FOREIGN AID AND ECONOMIC GROWTH by Ketsia S. Dimanche A Thesis Submitted to the Faculty of The Wilkes Honors College in Partial Fulfillment of the Requirements for the Degree of Bachelor of Arts in Liberal Arts and Sciences with a Concentration in Economics Wilkes Honors College of Florida Atlantic University Jupiter, Florida May 2010
FOREIGN AID AND ECONOMIC GROWTH by Ketsia S. Dimanche This thesis was prepared under the direction of the candidate s thesis advisor, Dr. Kanybek Nur-tegin, and has been approved by the members of her supervisory committee. It was submitted to the faculty of the Honors College and was accepted in partial fulfillment of the requirements for the degree of Bachelor of Arts in Liberal Arts and Sciences. SUPERVISORY COMMITTEE: Dr. Kanybek Nur-tegin Dr. Keith Jakee Dean, Wilkes Honors College Date ii
ACKNOWLEDGEMENTS A special thanks to my mother, Monique Dimanche, for inspiring me with this thesis topic. Without her encouragements, I would not be able to complete it. I thank my father, sisters and friends for their unconditional support and my advisors for their guidance. iii
ABSTRACT Author: Title: Institution: Thesis Advisor: Degree: Concentration: Ketsia S. Dimanche Foreign Aid and Economic Growth Wilkes Honors College of Florida Atlantic University Dr. Kanybek Nur-tegin Bachelor of Arts in Liberal Arts and Sciences Economics Year: 2010 The effect of foreign aid on economic growth is still ambiguous in the economic literature. In this paper, we analyze the effect of foreign aid on economic growth in 79 developing countries from the Latin America Caribbean Region, Africa, and Asia. Using data for the year 2000, we find that foreign aid has an insignificant negative effect on economic growth when we control for other factors. iv
TABLE OF CONTENTS I. INTRODUCTION...1 II. LITERATURE REVIEW...4 III. ECONOMETRIC MODEL...8 IV. DATA...12 V. RESULTS...16 VI. CONCLUSION...19 APPENDIX...20 REFERENCES...23 LIST OF TABLES Table 1. Summary of Predicted Effects of Determinants on Growth in Literature...11 Table 2. Descriptive Statistics of Dependent and Independent variables...13 Table 3a. Least Squares Estimation Results: Main Regression...17 Table 3b: Least Squares Estimation Results: Second Regression...17 Table 4. RESET Test Results...20 Table 5. R 2 from Auxiliary Regressions and VIF...21 Table 6. Countries Used in the Analysis...22 v
I. INTRODUCTION Economic growth and development are in the forefront of international relations and policy making. Developed countries such as the United States and international organizations such as the World Bank and the International Monetary Fund provide assistance that attempts to help the economic growth of developing countries. The United Nations adopted the Millennium Development Goals (MDGs) in September 2000 (United Nations, 2009). The MDGs are representative of the partnership between countries, rich and poor, to achieve a common goal of commitment to increase overall living conditions. Living conditions can improve when a country experiences economic growth. Since their introduction in 2000, the principal goal of MDGs is to reduce world poverty by half by the year 2015. Poverty is measured by the proportion of the population living with an income below one dollar per day (United Nations, 2009). These goals not only target the reduction of poverty but they also include: achieving universal primary education; promoting gender equality while empowering women; reducing child mortality by twothirds; improving maternal health; combating diseases; ensuring environmental sustainability; and developing a global partnership for development (United Nations, 2009). Since poverty is one of the attributes of underdevelopment, development assistance or foreign aid is one way to achieve these goals. Foreign aid is defined as the voluntary transfer of public resources from one country to another to better the human condition in the country receiving the aid (Lancaster, 2007: 9). Foreign aid is measured by the net flow of resources which may or may not facilitate growth. There are two measurements of aid used in foreign aid 1
research: effective development assistance (EDA) and official development assistance (ODA). EDA is the sum of grants and the grant equivalents of official loans whereas ODA includes both the direct grants and concessional loans for which the grant component is above 25 per cent (Daalgard and Hansen, 2001: 25). EDA combines aid in the form of direct grants and the portion of loans that are grants adjusted for inflation to reflect the real cost of providing the aid. ODA comprises direct grants and loans that have a grant component greater than 25 percent. Although the definitions of EDA and ODA differ, the results from using either measurement are comparable when studying the effect of foreign aid on economic growth. There is no clear consensus on the effect aid has on economic growth. On one hand, foreign aid tends to positively affect the growth of an economy if the country possesses good economic policies such as openness to trade and a low inflation rate (Burnside and Dollar, 2000: 847). On the other hand, however, foreign aid can be detrimental to growth in the long run because recipient countries are more likely to become dependent on foreign aid (Rajan and Subramanian, 2005: 5). The purpose of aid, in most cases, is to promote the economic growth of these recipient countries. If, in the end, these countries depend on the donor countries to sustain and finance their activities, then the aid contradicts its original purpose. Also, the returns to aid are believed by some to be negative in the long run because the increased inflows of aid defeat their original purpose and allow these disbursements to be wasted in inefficient economic activities (Lansink and White, 2001). Thus, the long run negative returns of aid indicate that a country has potentially received too much aid (Lansink and White, 2001: 48). In other 2
words, the levels of aid determine the relationship between growth and aid in the long run. In this paper, we look at the effect of foreign aid on economic growth for developing countries. This research is needed to assess the effectiveness of aid giving on promoting economic growth. Most studies have analyzed the overall effect of aid on economic growth for Asian countries (Asteriou, 2009 and Burke and Ahmadi-Esfahani, 2006) and Sub-Saharan countries (Gomanee et al., 2005) in a panel-data framework. My research will provide more insight on whether aid giving to developing countries is effective on growth and add to the literature on the effect of aid on growth. The paper is organized as follows: Section II provides a brief review of the relationship between foreign aid and economic growth in the literature. In Section III, the estimation procedure is explained. Section IV describes the data, Section V discusses the results, and Section VI concludes. 3
II. LITERATURE REVIEW The relationship between foreign aid and economic growth is often describes by the twogap model of Chenery and Strout (1966). The first gap is the financing gap where investment necessary for growth is related to domestic saving and the second gap is the relationship between imports and foreign exchange. Whenever a country receives aid, the aid fills that gap. Easterly (2003) focuses on the financing gap because foreign aid is believed to increase investment which in turn increases economic growth. When testing 88 aid recipient countries for the years 1965 to 1995, he finds that only six countries had increases in investment and four countries had a positive relationship between growth and investment. In sum, the relationship between aid and growth is unclear. Many researchers such as Rajan and Subramanian (2005), Lansink and White (2001), Gong and Zou (2001), and Knack (2001) argue that aid affects growth negatively. The negative effects are manifested through numerous channels. Rajan and Subramanian (2005) suggest that aid has detrimental long-term effects on economic growth of developing countries through a decrease in employment. In a study of 40 countries for the period 1980-1989 and 28 countries for 1990-1999, they examine whether labor-intensive industries have a relatively slower growth rate in countries with high aid inflows by controlling for within-country effects. Controlling for differentials in industries within countries, Rajan and Subramanian (2005) assume that aid giving is driven by trade and exchange rate policies (10). They find that aid inflows undermine the competitiveness of a country s exporting sectors because aid causes an overvalued exchange rate that has 4
adverse consequences on the growth of traded goods sector in the recipient country (Rajan and Subramanian, 2005: 5). Reduced competitiveness in the traded good sector decreases employment growth. Decrease in employment growth can trigger the country s future dependency on aid (Rajan and Subramanian, 2005: 7). As a result, there will be an increase in the level of aid that the country receives. In a cross-country analysis, Knack (2001) indicates that high levels of aid erode governance quality through increases in corruption and rent-seeking that cause political instability and the weakening of the quality of the state bureaucracies in the recipient countries (313). In relation to the dependency of countries on more foreign aid, Lansink and White (2001) find that with increases in aid flows, high levels of aid may signify, or induce aid dependence, rather than lay the basis for self-reliant development as aid is intended to (43). Because of that dependence, Lansink and White (2001) believe that an aid Laffer curve may exist when high aid inflows negatively affect a country s productivity (49). A Laffer curve is one that illustrates an increase of total benefits up to a maximum level of aid. Any aid level that is below that maximum will increase growth and any aid level above that maximum will decrease growth. The idea of the Laffer curve demonstrates that there is a limit to the amount of aid a country can receive. Therefore, in their study of 138 countries from Sub-Saharan Africa, Latin America, and Asia for the sub-periods 1975-1979, 1980-1984, 1985-1989 and 1990-1992, the quadratic term of aid is included to test whether the Laffer curve exists. Their results confirm the existence of the aid Laffer curve. 5
The relationship of aid on growth depends on the governments capacity to effectively manage the aid they receive to promote the development of their economies. According to Gong and Zou (2001) an increase in the amount of foreign aid a country receives reduces capital accumulation and labor supply and increases consumption in the long-run. Using an optimal growth model with foreign aid, foreign borrowing, and endogenous-leisure-and-consumption choices, Gong and Zou (2001) find that an increase in aid raises income. An income increase causes consumption to increase and investment to decrease. This is a channel through which aid negatively affects the growth process through dissipation of increase income in consumption. On the other hand, there are several mechanisms through which aid can positively contribute to economic growth. Aid increases investment and the ability to import goods and technology (Morrissey, 2001: 41). The ability to invest and increase goods availability and technology in a country will in turn affect the growth rate. In a study of 77 developing countries over the 1971-1990 period, Fayissa and El-Kaissy (1999) show that foreign aid positively affects economic growth. Similarly, Asteriou (2009) finds a positive relationship of aid on economic growth for five South Asian countries for the years 1975 to 2002 when they control for short-run dynamics and country-specific effects. Gomaneee et al. (2005) analyze the relationship of aid and economic growth for 25 Sub-Saharan countries over the period 1970 to 1997 and find that there is a significant positive effect of aid on economic growth. In the case of the Sub-Saharan countries, investment serves a transmission mechanism through which greater economic performance is achieved. 6
In addition, some researchers argue that the positive effect of aid on growth is conditional on some economic factors. Burnside and Dollar (2000), Collier and Dollar (2002), and Easterly et al. (2004) show that aid can positively affect economic growth if the recipient country possesses good fiscal, monetary and trade policies but has little impact where these economic policies are poor. Burnside and Dollar (2000) use data from 56 countries for six four-year time periods from 1970-1973 until 1990-1993. In their model, aid has a positive impact on growth when it is interacted with good policy. In other words, aid accelerates growth when poor countries have sound macroeconomic policies, whereas aid is dissipated in unproductive government expenditure in highly distorted economies (Burnside and Dollar, 2000: 847, 864). In order to maximize the effectiveness of aid in developing countries, aid should be allocated to countries with good policies and great amounts of poverty (Collier and Dollar, 2002: 1482). Easterly et al. (2004) also assess the relationship between aid, policy and economic growth using a larger sample than that of Burnside and Dollar (2000) by extending the years to 1997 instead of 1993 and adding additional countries for a total of 62 countries instead of 56. They find that when expanding Burnside and Dollar s data set, the results are not robust and the question of aid effectiveness is still inconclusive. The debate over the effect of foreign aid on economic growth is on-going and requires further study. 7
III. ECONOMETRIC MODEL The effect of aid on economic growth is still ambiguous in the literature. Therefore, we construct an econometric model to test aid s effect on economic growth for the year 2000. To study determinants of economic growth, Barro and Sala-i-Martin (1999) use an empirical framework that relates growth rate of GDP per capita to two different sets of variables. The first set is that of initial level of state variables such as the stock of physical capital and the stock of human capital in the forms of educational attainment and health. The second set is control or environmental variables such as the ratio of government consumption to GDP, the ratio of domestic investment to GDP, movements in the terms of trade, etc (421). Following Barro and Sala-i-Martin ((1999) and other previous research on foreign aid (Burnside and Dollar, 2000, Daalgard et al., 2004, and Gomanee et al., 2005), we investigate the effect of foreign aid on economic growth using a linear model: g = β 0 + β 1 X + β 2 P + β 3 A + β 4 A 2 + ε, (1) where g is the growth rate of real GDP per capita; vector X is a vector of variables that include the logarithm of GDP per capita, population growth, investment, and human capital; vector P is a vector of macroeconomic policies that might affect economic growth such as trade, government consumption and inflation; A is foreign aid; and A 2 is the square term of foreign aid. The first independent variable is the logarithm of initial GDP. Logarithm of initial GDP accounts for the effects of convergence between countries (Burnside and Dollar, 2000, Daalgard et al., 2004, Easterly et al., 2004, and Lansink and White, 2001). The 8
logarithm of initial GDP per capita corrects for the difference in the growth rates of countries that are operating below their steady-state levels (Weil, 2009: 66-67). Another independent variable is population growth rate which is used to represent the effect of population on income per capita. The quantity of capital available in a country is fixed. Thus, an increase in the population growth would result in less capital being accessible for each worker (Weil, 2009: 95). This relationship between population growth and capital per worker affects economic growth and is included in the model for this reason. The ratio of gross capital formation to GDP is included to capture the effect of the growth rate of capital stock on the growth rate of GDP per capita. Another factor that influences growth is investment in human capital. Human capital can be seen as the key input to the research sector, which generates the new products or ideas that underlie technological progress (Barro, 1991: 408-409). Therefore we include education in the form of secondary school enrollment in the model. The variable of interest in the model, aid, is the official development aid (ODA). Aid is measured as the net flow of resources from one country to the other. ODA, specifically, comprises of direct grants and concessional loans with the grant component being above 25% (Lancaster, 2007). The disbursement of foreign aid can either be integrated into physical capital formation or into human capital investment depending on whether the aid is capital or knowledge intensive. Moreover, the quadratic term of the ratio of aid to GDP is included to take account of the diminishing marginal returns characteristic of aid. If the Laffer curve assumption for aid holds, we expect the coefficient for aid squared to be negative. Lastly, two variables of policies are incorporated in the model. They are government consumption expenditures, inflation 9
rate, and trade as the sum of exports and imports. As Knack (2001) mentions, high level of aid can increase government spending. Table 1 summarizes the effects of these determinants in growth literature. 10
Table 1: Summary of Predicted Effects of Determinants on Growth in Literature Determinants Effect on Growth Source Initial GDP per Capita Negative Burnside and Dollar, 2000; Daalgard et al., 2004; Daalgard and Hansen, 2001; Easterly et al., 2004; Lansink and White, 2001 Positive Collier and Dollar, 2002; Gomanee et al., 2005 Aid Positive Burnside and Dollar, 2000; Asteriou, 2009; Burke and Ahmadi-Esfahini, 2006; Lansink and White, 2001; Daalgard et al., 2004; Daalgard and Hansen, 2001; Gomanee et al., 2005 Negative Collier and Dollar, 2000 Aid Squared Negative Daalgard et al., 2004; Gomanee et al., 2005; Collier and Dollar, 2002; Lansink and White, 2001; Daalgard and Hansen, 2001 Investment Positive Barro and Sala-i-Martin (1999); Burke and Ahmadi- Esfahni, 2006; Gomanee et al., 2005 School Negative Lansink and White, 2001 Positive Baro and Sala-i-Martin (1999); Gomanee et al., 2005 Population Growth Negative Weil, 2009 Trade Positive Barro and Sala-i-Martin, 1999; Daalgard et al., 2004; Lansink and White, 2001; Gomanee et al., 2005 Negative Collier and Dollar, 2000 Inflation Negative Daalgard et al., 2004; Collier and Dollar, 2000; Lansink and White, 2001; Gomanee et al., 2005 Government Consumption Negative Barro and Sala-i-Martin, 1999; Gomanee et al., 2005 11
IV. DATA All data used in our estimation procedure is from the World Bank s World Development Indicators (WDI) database. WDI assists in measuring the progress of development. From the database, we chose 79 countries from the Latin America Caribbean Region, Africa, and Asia for the year 2000. The list of countries is based on the availability of data for each variable, and the full list is available in the Appendix. For the school variable, we had seven countries with missing observations. We opted to substitute the missing values with the value for the last year before 2000 for which the countries had available data for secondary school enrollment. Table 2 offers the descriptive statistics for the dependent and independent variables. Before proceeding to the results of the regressions, we test for model misspecification. To detect omitted variable and incorrect functional form of the model, we use the Regression Specification Error Test (RESET) diagnostic test (Hill et al., 2008). We augmented our regression model with the squared term of the predicted value to create an artificial model. The model is g = β 0 + β 1 X + β 2 P + β 3 A + β 4 A 2 + γ 1 ĝ 2 + ε, (2) where ĝ is the predicted value obtained from the results of equation one. A test of misspecification is a test where H 0 : γ 1 = 0 H 1 : γ 1 0 Rejection of the null hypothesis implies that the original model is inadequate and can be improved. A failure to reject the null hypothesis implies that the test has not been able to 12
Table 2: Descriptive Statistics of Dependent and Independent Variable Variable Measurement Mean Std Deviation Min Max Growth Annual % Growth Rate of GDP per 1.012 4.059-16.550 9.237 capita LnGDP Logarithm of per capita GDP Initial Income 7.006 1.266 4.442 9.711 POP Annual growth rate Population Growth 1.827 0.940-0.3344 4.683 SCHOOL % gross Secondary School 52.635 29.017 6.068 113.104 Enrollment INVEST % of GDP Gross Capital Formation 21.891 8.346 3.454 50.509 Aid % of GDP Official Development 5.590 7.330-0.0188 37.265 Assistance Aid 2 % of GDP Official Development 84.297 199.381 0 1388.697 Assistance Squared TRADE % of GDP Sum of imports and 81.847 42.202 21.719 220.407 Exports GVTCONSUMP % of GDP General Government 14.241 5.993 4.571 37.494 final Consumption INFLATION Annual % Inflation, consumer 14.974 58.622-0.936 513.906 prices Sample Size: 79 13
detect any misspecification. Because we fail to reject the RESET null hypothesis, the test was not able to detect any misspecification in the model. The result for the RESET test is included in the Appendix. We test the model for homogeneity of variance with the Park Test for homoskedasticity. We squared the residuals and regressed it on every independent variable. If the coefficient from the regression is significant, homoskedasdicity cannot be assumed. The insignificance of the coefficients for the Park Test allows us to assume homoskedasticity. According to Hill et al. (2008), when a variable and its square are included in the same model, problems of collinearity might occur. To verify the assumption of noncollinearity, we carried out a series of auxiliary regressions. The auxiliary regression includes one of the independent variables as the dependent variable. The new dependent variable, in turn, is regressed on the other explanatory variables. If any of the resultant coefficient of determination, R 2, is greater than 0.9, the implication is that a large portion of the variation in the predictor is explained by the variation in the other explanatory variables. Then, there may be a collinearity problem (Lomax, 2007). We have found that the only variable with a coefficient of determination greater than 0.9 is the foreign aid variable (refer to table 5). To ensure the accuracy of the auxiliary regressions results, we also examined the variance inflator factor (VIF) of each explanatory variable. The largest VIF should be less than 10 in order to satisfy the assumption of noncollinearity (Lomax, 2007). We also discovered that the only variable with a VIF greater than 10 is the foreign aid variable. To correct for the violation of the noncollinearity assumption, we removed the quadratic term of aid. After the removal, the largest VIF for the new regression is less 14
than 10. Therefore, we proceed with the results of that second regression without the quadratic term of aid. We include the results of the main regression for comparison and to know whether the Laffer curve effect is manifested. 15
V. RESULTS The results of the regression analysis in Tables 3a and 3b are consistent with previous studies in economic growth literature. The coefficient estimates from both tables are very similar with the exception of the aid variable. It is worth noting that the coefficient of the quadratic term of aid is not negative, 0.014 (0.005). Unlike the results of Lansink and White (2001), the positive coefficient means that our results does not confirm the existence of a Laffer curve for foreign aid. The subsequent interpretation is based on the second regression results from Table 3b. Just as Barro and Sala-i-Martin (1999) concluded, initial income per capita is negative, -1.054 (0.695). The negative coefficient denotes conditional convergence among developing countries. Conditional convergence means that holding the other explanatory variables constant, higher growth for countries with a lower starting GDP per capita is predicted. We have found that the estimated coefficient for the trade variable is positive, 0.010 (0.011). A one-percent increase in the sum of exports and imports will increase growth by 0.01 percentage point for the year 2000. Although the sign of the trade variable is consistent with some of the existing literature on the effect of trade as a determinant of growth, it is not statistically significant. Our results indicate that a one-percent increase in consumer prices will decrease economic growth by 0.019 percentage point, -0.019 (0.008). The variable is significant at the 5% level and its sign is consistent with the finding of previous studies. In addition, the estimated coefficient of government consumption expenditure is negative, -0.186 (0.074). 16
The result means that a one-percent increase in government consumption expenditure is associated with a fall in the growth rate of GDP per capita by 0.186 percentage points. Just as Knack (2001) suggested, the negative coefficient of government consumption might be a channel through which aid is diverted to. The coefficient for school is positive but insignificant, 0.037 (0.029). An increase in the percentage of gross enrollment in secondary school will increase GDP per capita growth by 0.037 percentage point. The coefficient of investment is also positive, 0.153 (0.058) and significant at the 5% level. For a percentage increase in gross capital formation, the growth rate of GDP per capita increases by 0.153 percentage point. Although the coefficient of the population growth rate has the expected sign, -0.517 (0.539), we have found it to be statistically insignificant. Moreover, the coefficient for foreign aid is negative, -0.064 (0.078). For an increase in the percentage of foreign aid a country receives, the growth rate of GDP per capita decreases by 0.064 percentage point. A negative coefficient implies that aid has an adverse effect on the growth rate of an economy. It might also mean that low growth rates are influencing the effectiveness of aid instead. This result is consistent with Jensen and Paldam s (2006) claim that aid has an insignificant effect on growth because aid might have other goals besides enhancing the growth of an economy. 17
Table 3a: Least Squares Estimation Results: Main Regression Coefficient Standard t-statistic Error Constant 11.055** 4.570 2.419 LnGDP -1.694** 0.711-2.384 Trade 0.014 0.011 1.233 INFLATION -0.021*** 0.007-2.925 GOVTCONSUMP -0.141 0.073-1.928 SCHOOL 0.035 0.028 1.231 INVEST 0.165*** 0.056 2.931 POP -0.444 0.519-0.856 Aid -0.498*** 0.182-2.741 Aid 2 0.014** 0.005 2.617 R-squared 0.383 F-statistic 4.763 F-probability 0.000 Countries 79 *** significant at the 1% level; ** significant at the 5% level Table 3b: Least Squares Estimation Results: Second Regression Coefficient Standard t-statistic Error Constant 6.491 4.397 1.476 LnGDP -1.054 0.695-1.518 Trade 0.010 0.011 0.912 INFLATION -0.019** 0.008-2.574 GOVTCONSUMP -0.186** 0.074-2.520 SCHOOL 0.037 0.029 1.264 INVEST 0.153** 0.058 2.628 POP -0.517 0.539-0.960 Aid -0.064 0.078-0.829 R-squared 0.322 F-statistic 4.155 F-probability 0.000 Countries 79 *** significant at the 1% level; ** significant at the 5% level 18
VI. CONCLUSION In this paper, we investigated the effect of foreign aid on economic growth for developing countries. When we controlled for other factors, foreign aid has an insignificant negative effect on economic growth. Our results indicated that the Laffer curve effect was not present in our analysis. The conclusion is not new but it is additional evidence to the negative relationship of aid and growth in the literature. We have also found that there is conditional convergence among developing countries. The results are consistent with previous studies, such that trade, education, and investment positively affect growth and inflation while government consumption and the population growth rate negatively affect growth. Our conclusion of the negative relationship between foreign aid and economic growth supports Alesina and Dollar s (2000) claim that foreign aid has not been successful at promoting growth and reducing poverty because of poor performance of bureaucracies in the receiving countries. Similarly, the negative effect of aid on economic growth can be associated with the likelihood of an increase in government consumption as a country receives additional aid. As Knack (2001) argues, high levels of aid erode governance quality through increases in corruption and rent-seeking, which in turn, increase political instability. This negative relationship causes us to think more carefully about how aid giving can be made more effective. Just as the interest to improve living conditions through economic growth in developing countries has become an important topic in foreign policy, donors need to take into account the negative relationship 19
between foreign aid and economic growth when making decisions to provide assistance to other countries. This field still requires further research. 20
APPENDIX Table 4: RESET Test Results Coefficient t-statistic p-value Constant 11.488 2.522 0.014 LnGDP -1.779-2.508 0.015 Trade 0.014 1.257 0.213 INFLATION -0.012-1.141 0.258 GOVTCONSUMP -0.139-1.582 0.118 SCHOOL 0.033 1.197 0.235 INVEST 0.180 3.153 0.002 POP -0.472-0.914 0.364 Aid -0.535-2.927 0.005 Aid 2 0.016 2.814 0.006 PREDICTEDGROWTH 2-0.059-1.341 0.184 21
Table 5: R 2 from Auxiliary Regressions and VIF Predictors R 2 VIF with Aid 2 VIF after removing Aid 2 lngdp 0.801 5.496 4.845 Trade 0.253 1.453 1.435 INFLATION 0.226 1.226 1.214 GOVTCONSUMP 0.339 1.297 1.224 SCHOOL 0.697 4.515 4.510 INVEST 0.410 1.493 1.484 POP 0.331 1.613 1.608 Aid 0.924 12.042 2.025 Aid 2 0.884 8.102 22
Table 6: Countries Used in the Analysis Argentina Algeria* South Africa Bahamas Bangladesh Sri Lanka* Barbados Botswana Sudan Belize Burundi* Swaziland Bolivia Cameroon Tanzania* Brazil Central African Republic* Togo Chile Congo, Dem. Rep.* Tunisia Columbia Congo, Rep. Uganda Costa Rica Gambia Zambia Dominica Ghana Zimbabwe Dominican Republic Guinea-Bissau Cambodia Ecuador India Mongolia El Salvador Indonesia Benin Grenada Iran Cote d Ivoire Guatemala Fiji Ethiopia Guyana Jordan Gabon Haiti Kenya Senegal Honduras Lesotho Jamaica Malawi Mexico Malaysia Nicaragua Mali Panama Mauritania Paraguay Mauritius Peru Mozambique St. Kitts and Nevis Nepal St. Lucia Niger St. Vincent and the Grenadines Pakistan Suriname Papua New Guinea* Trinidad and Tobago Philippines Uruguay Seychelles Venezuela Solomon Islands * denotes countries with missing value for school for the year 2000 23
REFERENCES Alesina, Alberto, and David Dollar. 2000. Who Gives Foreign Aid to Whom and Why? Journal of Economic Growth, 5 (1): 33-63. Asteriou, Dimitrios. 2009. Foreign Aid and Economic Growth: New Evidence From a Panel Data Approach from Five South Asian Countries. Journal of Policy Modeling, 31(1): 155-161. Barro, Robert J. 1991. Economic Growth in a Cross Section of Countries. The Quarterly Journal of Economics, 106 (2): 407-443. Barro, Robert J., and Xavier Sala-i-Martin. 1999. Economic Growth. The MIT Press: Cambridge. Burke, Paul J., and Fredoun Z. Ahmadi-Esfahani. 2006. Aid and Growth: A Study of South Asia. Journal of Asian Economics, 17 (2): 350-362. Burnside, Craig, and David Dollar. 2000. Aid, Policies, and Growth. The American Economic Review, 90 (4): 847-868. Chenery, Hollis B., and Alan M. Strout. 1966. Foreign Assistance and Economic Development. American Economic Review, 56 (4): 679-733. Collier, Paul, and David Dollar. 2002. Aid Allocation and Poverty Reduction. European Economic Review, 46 (8): 1475-1500. Daalgard, Carl Johan, and Henrik Hansen. 2001. On Aid, Growth and Good Policies. Journal of Development Studies, 37 (6): 17-41. Daalgard, Carl-Johan, Henrik Hansen, and Finn Tarp. 2004. On the Empirics of Foreign Aid and Growth. The Economic Journal, 114 (June): F191-F216. 24
Easterly, William. 2003. Can Foreign Aid Buy Growth? Journal of Economic Perspectives, 17 (3): 23-48. Easterly, William, Ross Levine, and David Roodman. 2004. Aid, Policies, and Growth: Comment. The American Economic Review, 94 (3): 774-780. Fayissa, Bichaka, and Mohammed I. El-Kaissy. 1999. Foreign Aid and the Growth of Developing Countries (LDCs): Further Evidence. Studies in Comparative International Development, 34 (3): 37-50. Gomanee, Karuna, Sourafel Girma, and Oliver Morissey. 2005. Aid and Growth in Sub- Saharan Africa: Accounting for Transmission Mechanisms. Journal of International Development, 17 (8): 1055-1075. Gong, Liutang, and Heng-fu Zou. 2001. Foreign Aid Reduces Labor Supply and Capital Accumulation. Review of Development Economics, 5 (1): 105-118. Hill, Carter R., William E. Griffiths, and Guay C. Lim. 2008. Principles of Econometrics. Wiley: Hoboken. Jensen, Peter Sandholt, and Martin Paldam. 2006. Can the Two New Aid-Growth Models Be Replicated? Public Choice, 127 (1-2): 147-175. Knack, Stephen. 2001. Aid Dependence and the Quality of Governance: Cross-Country Empirical Tests. Southern Economic Journal, 68 (2): 310-329. Lancaster, Carol. 2007. Foreign Aid: Diplomacy, Development, Domestic Politics. The University of Chicago Press: Chicago. Lansink, Robert, and Howard White. 2001. Are There Negative Returns to Aid? Journal of Development Studies, 37 (6): 42-65. 25
Lomax, Richard G. 2007. Statistical Concept: A Second Course. Lawrence Erlbaum Associates: Mahwah. Morrissey, Oliver. 2001. Does Aid Increase Growth? Progress in Development Studies, 1 (1): 37-50. Rajan, Raghuran G. and Arvind Subramanian. 2005. What Undermines Aid s Impact on Growth? International Monetary Fund, Working Paper 05/126. United Nations. 2009. Millennium Development Goals. United Nations. Weil, David N. 2009. Economic Growth. Pearson Education: Boston World Bank. 2009. World Development Indicators. The World Bank Group. 26