Civil Conflict, researchers found that "an increase in U.S. food aid increases the incidence and duration of civil conflicts [and these] effects are most pronounced in countries with a recent history of civil conflict." 74 In sum, they found food aid to prolong civil conflict, a nd that this effect was increased in a region who had already experienced civil conflict. =>?@!A@B@0!0@27@A!C7.8@3:.!CD.E:.!-;0?@!4?1I8@;F4!-G!?.82;1F2A12;!217!200-:@7!CD.E:.!F-! /1A/.8M@;F!F?@!41@3@!F?2F!?27!B@@;!I02/@7!-;!H12GA2!B?@!G--7!217!204-! 200-:@7!CD.E:.! F-!G@@7!?14!2A8
! #' country is able to recover more quickly from the negative growth incurred during conflict and thus takes in less aid Meanwhile, a country that has already experienced negative growth from conflict, then experiences the negative effects of introducing aid with a multiplier effect. Nunn and Qian remark that food aid is often a target of armed factions that either steal and sell the aid to those in need, or simply use i t to feed their own forces. In the Biafran case, this example is all too true. The Crisis Caravan suggests that shipments of humanitarian aid caused the Biafran conflict to last years longer than it would have otherwise. 75 This also means that in this cas e aid caused more harm, although "[m] ost of them [donors] were genuine in their humanitarian efforts, but little did they know that most of their contributions were used to purchase arms and ammunition which prolonged the war and thereby increased and heig htened the sufferings of those they were trying to help." 76 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! 75 Polman, Linda. 2010 76 Atofarati, Abubakar A. "The Nigerian civil war: Causes, strategies and lessons learnt." retrieved from htpp://www. africamasterweb. com/ BiafranWarCauses. html (1992).
! #) There is something to be said that these governments were not totally unwise to the effects their aid was having. "Some foreign governments covertly encouraged and sustained rebellion under the guise of humanitarianism by secretly giving weapons and other material to Biafra." 77 According to my data, GDP growth a nd total aid to Nigeria did not have a clear correlation during the 90's in Nigeria as demonstrated in the graph above While this finding does not support the findings of aid having a negative relationship to growth, it is also important to note that this project only looks at snapshots of the aid effect, not it's cumulative effect. In addition, I do not measure for th e effect of prolonged civil conflict had on Nigeria in the immediate time following the end of the civil war. A larger concern here is that humanitarian and development aid has become another tool for dominant world powers to maintain control over the dev eloping world. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! 77 Atofarati, Abubakar A. 1992
! #( In the multiple re gression analysis: F(4,6)=.67, with the p value too large to be anywhere close to significant. What this means is that the variables that yielded a statistically significant measure of the Asia cases, and a statistically i nsignificant measure of the Sub Saharan cases, has an even lower significance level in the case of Nigeria. In the Sub Saharan cases overall, the same regression model explained about 4% of the change in growth and the p value correlated to a highly signif icant p value but an insignificant R 2 In the case of Nigeria the R 2 fell to .3101, meaning it explains only 3% of the variance in GDP growth and the p value was not significant The ramifications of civil conflict are outside the scope of my project, but it cannot be ignored that aid in its present forms has had proven negative consequences. Although the Biafran conflict falls outside my timeframe, civil war typically reduces growth at rate of 2.3 percent. 78 This is a critical part of the comparative historical analysis because it proves that aid by prolonging the civil conflict decreased growth for at least a year beyond what the civil war would have caused had aid not been introduced. In this particular ca se, aid had no clear effect on the overall growth of Nigeria. From this case study, I am not able to prove my hypothesis that aid has a malignant effect in the Sub Saharan cases. I suspect that in the case of Nigeria aid did more harm than good by prolonging the civil conflict. However, since this time period is not included in my data I canno t confirm nor deny its aggregate effect on the growth of the country. I did find that aid had no significant effect on gr owth in time period of my study. I c a n infer that in the past aid had a demonstr a ble negative effect on growth. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!!!!!!!!!!!!!! ') P-001@A!"'5!"++)J!
! $+ Conclusions In conclu sion, I cannot confirm my hypothesis that aid is had a malignant effect in Sub Saharan Africa, but I am able to say it did not have a statistically significant positive effect. In the South and So uthe ast Asia cases, I was unable to confirm the significance of the relationship betwee n aid and growth. T he l ack of equal am ounts of data from both regions prevented a purely comparative study. There also is a natural disparity between the number of cases. I was aware of this bias in the original design of the methods of the project, but I felt that the influence of different aid policies was clear enough to compare the two time periods. In order to measu re this accurately I had to en sure that the quantitative tests did not reflect the bias in number of cases. I used the GDP growth variable rather than raw GDP in order to control for the original GDP. This avoids treating small countries with smaller GDP's the same as the larger countries. For example, it would be unfair to s ay that India has a stronger econom y than Malaysia just based on GDP, since the latter is a much smaller country and would have a lower production ceiling. I adjusted to the disparity between number of cases in my analysis of the data by being more conservative with the conclusions about the Asian cases Since I had more complete data and a higher number of observations from the Sub Saharan Africa cases, I was able to be more confident in the relationships. The multiple regr ession analyse s showed that the traditional model of aid, based on growth and aid income, fit the Asia models at a moderate level (statistically significant), and the Africa cases at sta tistically insignificant level. What can be conclu ded from thi s is that the aid model explains growth for the cases of South and Southeast Asia better than the case of Sub Saharan Africa in the 1990' s. The bigger picture of this finding is that the way aid is being evaluated for
! $* effectiveness is not an accurate measure in the Sub Saharan Africa case. This could mean that the foundations for aid policy are not sound and therefore may need to be reformed. The re evaluation of how aid policy is formulated is the most important practical application of t his project as well as the nex t step for continued research. This paper established that the common factors included in evaluat ions of aid effectiveness do not fit the relationship between aid and growth in Sub Saharan Africa. In the case of South and Southe ast Asia, the model explained growth at a statistically si gnificant level, so this is the central divergence between the cases. The next step in research on this topic is to further investigate the differences in aid type and delivery in order to sh ow what types of aid are effective in what context. It would also be i mportant to look at global financial policies in relation to aid policie s. My initial research as well as the historical research of the Nigeria case study both suggested that there were structures in the global market preventing growth in various ways. Whether or not that is a function of the aid itself or if aid is a symptom o f these structures is another topic that would need to be investigated. There are many directions for aid research to continue in, and as aid is a tool of development policy, it is worth researching all of its effects and impacts. In conclusion, this project started with the very simple observation of a split between t he developed world and the under developed world. Foreign aid is a mechanism intended for supporting the developing world, but in the Sub Saharan Africa case it is clear it is not responding as well, or even at all, to aid. It is the responsibility of the world to find and support the best p o licies a nd mechanisms to develop the developing worl d. This study although limited in scope, establishes the need to change how we evaluate aid, and confirms that aid did not have a positive effect on aid in Sub Saharan Africa in the 1990's.
! $" Works Cited 1. Acemoglu, Daron. "Root causes." Finance & Development 40, no. 2 (2003): 27 43. 2. Atofarati, Abubakar A. "The Nigerian civil war: Causes, strategies and lessons learnt." retrieved from htpp://www. africamasterweb. com/BiafranWarCauses. html (1992). 3. Boone, Peter. "Politics and the effectiven ess of foreign aid." European economic review 40, no. 2 (1996): 289 329. 4. Booth, Anne. "Colonial revenue policies and the impact of the transition to independence in South East Asia." Bijdragen tot de taal land en volkenkunde/Journal of the Humanities and Social Sciences of Southeast Asia 169, no. 1 (2013): 37 67. 5. Booth, Anne. "Colonial revenue policies and the impact of the transition to independence in South East Asia." Bijdragen tot de taal land en volkenkunde/Journal of the Humanities and Social Scie nces of Southeast Asia 169, no. 1 (2013): 37 67. 6. Burnside and Dollar Burnside, Craig, and David Dollar. "Aid, policies, and growth." American economic review (2000): 847 868. 7. Carlsson, Jerker, Gloria Somolekae, and Nicolas Van de Walle, eds. Foreign aid in Africa: Learning from country experiences Nordic Africa Institute, 1997. 8. Clist, Paul, and Oliver Morrissey. "Aid and tax revenue: signs of a positive effect since the 1980s." Journal of International Development 23, no. 2 (2011): 165 180. 9. Collier, Paul. The bottom billion: Why the poorest countries are failing and what can be done about it Oxford University Press, USA, 2008. 10. De Long, J. Bradford, and Barry Eichengreen. The Marshall Plan: History's most successful structural adjustment program No. w3899. National Bureau of Economic Research, 1991. 11. Diamond, Jared. "Guns, germs, and steel." (1997). 12. Doucouliagos, Hristos, and Martin Paldam. "Aid effectiveness on growth: A meta study." European journal of political economy 24, no. 1 (2008): 1 24. 13. Easter ly, William, Ross Levine, and David Roodman. New data, new doubts: A comment on Burnside and Dollar's" aid, policies, and growth"(2000) No. w9846. National Bureau of Economic Research, 2003. 14. Ghani, Ejaz, and Homi Kharas. "The service revolution in South Asia." SERVICE REVOLUTION (2010): 15. Hansen, Henrik, and Finn Tarp. "Aid effectiveness disputed." Foreign aid and development: Lessons learnt and directions for the future (2000): 103 128. 16. Helen Chapin Metz, ed. Nigeria: A Country Study Washington: GPO for the Library of Congress, 1991. 17. Heraclides, Alexis. "Secessionist minorities and external involvement." International Organization 44, no. 3 (1990): 341 378. 18. Kalima Barbara. Reality of Aid: Africa Edition 2003 04 AFRODAD, 2004. 19. Maipose, Gervase, Gloria S omolekae, and Timothy Johnston. "Effective Aid Management." Foreign Aid in Africa (1997): 16.
! $# 20. Maswana, Jean Claude. "Economic Development Patterns and Outcomes in Africa and Asia." (2006). 21. Minoiu, Camelia, and Sanjay G. Reddy. "Development aid and economic growth: A positive long run relation." The Quarterly Review of Economics and Finance 50, no. 1 (2010): 27 39. 22. Mosley, Paul. "Aid effectiveness: The Micro Macro Paradox." Ids Bulletin 17, no. 2 (1986): 22 27. 23. Moyo, Dambisa. Dead aid: Why aid is not working and how there is a better way for Africa Macmillan, 2009. 24. Nowak, Wioletta. "The evolution of development assistance." Journal of US China Public Administration 11, no. 5 (2014): 454 462. 25. Nunn, Nathan, and Nancy Qian. "Aiding Conflict: The Impact of U.S. Food Aid on Civil War." NBER, www.nber.org/papers/w17794 26. "Overview of an Alternative Long Term Development Strategy." Africa's Recovery in the 1990s: From Stagnation and Adjustment to Human Development by Giovanni Andrea Cornia et al., St. Martin's Press, 1995, pp. 159 190. 27. Pivovarsky, Mr Alexander, Mr Benedict J. Clements, Mr Sanjeev Gupta, and Mr Erwin Tiongson. Foreign aid and revenue re sponse: does the composition of aid matter? No. 3 176. International Monetary Fund, 2003. 28. Polman, Linda. The crisis caravan: what's wrong with humanitarian aid? Metropolitan Books, 2010. 29. Siddiqui, Rukhsana A., ed. Subsaharan Africa in the 1990s: Challenges to democracy and development Greenwood Publishing Group, 1997.