The Value Chain of Foreign Aid pp 165237  Cite as
The utilization of ODA
Chapter
Abstract
Having analyzed the supply side of development finance (Chap. 2) as well as the process of allocating official funds (Chap. 3) and the manifold implications associated therewith, Chap. 4 finally addresses the “demand side” of ODA and the potential channels of aid affecting poverty reduction in a recipient country. The enquiries here are fourfold:

How is ODA actually disbursed and utilized in lowincome countries (Sect. 4.1)?

Does foreign aid lead to poverty reduction via economic growth (Sect.4.2)?

What are the interdependencies between ODA, poverty reduction, inequality and economic growth (Sect. 4.3)?

What are the macroeconomic consequences of ODA in recipient countries, especially in cases of sudden, large increases of ODA (Sect. 4.4)?
Keywords
Poverty Line Poverty Reduction Recipient Country Poverty Trap Debt Relief
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Preview
Unable to display preview. Download preview PDF.
References
 453.See Harrod (1939) and Domar (1946).Google Scholar
 455.For a critical review of the HarrodDomar model, see Frenkel and Hemmer (1999), pp. 9–25.Google Scholar
 456.See Gylfason (1998), p. 42.Google Scholar
 457.Easterly (1999), p. 2 expands that Domars intention was to discuss shortterm recessions and investment in the United States and not to model longrun economic growth. Empirical studies that have examined the link between private investment and growth usually confirm that there is a positive relationship. For public investment, the results vary. On an aggregate level, the evidence on the impact of investment is mixed. At least, the predictions made by Lewis (1954, 1976, 1978), Rostow (1960) and other development economists in the 1950s, 1960s and 1970s that a higher investment ratio will lead to higher growth have not been fulfilled. Moreover, the composition of public investment matters more than the quantity of public investment itself. See Devarajan et al. (2002), p. 11. Their crosscountry study on the relationship between public investment, private investment and growth by shows no positive correlation for a sample of African countries.Google Scholar
 458.See Gillis et al. (1983), p. 122.Google Scholar
 459.See Chenery and Strout (1966) and Rostow (1960). Rostow was also economic advisor under the Presidents Eisenhower, Kennedy and Johnson and pledged for increased foreign aid to developing countries for political reasons.Google Scholar
 460.See also Diwan (1968), p. 529.Google Scholar
 461.Easterly (1997), p. 8.Google Scholar
 462.For support of this argumentation, see for example Rahman (1968) and Weisskopf (1972).Google Scholar
 463.See Fischer (1991), Fischer (1993), Easterly (1993).Google Scholar
 464.See Barro (1991) and Barro and SalaiMartin (1995).Google Scholar
 465.See Durberry et al. (1998).Google Scholar
 466.Hjertholm et al. (1998), p. 4.Google Scholar
 467.Ibid., p. 4. Also, a certain capitaloutput ratio, in analogy to the HD model, is taken as a datum.Google Scholar
 468.Again, a separation can be made between the ex ante trade gap and the ex post trade gap. An empirical examination on the role of foreign exchange for economic development is provided by Manne (1963). He finds that a small increase of foreign exchange would increase Mexico’s growth rate by 2.5 percentage points.Google Scholar
 469.See Nissanke and Ferrarini (2001), p. 6.Google Scholar
 470.See Chenery and Strout (1966), McKinnon (1964).Google Scholar
 471.Chenery and Strout (1966), p. 689.Google Scholar
 472.For an analysis of the elasticity of substitution between domestic and foreign inputs in twogap models, with special focus on Argentina, see Michalopoulos (1975). The author concludes that at least for some countries twogap models provide only a very limited basis for projecting foreign exchange needs.Google Scholar
 473.See Bacha (1984), p. 264 and Bacha (1990), p. 286–288.Google Scholar
 474.See Bacha (1990) for an illustration. See also Taylor (1990, 1991 and 1994) for a discussion of a threegap model and Iqbal (1995) for an application to Pakistan.Google Scholar
 475.World Bank economists developed a Minimum Standard Model (MSM) when Chenery became chief economic advisor to the World Bank. The model also had two revisions: The Revised Minimum Standard Model (RMSM) and the Revised Minimum Standard Model Extended (RMSMX). In these models the savings, foreign exchange and fiscal gap is calculated. Easterly (1997), p. 13 cites Jos Verbeek, the World Bank’s RMSMX coordinator, who estimates that 90% of all World Bank economists use some kind of RMSM model to make projections.Google Scholar
 476.See Gottschalk (2000).Google Scholar
 478.See Bruton (1969), p. 440 and 445. Providing an economy with foreign aid may lead to the situation where original distortion (why foreign aid was used for in the first place) is not eliminated but upheld.Google Scholar
 479.See Thorbecke (2000), p. 27.Google Scholar
 480.See Easterly (1997). Easterly’s article recalls the “history” of the HD and the gap model approach and shows how merely dogmatic, historical and political factors favored its continuous application.Google Scholar
 481.See ibid., p. 5.Google Scholar
 482.See Ranaweera (2003). See also there for an exact description of the RMSM variables and computing procedure.Google Scholar
 483.See Nissanke and Ferrarini (2001), p. 4. The authors note that Easterly’s criticism has been widely recognized and that it could be applied to any other theoretical model, “when models are applied mechanically for practical purposes without due calibrations.”Google Scholar
 484.See Hussain (2001), p. 91.Google Scholar
 485.See Hussain (1997).Google Scholar
 486.See Thirlwall (1979) and Thirlwall and Hussain (1982).Google Scholar
 487.For an extensive review of Thirlwall’s Law, see for example Dutt (2002) and Perraton (2003). A timeseries test of Thirlwall’s Law for Brazil is provided by Bértola et al. (2002) who confirm the existence of a longrun relationship between Brazilian GDP, terms of trade, and world income. Thirlwall’s law acts as a basis of a large body of postKeynesian growth models.Google Scholar
 488.See Hussain (2001), p. 95. A formal description of the model is also provided there.Google Scholar
 489.See in particular Krugman (1989).Google Scholar
 490.Hussain adopts the projections of the UN Economic Commission for Africa, in which a target growth rate of 7% annually on average is set. See ECA (1999).Google Scholar
 492.Hjertholm et al. (1998), p. 3.Google Scholar
 494.See Hjertholm (1998), pp. 29–34 for details.Google Scholar
 495.For an extensive overview of these issues see Hjertholm et al. (1998), p. 7–28.Google Scholar
 496.Barro and SalaiMartin (1995), p. 49.Google Scholar
 497.Nelson (1956), p. 894.Google Scholar
 498.For a complete review of the model, see Nelson (1956), pp. 894ff.Google Scholar
 499.See ibid., pp. 897–898.Google Scholar
 500.Although it seems that this “population trap” is valid in most of the poor African countries, it cannot be generalized. For an analysis of the historical evolution of the relationship between population and growth, see Galor and Weil (2000).Google Scholar
 501.See Barro and SalaiMartin (1995), p. 49.Google Scholar
 502.For further details, see Barro and SalaiMartin (1995), Chap. 1.Google Scholar
 504.See Bloom et al. (2003) and Quah (1997).Google Scholar
 505.Furtado (1961), p. 67.Google Scholar
 507.Thomas et al. (2002), p. 52.Google Scholar
 508.RosensteinRodan (1943).Google Scholar
 509.Leibenstein (1957).Google Scholar
 510.See RosensteinRodan (1943, 1961, 1984). See also Nelson (1956) and Leibenstein (1957).Google Scholar
 511.See Murphy et al. (1989). Further works examine the big push under the label of “threshold externalities”. See Azariadis and Drazen (1990) as well as Ghosh and Wolf (1990). Bowman and Anderson (1963) stress the importance of a minimum literacy rate for rapid economic growth.Google Scholar
 512.See Murphy et al. (1989).Google Scholar
 513.See ibid. A concept arising from this is the theory of balanced growth. See for example Nurkse (1953). A contrasting view is taken by Hirschman (1958) who advocated unbalanced growth, pointing to the forward and backward linkages in production.Google Scholar
 514.See Murphy et al. (1989).Google Scholar
 515.See also Shams (1999), p. 2. As RosensteinRodan (1961), p. 63 remarks, free international trade reduces the size of the minimum push required.Google Scholar
 516.Sauer et al. (2003).Google Scholar
 517.See Vellutini (2002), p. 453.Google Scholar
 518.See Easterly (2006), p. 37ff.Google Scholar
 519.Ibid., p. 40.Google Scholar
 520.See Kraay and Raddatz (2005).Google Scholar
 521.Kraay and Raddatz (2005), p. 3 and 31.Google Scholar
 522.See Dorn et al. (1998), p. 26.Google Scholar
 523.See Obstfeld and Rogoff (1998).Google Scholar
 524.See Easterly (1999).Google Scholar
 525.See Bhattacharya and Clements (2004), p. 49.Google Scholar
 526.See Nissanke and Ferrarini (2001), p. 6.Google Scholar
 527.Usually, investors will demand higher interests for the higher insolvency risk. Models of the Ponzi debt games class deal with the idea whether governments can “play a game” in which they roll over debt indefinitely in order to never increase taxes. Blanchard and Weil (2001) show that this approach of rolling over debt forever is feasible under certain conditions.Google Scholar
 529.Krugman (1988), p. 2. One of the first contributions on this topic is Sachs (1984, 1986).Google Scholar
 530.More analytically, “a country has a debt overhang problem when the expected present value of potential future resource transfers is less than its debt.” Krugman (1988), p. 5.Google Scholar
 531.See Greene and Villanueva (1991) and Savvides (1992).Google Scholar
 532.See Clements et al. (2003), p. 3–4.Google Scholar
 533.See Chowdhury (2001), p. 4.Google Scholar
 534.Such a concern was already raised by Krugman (1988), p. 14 who pointed out that there is no reason for a country to adjust, because the reward goes solely to its creditors. For the creditors, in turn, this means that they should give less incentives for policy adjustment than necessary in order to receive at least a part of their debt.Google Scholar
 535.Cohen (1993).Google Scholar
 537.See Clements et al. (2003) for a review of empirical studies. Estimations on the threshold level (with respect to GDP) vary widely.Google Scholar
 538.See Clements et al. (2003), p. 18.Google Scholar
 539.See Krugman (1988), p. 10 for this argument. For example, a rise in commodity prices for certain goods might enable resourcerich countries to be able to repay their outstanding debt because of larger export earnings. Another important point to consider is the fact that although the creditors taken as one group have a collective interest in prolonging their loans, each individual creditor’s best strategy is to opt out and let the other creditors extend their loans.Google Scholar
 540.See Griffin (1970), Griffin and Enos (1970) and Papanek (1973).Google Scholar
 541.See Voivodas (1973).Google Scholar
 542.See, for example, Hansen and Tarp (1999) who recall 131 regression studies.Google Scholar
 543.See ibid., p. 6.Google Scholar
 544.See Hansen and Tarp (1999), p. 9.Google Scholar
 545.Ibid, p. 1.Google Scholar
 546.See Mosley (1987).Google Scholar
 547.See Dalgaard et al. (2004), Dalgaard and Hansen (2001) and Moreira (2003).Google Scholar
 548.A noteworthy exemption is Ghosh and Wolf (1998) who leave the path of regression studies and present a decision tree model.Google Scholar
 549.Ravallion (2003a), p. 1 is right in pointing out that all (even small) differences in the methodology used must be carefully considered: “[D]ramatically differing positions taken in this debate [on poverty] often stem from differences in the concepts and definitions used and differences in data sources and measurement assumptions.”Google Scholar
 550.See Collier and Dehn (2001). More generally, Chauvet and Guillaumont (2001) show that in economically instable countries (e.g., fluctuating terms of trade) aid is more effective, because it gives a country the opportunity to deal with economic shocks without changing policy reforms. In contrast, political instability reduces the effectiveness of aid.Google Scholar
 551.See Collier and Hoeffler (2004).Google Scholar
 552.See Dalgaard et al. (2004). Earlier models that analyze the link between economic development and the geography of countries go back to Bloom and Sachs (1998) and Gallup et al. (1999).Google Scholar
 553.See Easterly and Levine (2002).Google Scholar
 554.See Beck et al. (2004).Google Scholar
 555.See Engerman and Sokoloff (2005).Google Scholar
 556.See Dalgaard et al. (2004), p. 192.Google Scholar
 557.See Svensson (1999).Google Scholar
 558.See Guillaumont and Chauvet (2001).Google Scholar
 559.See Islam (2003).Google Scholar
 562.See Ravallion (2003a), p. 4.Google Scholar
 566.See Watts (1968).Google Scholar
 567.See Zheng (1993). Originally, Sen (1976) presented three axioms to be central for poverty measurement. These are the focus axiom, the monotonicity axiom and the transfer axiom. The focus axiom states that the measure is invariant to changes of income of the poor. Monotonicity entails that every reduction of income of the poor must be associated with an increase in poverty. The transfer axiom holds that any transfers reducing the inequality among the poor reduce poverty. Later, the axiom of additive decomposability has been added. It means that aggregate poverty can be written as a population weighted mean of the poverty measures across different groups. Zheng (1993) and other authors have further enhanced this set of axioms.Google Scholar
 568.See Agence Francaise de Développement et al. (2005), p. 81f.Google Scholar
 569.See Ravallion and Chen (2003). See there for an application of this index to China.Google Scholar
 570.See Sen (1997).Google Scholar
 574.Ravallion (2003a), p. 7.Google Scholar
 576.Dagdeviren et al. (2004), p. 126.Google Scholar
 578.See Klasen (2003).Google Scholar
 579.See Kuznets (1955).Google Scholar
 580.See Todaro (1997), p. 163.Google Scholar
 581.Shorrocks and van der Hoeven (2004), p. 1.Google Scholar
 582.See Thornton et al. (1978). See also Anderson (1964).Google Scholar
 583.Thornton et al. (1978), p. 394. In their study, the authors use a definition of poverty with an increasing threshold for the period between 1963 and 1974.Google Scholar
 584.Hirsch (1980) showed that the results of the Thornton et al. model are sensitive to various specifications and that the original specification used by the authors may be inappropriate. See Hirsch (1980), p. 151. Most of the significant results vanish if a relative instead of an absolute poverty measure is used (Hirsch takes a “percentage change in poverty” variable).Google Scholar
 585.See Norton (2002), p. 264.Google Scholar
 586.See Norton (2002), p. 269. He acknowledges, however, that this result can be partly attributed to the difficulties of measuring the true incomes of the poor. Additionally, the simulations are based on a ceteris paribus clause, which is problematic because the incomes of the poor and the rich both grow, as the mere existence of trickleup and trickledown effects shows.Google Scholar
 587.See Aghion and Bolton (1997).Google Scholar
 588.Loayza and Raddatz (2005), p. 18.Google Scholar
 589.See UN (2000), World Bank (2000b).Google Scholar
 590.See Ahluwalia and Chenery (1974). Their work on redistributive mechanisms for poverty reduction remained rather ephemeral with the upcoming neoliberalism in the 1980s.Google Scholar
 591.A theoretical explanation of immiserizing growth was introduced by Bhagwati (1958) and Bhagwati (1988). It goes back to works of Edgeworth (1894) who called it indemnifying growth.Google Scholar
 592.See Ravallion (2004a), p. 6.Google Scholar
 593.See White and Anderson (2001), Kakwani and Pernia (2000).Google Scholar
 594.See White and Anderson (2001).Google Scholar
 595.See Kakwani and Pernia (2000).Google Scholar
 596.Note that this index, applied to the simplest definition of propoor growth (growth is propoor if it reduces poverty) implies a value of greater 0 to represent propoor growth. See Kakwani and Pernia (2000) and also Baulch and McCulloch (2000).Google Scholar
 597.See Kakwani and Son (2002).Google Scholar
 598.See Ravallion (2004a), p. 4. Because the growth incidence curve focuses on percentiles, it is not the same as the mean growth rate in the income of all poor. Son (2004) shows the difference in an example.Google Scholar
 599.See Son (2004).Google Scholar
 600.See Atkinson (1987). Atkinson provided a link between secondorder dominance and poverty reduction. Son (2004) shows that under Atkinson’s theorem, an upward shift of the generalized Lorenz curve results in poverty reduction using a distributional definition of propoor growth. This holds for all poverty lines and all poverty measurements.Google Scholar
 601.See Lopez (2004a), pp. 4–5.Google Scholar
 603.See Klasen (2005), pp. 5–6. He cites the example of schooling. A 20% increase in years of schooling of the poor might be equivalent to 1/4 year of additional schooling, whereas for the rich, a moderate 10% increase might mean one more year. This can hardly be regarded as propoor.Google Scholar
 604.See Kakwani and Pernia (2000), p. 4.Google Scholar
 606.See Ahluwalia and Chenery (1974), p. 209.Google Scholar
 607.Harrod and Domar showed that investment has a dual character. Not only does demand increase via multiplication effects, but also does supply expand, because investment increases the production capacity. According to Harrod and Domar, economic growth depends on the savings rate, the ratio of capital to output (capital coefficient) and the depreciation rate. The HarrodDomar equation \( \frac{{\Delta Y}} {Y} = \frac{s} {k} \) explains that the growth rate of a country (left side of the equation) is proportional to its savings rate and inversely proportionate to the capital coefficient (right side of the equation). Thus, growth can only be positively affected by more investment (= higher s due to the ex post identity of savings and investment) or a lower capitaloutput ratio. Since the latter is assumed to be held constant, it follows: The higher the investment rate (i.e. the accumulation of physical investment via domestic savings), the higher the growth rate of the succeeding period. See Harrod (1939) and Domar (1946).Google Scholar
 608.See Loayza and Raddatz (2005).Google Scholar
 609.See Datt and Ravallion (2002) and Ravallion and Datt (1996).Google Scholar
 610.Kappel et al. (2005), p. 27. For specific literature on the agricultural sector, see Timmer (1988) and Ravallion (1995).Google Scholar
 612.Klasen (2005), p. 19.Google Scholar
 613.Rector and Lauder (1995) argue for such a welfare dependency hypothesis claiming that transfers diminish the incentives of poor people to search for jobs.Google Scholar
 614.See Coady et al. (2002), p. 5ff.Google Scholar
 617.Ravallion (2003b), p. 15.Google Scholar
 618.See Dagdeviren et al. (2004), p. 126.Google Scholar
 620.Dollar and Kraay (2002), Lopez (2004b) and Besley and Burgess (2003) provide evidence.Google Scholar
 621.See Ravallion and Chen (2003).Google Scholar
 622.To cite just two studies: Bibi (2005) for Tunisia and Contreras (2001) for Chile.Google Scholar
 623.See several World Development Reports (e.g., 1990) as well as Ravallion (1995) and Fields (2001).Google Scholar
 624.Dollar and Kraay (2002).Google Scholar
 627.See Kraay (2006), forthcoming. Average incomes are defined as the average annual growth rate over the spell of household average income. Changes in relative incomes are measured using the average annual proportional change in the Gini index as well as the discretetime distribution component of the change in the headcount measure of poverty. Kraay’s results show that growth in average incomes accounts for 70 percent of the variance in the short run, and 97 percent in the long run. Changes in relative incomes account for only 30 percent of the variance of changes in the headcount measure of poverty.Google Scholar
 628.See Agence Francaise de Développement et al. (2005).Google Scholar
 629.See Kuznets (1955).Google Scholar
 630.See Barro (2000).Google Scholar
 632.See Dollar and Kraay (2002).Google Scholar
 633.See Deininger and Squire (1998) and Bruno et al. (1998).Google Scholar
 634.See, for example, Dollar and Kraay (2002) or Ravallion (1997).Google Scholar
 635.See Ravallion (2001).Google Scholar
 636.See Ravallion (2004a), p. 8.Google Scholar
 637.See Myrdal (1984), p. 154.Google Scholar
 638.See Alesina and Rodrik (1994) or Perotti (1996).Google Scholar
 639.See Dasgupta and Ray (1986) as well as Ravallion (1992).Google Scholar
 640.See Strauss (1986).Google Scholar
 641.See Alesina and Rodrik (1994).Google Scholar
 642.See Alesina and Perotti (1996), Demombynes and Özler (2002).Google Scholar
 643.See Kaldor (1956).Google Scholar
 644.See Forbes (2000) or Li and Zou (1998).Google Scholar
 645.See, for example, Persson and Tabellini (1994) and Birdsall et al. (1995).Google Scholar
 646.Klasen (2003).Google Scholar
 647.See Ravallion (1997, 2001) and Son and Kakwani (2004).Google Scholar
 648.See Lopez (2004b), p. 17.Google Scholar
 650.See, for example Klasen (2005).Google Scholar
 651.See Fields (1980), Bourguignon and Morrison (1998) and Ravallion and Datt (1999), p. 9.Google Scholar
 652.The case of Uganda shows that periods of success and failure can follow very rapidly on another. See Kappel et al. (2005).Google Scholar
 653.See, for example, Aiyar et al. (2005) on a sample of five African countries and Andrews et al. (2005) for Ethiopia.Google Scholar
 654.See Okun (1975).Google Scholar
 655.Elbadawi (1999) finds that beyond a certain threshold, more ODA negatively affects the volume of exports (sample of 62 countries).Google Scholar
 656.See Devarajan et al. (2002).Google Scholar
 657.See Heller and Gupta (2002).Google Scholar
 659.World Bank and IMF (2005).Google Scholar
 660.Clemens and Radelet (2003), p. 5.Google Scholar
 661.See Collier and Hoeffler (2004).Google Scholar
 662.See Collier and Dollar (2002) as well as Heller and Gupta (2002).Google Scholar
 663.See Heller (2005), p. 10.Google Scholar
 665.See Moss and Subramanian (2005), pp. 3–4. Data is from 2003 (and 2004 where available).Google Scholar
 666.See Powell (2003), p. 4.Google Scholar
 667.See Ndikumana (2004).Google Scholar
 668.For each extra dollar of debt relief, recipients of debt relief received approx. 9.6 cents of ODA and 23.8 cents of concenssional loans. The results for additionality of debt forgiveness are less significant and depend on the estimation method. See Ndikumana (2004), p. 336.Google Scholar
 669.Ndikumana (2004), p. 337.Google Scholar
 670.See Clemens and Radelet (2003), p. 6.Google Scholar
 671.See Round and Odedokun (2004) and Schweinberger and Lahiri (2006).Google Scholar
 672.See Bhaduri and Skarstein (1996) who assess the impact of foreign aid on domestic demand.Google Scholar
 673.See Collier (1999).Google Scholar
 674.See Aiyar et al. (2005).Google Scholar
 676.See Feyzioglu et al. (1996).Google Scholar
 678.See Aiyar et al (2005), p. 29.Google Scholar
 680.Gupta et al. (2004).Google Scholar
 681.See Meltzer (2000).Google Scholar
 682.See Sanford (2002).Google Scholar
 685.See Mascarenhas and Sandler (2005), p. 1096.Google Scholar
 686.See Clements et al. (2004), p. 48. The study included more than 100 countries and covers 30 years. Their results prove robust for lagged values, twoway causality and endogeneity between aid and revenues.Google Scholar
 687.This term was first used in 1977 by The Economist to describe the gas discoveries on the Netherlands in the 1960s. See also Chowdhury (2004), pp. 5–19 for a detailed survey.Google Scholar
 688.Similar effects were observed for minerals in Australia and oil in the UK and Norway. Also the subSaharan country Gabon is a case where oil discoveries in the 1970s have led to the Dutch Disease effect of shrinking industrial and agricultural sectors because of the appreciation of the currency and capital movements to the oil sector. See Zafar (2004).Google Scholar
 689.See Corden and Neary (1982) and Snape (1977) for theoretical models of the Dutch Disease effect.Google Scholar
 690.See Nkusu (2004), pp. 9–13 for an analytical framework.Google Scholar
 691.Mavrotas (2003), p. 17.Google Scholar
 692.Small countries experiencing a large, temporary inflow of foreign exchange, e.g. foreign aid, are found to be more affected by Dutch Disease effects. See Bhaduri and Skarstein (1996), p. 195.Google Scholar
 693.See Aiyar et al. (2005), p. 31.Google Scholar
 694.See, for example, Younger (1992). See also Adam and Bevan (2003) for Uganda who point to the complex relationship between aid flows, real exchange rates, export volumes and welfare.Google Scholar
 695.See Andrews et al. (2005), p. 35.Google Scholar
 696.See Sackey (2001) and Nyoni (1998). The Ghana case refers to the period 1962–1996, the Tanzania case to 1967–1993.Google Scholar
 697.See Langhammer (2002), p. 13.Google Scholar
 698.Heller (2005), p. 11.Google Scholar
 699.See Alesina and Weder (2002), p. 1135.Google Scholar
 700.See Svensson (2000).Google Scholar
 701.See Bräutigam and Knack (2004), p. 276.Google Scholar
 702.See Knack (2001). The corruption variable is used as part of a set of governance indicators including bureaucratic quality and rule of law.Google Scholar
 703.See Tavares (2003).Google Scholar
 704.See Bulir and Hamann (2003). The authors estimate aid volatility to be 30–60% of the mean.Google Scholar
 705.See World Bank (2005b), p. 106.Google Scholar
 706.See Eifert and Gelb (2005), p. 24.Google Scholar
Copyright information
© PhysicaVerlag Heidelberg 2007