Available online at www.sciencedirect.com
Journal of Policy Modeling 30 (2008) 523–539
Foreign aid, growth and poverty: A policy framework for Niger夽 Emmanuel Pinto Moreira a,∗ , Nihal Bayraktar b a
World Bank, Mail Stop #: J7-703, 1818 H St. NW., Washington, DC 20433, USA b School of Business Administration, Penn State University-Harrisburg, 777 W. Harrisburg Pike, Middletown, PA 17053, USA
Received 31 October 2006; received in revised form 31 January 2007; accepted 28 March 2007 Available online 18 May 2007
Abstract This paper extends the dynamic macroeconomic framework developed by Ag´enor et al. [Ag´enor, P. -R., Bayraktar, N., & Aynaoui, K. E. (2006, July). Roads out of Poverty? Assessing the Links between Aid, Public Capital, Growth, and Poverty Reduction. World Bank, Revised.]. As in the original model, linkages between foreign aid, public investment (education, infrastructure, and health) and growth are explicitly captured, but this time in a fixed nominal exchange rate regime. Although the nominal exchange rate is fixed, the relative price of domestic goods is endogenous, thereby allowing for potential Dutch disease effects associated with increases in aid. The impact of policy shocks on poverty is assessed by using partial growth elasticities. A policy experiment of increasing foreign aid illustrates the dynamic trade-offs between growth and poverty reduction in Niger. © 2007 Society for Policy Modeling. Published by Elsevier Inc. All rights reserved. JEL classification: F35; O11; O2; O4 Keywords: Niger; Foreign aid; Growth; Poverty; Public Investment
1. Introduction An abundant literature has recently been devoted to the economic effects of foreign aid on recipient low-income countries (LICs). It has led to much controversy among development practitioners. On the one hand, Burnside and Dollar (2000) argue that the effectiveness of foreign aid in enhancing growth in a country is conditional to the implementation of good policies. On the other hand, developing economists have particularly challenged the robustness of the 夽 ∗
A detailed version can be downloaded on http://www.personal.psu.edu/nxb23/PintoMoreira-Bayraktar-PSU-WP.pdf. Corresponding author. Tel.: +1 202 458 1834. E-mail address:
[email protected] (E. Pinto Moreira).
0161-8938/$ – see front matter © 2007 Society for Policy Modeling. Published by Elsevier Inc. All rights reserved. doi:10.1016/j.jpolmod.2007.03.002
524
E. Pinto Moreira, N. Bayraktar / Journal of Policy Modeling 30 (2008) 523–539
dependence of the aid-growth link on the policy regime. Guillaumont and Chauvet (2001) and Chauvet and Guillaumont (2004) point out that policies themselves depend on aid, and that the effectiveness of aid depends on the degree of economic vulnerability and domestic instability. Dalgaard and Hansen (2001) argue that foreign aid spurs growth irrespective of the quality of the country’s policies, but with decreasing marginal returns. Easterly, Levine, and Roodman (2003), Easterly (2003), and Roodman (2003) find that the relationship between aid and policies is not significant, and stress that Burnside and Dollar’s finding is not robust to alternative definitions of aid, policies, and long-run growth. The aid literature suffers however from two shortcomings. First, it has mainly relied on crosscountry regressions to assess the impact of aid on growth and has thus failed to analyze the macroeconomic linkages between aid, policy variables and growth in a general quantitative equilibrium framework. Second, this literature has not paid attention to the poverty issue, in particular the mechanisms through which aid can impact poverty. Understanding these effects, and quantifying them, is important for LICs to design their poverty reduction strategies and donors to enhance the quality and depth of the policy debate, with a view to improve their policy advice. This paper extends the dynamic macroeconomic framework developed by Ag´enor, Bayraktar, and El Aynaoui (2006), and applies to Niger, one of the poorest countries in the world, to derive policy implications useful for both policymakers and donors providing aid to a typical LIC. According to the latest household survey available (1989–1993), 63% of the population live below the poverty line, and 34% considered extremely poor in Niger. Social indicators and living conditions are extremely fragile and precarious. The UNDP Human Development Index ranked Niger 177th out of 177 countries in 2005. Niger is heavily dependent on foreign aid, which has been volatile over the past years. As in the original model, key features are the fiscal and supply-side effects of aid, as well as the stock and flow effects of public investment, but this time in a fixed exchange rate regime. Public capital in health and infrastructure has a direct effect on output and the marginal productivity of inputs used in production. Public capital in education also plays an indirect role in the production process, given that “raw” labor must be educated to become productive. At the same time, potential congestion effects associated with the use of public services are taken into account. The domestic (composite) good is an imperfect substitute with the foreign good, and its relative price is endogenous. As a result, the model allows us to analyze potential Dutch disease effects that may be associated with large aid flows in a fixed-exchange rate economy such as Niger (through increases in domestic prices), in both the short and the long run. In addition, the model captures explicitly the link between aid and public investment, and possible adverse effects of large inflows of foreign aid on tax effort. The impact of policy shocks on poverty is assessed by using partial elasticities relating consumption growth to poverty. The model is used to simulate a variety of policies that could be important for helping Niger design and quantify a medium-term strategy aimed at fostering economic growth and reducing poverty. The remainder of the paper is organized as follows. Section 2 describes the model. Section 3 presents parameter estimates and the calibration procedure. Section 4 presents simulation results associated with an increase in foreign aid. The sensitivity analysis is reported in Section 5. Section 6 summarizes the main policy lessons of the analysis. 2. The model As noted earlier, the framework that we develop in this paper for a typical low-income country in a fixed exchange rate regime follows in many essential aspects the model presented in Ag´enor
E. Pinto Moreira, N. Bayraktar / Journal of Policy Modeling 30 (2008) 523–539
525
et al. (2006, July). As in the original formulation, we specify a one-sector, two-good model that accounts for the fiscal and supply-side effects of aid, as well as the supply- and demand-side effects of public capital formation. We, however, amend the original specification to better reflect some of the important characteristics of a low income economy following a fixed nominal exchange rate regime as well as other relevant changes, such as a disaggregation of domestic taxes into direct and indirect taxes. We also drop the distinction between food aid and nonfood aid. 2.1. Production of goods The economy produces one (composite) good, which is an imperfect substitute for an imported good. Net domestic production, Y, requires educated labor, LEP , private capital, KP, and public capital in health and infrastructure, KGhea and KGinf, respectively. KGinf improves the productivity of the private factors used to generate output, because it facilitates the production process. KGhea improves the quality of labor employed in production. In order to account for differences in the degree of substitutability between inputs, we adopt a nested CES production structure. At the lowest level, LEP and KGhea produce the composite input T (“effective” labor): −1/ρT KGhea −ρT −ρT T = AT βT LEP + (1 − βT ) , (1) POPθH where θH ≥ 0 and σT = 1/(1 + ρT) is the elasticity of substitution. KGhea is divided by the size of the population, POP, to account for congestion effects in the provision of health services. At the second level, T is used, together with KP to produce the composite input J: −1/ρJ
J = AJ[βJ T −ρJ + (1 − βJ)KP −ρJ ]
,
(2)
where σJ = 1/(1 + ρJ) is the elasticity of substitution. At the third level, J and KGinf are combined to give net domestic output: ⎡ −ρY ⎤−1/ρY KGinf −1 ⎦ Y = AY ⎣βY J −ρY + (1 − βY ) , (3) θI Y−1 where θI ≥ 0. It is assumed that KGinf becomes productive with a 1 year lag. The lagged value of output, Y−1 , is used here as an indicator of the intensity of use of public services in infrastructure. For a given value of θI, the higher the scale of production, the greater the potential for congestion effects. Domestic output is allocated between exports, X, and domestic sales, DOM, according to a constant elasticity of transformation (CET) function: 1/ρDE
Y = ADE[βDE XρDE + (1 − βDE)DOMρDE ]
,
(4)
where σDE = 1/(ρDE − 1), with 1 < σDE < ∞ measuring the elasticity of transformation between X and DOM. The value of production is given by PY × Y = PD × DOM + PX × X, where PD denotes the price of the domestic good, PX the domestic-currency price of exports, PY the net output deflator. Given these two equations, standard efficiency conditions imply that X/DOM = {(PX/PD)[(1 − βDE)/βDE]}σDE .
526
E. Pinto Moreira, N. Bayraktar / Journal of Policy Modeling 30 (2008) 523–539
2.2. Population and labor supply Population grows at the constant exogenous rate, n such that POP = (1 + n)POP−1 . Similarly, “raw” labor, LR, grows at the same rate as total population. We specify a three-level nested CES structure to show the transformation of raw labor into educated labor. At the first level, KGinf and KGedu produce a composite public capital in education, KGZ, which is defined as: −1/ρK
KGinf −ρK KGedu −ρK KGZ = AK βK + (1 − βK) , (5) Y θKE LRθKI where 0 < θKE < 1, 0 < θKI < 1, and σK = 1/(1 + ρK) is the elasticity of substitution. This equation indicates that the quality of education is not only determined by availability of KGedu but also by KGinf. KGedu is divided by the term LRθKGE in order to capture congestion effects in the education system (see Ag´enor, Bayraktar, Pinto Moreira, and El Aynaoui (2006)). Similarly, the public infrastructure input is divided by YθKGI in order to capture congestion effects related to KGinf. At the second level, the number of teachers on the government’s payroll, χLEG , and KGZ−1 produce a composite public education input, Z, which is defined as: −1/ρZ Z = AZ βZ(χLEG )−ρZ + (1 − βZ)(KGZ−1 )−rZ , (6) where χ is the share of teachers in total educated labor in the public sector and σZ = 1/(1 + ρZ) is the elasticity of substitution. At the third level, the transformation of raw labor into educated labor, LE, takes place through the education system. The “production function” for newly educated workers, LEN , depends on the quantity of raw labor in the economy in the previous period, LR−1 , as well as Z: −1/ρE
LEN = AE[βE(LR−1 )−ρE + (1 − βE)Z−ρE ]
,
(7)
where σE = 1/(1 + ρE) ≥ 0. Given the flow equation above, LE is given by LE = LE−1 + LEN . With LEG denoting the number of educated workers employed in the public education system, the stock of educated labor in production is thus LEP = LE − LEG . We also assume that wages are flexible, so that there is no open unemployment of educated labor. This assumption is consistent with much of the evidence for low-income Sub-Saharan African countries (see, for instance, Dorosh, Nssah, and Samba-Mamadou (1996), and Dorosh and Sahn (2000)). 2.3. Income and private expenditure Income from production accrues entirely to a representative household. Total income before taxes, YTOT, is thus: YTOT = PY Y + WG LEG − RP∗ ER FdebtP−1 + RD DdebtG−1 + ER UTR$,
(8)
where RP* is the interest rate on private foreign borrowing, FdebtP the stock of private foreign debt, DdebtG the stock of domestic public debt, RD the interest rate on that debt, UTR$ the foreign-currency value of private unrequited transfers (assumed exogenous), WG is the average wage paid in the public sector, and ER the nominal exchange rate, which is taken as fixed. Disposable income in nominal terms, Ydisp, can be defined as Ydisp = YTOT − DITAX, where
E. Pinto Moreira, N. Bayraktar / Journal of Policy Modeling 30 (2008) 523–539
527
DITAX denotes direct tax revenue. Total private consumption in real terms, CP, is defined as a constant fraction of disposable income such that CP = (1 − s) Ydisp/PQT, where 0 < s < 1 is the savings rate and PQT is the tax-inclusive composite market price of goods sold domestically. Private investment depends on the rate of growth in net domestic output (to capture an accelerator effect), private foreign capital, ER FP/NGDP, and the stock of public capital in infrastructure, relative to output, to capture complementarity effects1 : PQT IP KGinf PQT ER FP Y , , = IP , (9) NGDP Y−1 −2 NGDP −1 NGDP where IP is real private investment. The stock of private capital evolves over time according to KP = IP−1 + (1 − ␦P)KP−1 , where ␦P is a constant rate of depreciation. Total demand for goods sold on the domestic market, Qd, is given by Qd = (CP + CG) + (IP + IG), where CG and IG denote real current non-interest government spending and investment outlays, respectively. In standard Armington fashion, goods bought and sold on the domestic market, Q, are aggregated through a CES combination of imports, M, and DOM: −1/ρDM
Q = ADM[βDM DOM−ρDM + (1 − βDM)M−ρDM ]
,
(10)
where σDM = 1/(1 + ρDM) is the elasticity of substitution. The spending identity is given by PQ Q = PD DOM + PM M, where PQ is the price of the composite good before taxes, and PM the domestic-currency price of imports. Given these two equations, first-order conditions imply that M/DOM = {(PD/PM)[(1 − βDM)/βDM]}σDM . 2.4. Aid, government budget, and GDP at market prices The government collects taxes, and spends on salaries, goods and services, and interest payments. It also invests in education, health, infrastructure, and other items. It receives foreign aid. The deficit is financed by domestic borrowing and foreign borrowing (concessional). Total government spending in nominal terms, GTOT, is defined as GTOT = WG LEG + PQT(CG + IG) + RG∗ ER FdebtG−1 + RD DdebtG−1 ,
(11)
where WG LEG is the government wage bill, CG current non-interest expenditure on goods and services, IG total public investment, FdebtG the stock of foreign debt, RG* the interest rate on foreign debt, DdebtG the stock of domestic public debt, and RD is the interest rate on that debt. Both RG* and RD are assumed exogenous. The government budget balance, GBAL, is given by GBAL = TAX + AID − GTOT, where TAX is total tax revenue in nominal terms and AID is foreign grants measured in domestic-currency terms. TAX is defined as TAX = DITAX + INDTAX + tm ER PM* M, where DITAX (respectively INDTAX) denotes direct (respectively indirect) taxes, and tm is the tariff rate. AID is defined as AID = ER AID$, where AID$ is the foreign-currency value of foreign grants. The stock of domestic debt is defined as DdebtG = DB + DdebtG−1 , where DB is the flow of domestic borrowing, assumed exogenous. The financing constraint of the government implies that GBAL = DB + ER FG with FG denoting the flow of government borrowing abroad. We assume that the government finances its deficit through foreign borrowing, at the average interest rate RG* . 1
See Ag´enor (2004) for the complementarity between public investment and private capital formation.
528
E. Pinto Moreira, N. Bayraktar / Journal of Policy Modeling 30 (2008) 523–539
Total public investment, IG, is defined as the sum of investment in health, education, infrastructure, and “other” investment spending, IGoth: IG = IGedu + IGhea + IGinf + IGoth. Each component is given as a fixed fraction of total investment such that IGh = κh IG with h = edu, hea, inf, oth, and κh = 1. Direct tax revenue is given by an “effective” tax rate times the tax base such that DITAX = DITXR YTOT, where DITXR is assumed to be constant. Indirect tax revenue is also given by an “effective” tax rate times the tax base, domestic sales, PQ Q. The effective tax rate, INDTXR, depends on its lagged value and, the ratio of aid to GDP in order to capture a possible adverse effect of aid on fiscal effort2 : AID INDTAX = INDTXR INDTXR−1 , PQ Q. NGDP
(12)
It is assumed that CG has two components: CG = NMCG + δh h KGh−1 with h = edu, hea, inf. δh h KGh−1 is government spending on maintenance and NMCG is government spending on other goods and services (constant as a share of GDP). Total public investment, also as a share of GDP, depends positively on the lagged value of the tax ratio and aid as a share of domestic output3 : PQT IG = ig NGDP
TAX NGDP
AID , , NGDP −1
AID NGDP
2 .
(13)
To the extent that the coefficient of the linear term is positive and that of the quadratic term is negative, foreign aid would be positively related to public investment outlays only up to a certain level of aid, and would be negatively related thereafter. This specification allows us to capture the absorptive constraints that a large increase in aid may create in the institutional environment that is typical of many poor countries. Stocks of public capital in education, health, and infrastructure are given by KGh = αh IGh−1 + (1 − δh)KGh−1 ,
h = edu, hea, inf
(14)
where 0 < δh < 1 is a constant depreciation rate and 0 < = αh < = 1 is the efficiency parameter. The efficiency parameter is introduced to capture the possibility that a fraction of the resources invested in investment projects may not have a positive impact on the public capital stock as emphasized by Pritchett (1996) in the context of developing countries in general. In Eq. (14), we follow the linear specification proposed by Arestoff and Hurlin (2005) and relate the stock of public capital in sector h to the flow of investment in h. In the case of “full efficiency”, αh is therefore equal to 1. In the experiments reported below, we will also consider the case where αh < 1. Nominal GDP at market prices is given by NGDP = PY Y + INDTAX + tm PM* ER M, where PY Y = PQ Q + PX X − PM M.
2 See Franco-Rodriguez (2000) for a review of fiscal response models. Gupta, Clements, Pivovasrky, and Tiongson (2003) found that grants tend to have an adverse effect on revenue mobilization. 3 Clements, Bhattacharya, and Nguyen (2003). Nguyen show that public investment is negatively and nonlinearly related to foreign debt service.
E. Pinto Moreira, N. Bayraktar / Journal of Policy Modeling 30 (2008) 523–539
529
2.5. Balance of payments and foreign debt Measured in foreign-currency terms, the balance of payments is given by PX∗ X − PM∗ M − RG∗ FdebtG−1 − RP∗ FdebtP−1 + UTR$ + AID$ +FG + FP − ΔNFA = 0,
(15)
where FP denotes private capital inflows (assumed exogenous) and NFA the change in net foreign assets of the central bank. The foreign-currency value of the stock of private foreign debt, FdebtP, is thus defined as FdebtP = FP + FdebtP−1 , whereas the foreign-currency value of the stock of external public debt, FdebtG, is given by FdebtG = FG + FdebtG−1 . Total external debt, FdebtTot, is thus given by FdebtTot = FdebtP + FdebtG. Given that the nominal exchange rate is fixed, the balance of payments clears through adjustment in official reserves. 2.6. Market equilibrium and prices Market equilibrium requires Q = Qd. The price of the composite good before taxes, PQ, is a CES aggregation of the price of the domestically-produced good and the price of imports: 1−σDM )
PQ = [βDM PD1−σDM + (1 − βDM)PM1−σDM ] 1/(
.
(16)
The tax-inclusive price of the composite good is thus given by PQT = (1 + INDTXR) PQ. PY is determined by a dual CET price function: PY = [βDE PX1+σDE + (1 − βDE)PD1+σDE ]1/(1+σDE ) .
(17)
PD exhibits a disequilibrium price mechanism, adjusting partially towards its equilibrium value, EQPD such that PD = λPD EQPD + (1 − λPD)PD−1 ,
(18)
where EQPD = (PQ Q − PM M)/DOM from the equation for the spending identity (PQ Q) and λPD parameter measures the speed of price adjustment.4 The domestic-currency price of exports, PX, is given by PX = ER PX* , where PX* is the world price of exports (assumed exogenous). The domestic-currency price of imports, PM, is defined as the product of ER and the world price of imports, PM* (assumed exogenous), inclusive of tariffs such that PM = (1 + tm)ER PM* , where 0 < tm < 1 is the tariff rate. 2.7. Poverty analysis As noted earlier, this model is applied to Niger. Since there is no recent household survey for Niger, we therefore follow the methodology adopted by Ag´enor et al. (2006, July), which consists of relating the poverty rate directly to the growth rate of consumption per capita, as derived from the model. We use three partial elasticity values: a “neutral” or central value of −1, a value of −0.5, and a value of −1.5. In addition, we also use the “adjusted” elasticity formula proposed by Ravallion (2004, pp. 12–13). With a Gini coefficient equal to 50.5, this formula gives a partial growth elasticity of −9.3(1 − Gini)3ˆ = −1.13. Values above or below unity allow us to capture the case where growth is or is not distribution neutral. 4
See Buiter (1980) for details about this disequilibrium price mechanism.
530
E. Pinto Moreira, N. Bayraktar / Journal of Policy Modeling 30 (2008) 523–539
3. Parameter estimates and calibration In this section we present econometric estimates of some of the behavioral equations of the model and describe some features of the calibration procedure when the model is applied to Niger. 3.1. Parameter estimates Using annual time series, we estimated the two “fiscal” regressions (Eqs. (12) and (13)). The results for INDTXR indicate that while the lagged value of the dependent variable has a highly significant coefficient, AID/NGDP has a negative and significant effect on INDTXR. The estimated coefficients of the regression equation with IG/NGDP indicate that TAX/NGDP and AID/NGDP have a positive and significant effect on IG/NGDP and the coefficient associated with the squared value of AID/NGDP is negative and also significant. The last equation that we estimated relates to Eq. (9). The result shows that the rate of growth of real output and the ratio of the public capital stock in infrastructure to GDP have a positive effect on IP. Thus, the data provide evidence of a complementarity effect of KGinf on KP. All other parameters were determined either by using shares for the base period, by dwelling on the scant literature for Niger, or (when country-specific data were not available) by using plausible values for low-income developing countries in general–including the estimates compiled by Ag´enor et al. (2006, July) for Ethiopia in a similar setting.5 The elasticities of substitution were kept at relatively low values (σJ = 0.3, σT = 0.3, σY = 0.4, KGZ = 0.3, σZ = 0.2). While the share parameters βT and βY are taken to be 0.85, βJ is equal to 0.6.6 Measures of congestion effects were difficult to estimate, given the lack of information for developing countries in general and Niger in particular. As a result, we used relatively low values to avoid putting undue weight on these parameters (θKE and θKI = 0.9; θH = 0.4; and θI = 0.3). δh was set at 0.035, and δP at 0.06. 3.2. Calibration and baseline solution We calibrated the model for 2004. Data on national accounts, fiscal accounts, balance of payments (based on IMF estimates), and OECD data were combined to produce a consistent set of estimates. Capital stock data (both public and private) were derived relying on the perpetual inventory method. In solving the model, we use the nominal exchange rate as the num´eraire, and keep its value fixed in all the experiments, in accordance with Niger’s current exchange rate regime. Conducting policy experiments with the model requires building a baseline scenario over the period 2005–2015. In this scenario, foreign aid is kept constant in proportion of GDP at the 2004 level (about 10.7%).7 The baseline results are based on the assumption that public investment is “fully efficient”, in the sense that αh = 1 in Eq. (14). Thus, it is assumed that in the following years Niger will continue to implement institutional reforms that will help to improve governance, strengthen management of public resources, and eliminate much of the waste that has all too often characterized capital outlays in the past. 5 6 7
See Pinto Moreira and Bayraktar (2005) for details. See Ag´enor (2005) for detailed information on values of share parameters used in the literature. See Pinto Moreira and Bayraktar (2005) for details on other assumptions.
Table 1 Niger: trend-based projections, 2004–2015 2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
Poverty rate Ravallion’s (2004) adjusted elasticity (Gini = 50.5) Consumption per capita growth elasticity of −0.5 Consumption per capita growth elasticity of−1.0 Consumption per capita growth elasticity of−1.5
66.5 64.6 66.1 67.7
65.9 64.3 65.6 66.9
64.2 63.5 64.0 64.5
61.6 62.4 61.8 61.1
58.8 61.1 59.2 57.3
55.8 59.8 56.7 53.6
53.1 55 54.2 50.1
50.6 57.3 51.9 46.9
48.3 56.1 49.9 44.1
46.3 55.1 48.1 41.7
44.6 54.2 46.5 39.7
43.1 53.4 45.1 37.9
External sector (% of GDP) Current account Trade balance Exports of goods and NFS Imports of goods and NFS Aid, total Other
−6.0 −9.6 15.0 24.6 10.7 −7.0
−2.6 −7.0 14.9 21.9 10.7 −6.3
−1.1 −6.1 15.01 21.92 10.7 −5.6
0.0 −5.6 15.2 20.8 10.7 −5.0
0.8 −5.4 15.2 20.5 10.7 −4.5
1.4 −5.3 15.01 20.4 10.7 −4.0
1.7 −5.4 14.9 20.4 10.7 −3.5
1.9 −5.6 14.8 20.4 10.7 −2
2.1 −5.7 14.7 20.4 10.7 −2.9
2.2 −5.9 14.6 20.5 10.7 −2.6
2.3 −6.0 14.6 20.5 10.7 −2.4
2.4 −6.1 14.6 20.6 10.7 −2.2
Capital account Foreign direct investment Public borrowing Other
6.0 0.5 3.5 2.0
5.6 0.0 3.6 2.0
5.2 0.0 32.2 2.0
5.0 0.0 3.0 2.0
4.9 0.0 2.8 2.0
4.7 0.0 2.6 2.0
4.5 0.0 2.5 2.0
4.4 0.0 2.3 2.0
4.2 0.0 2.2 2.0
4.1 0.0 2.0 2.0
4.0 0.0 1.9 2.0
3.9 0.0 1.8 2.0
Government sector (% of GDP) Total revenue (including grants) Domestic taxes Indirect taxes on imports Foreign aid (grants)
21.2 4.6 5.9 10.7
20.7 4.8 5.3 10.7
20.8 5.0 5.1 10.7
20.8 5.1 5.0 10.7
20.8 5.2 4.9 10.7
20.8 5.3 4.9 10.7
20.8 5.3 4.9 10.7
20.9 5.3 4.9 10.7
20.9 5.4 4.9 10.7
21.0 5.4 4.9 10.7
21.0 5.4 4.9 10.7
21.0 5.4 4.9 10.7
Total expenditure Spending on goods and services (total) Wages and salaries Investment Interest payments
25.1 15.6 3.6 5.3 0.6
25.3 15.7 3.6 54.34 0.6
24.9 15.8 3.6 4.9 0.6
24.8 15.9 3.6 4.7 0.5
24.6 16.0 3.6 4.5 0.5
24.5 16.0 3.6 4.3 0.5
24.3 16.1 3.6 4.2 0.5
24.12 16.1 3.6 4.0 0.5
24.1 16.1 3.6 3.9 0.4
24.0 16.2 3.6 3.8 0.4
23.9 16.2 3.6 3.7 0.4
23.8 16.2 3.6 3.6 0.4
Overall fiscal balance including grants (cash basis)
−3.9
−4.6
−4.2
−4.0
−3.8
−3.6
−3.5
−3.3
−3.2
−3.0
−2.9
−2.8
3.9 3.5 0.4
4.6 3.6 1.0
4.2 32.2 1.0
4.0 3.0 1.0
3.8 2.8 1.0
3.6 2.6 1.0
3.5 2.5 1.0
3.3 2.3 1.0
3.2 2.2 1.0
3.0 2.0 1.0
2.9 1.9 1.0
2.8 1.8 1.0
Total financing Foreign financing Domestic borrowing
531
2005
E. Pinto Moreira, N. Bayraktar / Journal of Policy Modeling 30 (2008) 523–539
2004
532
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
Prices and real exchange rate GDP at factor cost deflator (% change) Composite good price (after indirect taxes, % change) Real exchange rate (% change)
−2.9 0.7 15.2
3.4 3.7 −0.7
4.1 4.1 −1.1
4.5 4.4 −1.4
4.6 4.4 −1.4
4.5 4.3 −1.3
4.2 4.0 30.0
3.8 3.7 −0.7
3.4 3.3 −0.3
3.0 3.0 0.0
2.7 2.8 0.2
2.5 2.6 0.4
Memorandum items Real GDP per capita at factor cost (% change) Real GDP per capita at market prices (% change) Private investment (% of GDP)
2.3 −1.5 8.2
0.6 5.4 8.0
2.1 3.8 7.8
3.2 4.6 7.4
3.8 5.1 7.1
4.1 4.9 6.9
4.1 4.7 6.6
4.0 4.4 6.4
3.9 4.0 6.1
3.6 3.7 5.9
3.4 3.4 5.7
3.1 3.1 5.5
Public investment (% of total public expenditure) Health (% of public investment) Infrastructure (% of public investment) Education (% of public investment) Other (% of public investment)
20.9 17.7 37.4 10.7 34.2
21.4 17.7 37.4 10.7 34.2
19.7 17.7 37.4 10.7 34.2
19.0 17.7 37.4 10.7 34.2
18.3 17.7 37.4 10.7 34.2
17.7 17.7 37.4 10.7 34.2
16.31 17.7 37.4 10.7 34.2
16.7 17.7 37.4 10.7 34.2
16.3 17.7 37.4 10.7 34.2
15.9 17.7 37.4 10.7 34.2
15.6 17.7 37.4 10.7 34.2
15.3 17.7 37.4 10.7 34.2
External debt (% of GDP) Educated labor (in % of population)
65.1 18.3
62.0 19.2
59.4 19.9
59.4 20.8
53.3 21.7
50.5 22.6
48.1 23.5
46.0 24.4
44.2 25.3
42.7 26.1
41.4 26.9
40.4 27.6
Note: The real exchange rate is defined as the growth rate of nominal exchange rate plus the growth rate of the import price index minus the growth rate of composite good price after indirect taxes. The “adjusted” elasticity formula proposed by Ravallion (2004) is −9.3 × (1 − Gini)3 = −1.13 where Gini index is 50.5 for Niger.
E. Pinto Moreira, N. Bayraktar / Journal of Policy Modeling 30 (2008) 523–539
Table 1 (Continued )
E. Pinto Moreira, N. Bayraktar / Journal of Policy Modeling 30 (2008) 523–539
533
The baseline projections for the period 2005–2015 are shown in Table 1. The results indicate that income per capita rises and poverty drops in Niger. In the worst case (a consumption growth elasticity of −0.5), the poverty rate drops by about 9.6 percentage points. Obviously if the current trends were to be maintained, the prospects of reducing poverty would be rather bleak and the MDG of halving poverty by 2015 would prove elusive despite increasing output and relatively high aid-to-GDP ratio. Indeed, even in the best case, it would take 15 years (that is by 2018) to reduce poverty by half in Niger. 4. Policy experiment: an increase in foreign aid Using the scenario shown in Table 1 as our baseline case, we now examine the effects of an increase in the level of foreign aid (which also implies changes in the level of public investment).8 This experiment is in line with Niger’s current PRSP, which emphasizes the need for an increase in foreign aid as a means to finance its development and poverty reduction strategy. It also acknowledges that Niger will still depend on foreign aid over the next decade. Moreover, this experiment assumes that the additional foreign aid would be provided in the form of grants, shifting the mix of foreign aid from “loan-dominated” to “grant-dominated.” This would also be in line with Niger’s debt sustainability requirements. Our experiment consists of a permanent increase in the aid-to-GDP ratio by 5 percentage points. The results are shown in Table 2. They are displayed as absolute differences from the baseline scenario. The policy experiment corresponds to the case of “full efficiency” where αh = 1 in Eq. (14). We investigate the case αh < 1 in Section 5. The direct effect of the increase in aid is on the budget. Higher inflows of aid are associated with a permanent improvement in the fiscal balance. Additional revenue allows the government to increase investment outlays. The increase in public investment raises over time the stock of public capital in infrastructure, which raises private investment. Thus, the rise in public investment “crowds in” private investment through a complementarity effect. The increase in private investment raises the stock of private capital over time; this, combined with the increase in the stock of public capital in infrastructure, increases the marginal productivity of all other production inputs. As a result, the growth rate of output per capita at factor cost reaches 1.8 percentage points by 2015. The increase in private capital accumulation raises the demand for educated labor, because private capital and educated labor are complementary factors in the production process. At the same time, the rise in public investment in education and infrastructure leads to an increase in the stock of capital in education and the “public education input”, and therefore to a higher “output” of educated workers. The increase in the stock of public capital in health raises the efficiency of educated labor in production. The productivity gains associated with the combined effect of improved “effective labor” and increased marginal productivity contribute to higher domestic output, which in turn raises consumption spending and lowers poverty. The growth rates of real disposable income and consumption per capita increase, reaching 2.2 percentage points by 2015. The impact on poverty reduction is quite significant. The poverty rate based on a partial elasticity of −1.5 falls by about additional 10.2 percentage points by 2015, compared to the baseline results. However, with an elasticity of −0.5, the drop is only 5.1 percentage points (see Table 2 and Fig. 1). With 8 Two additional types of policy experiments (a reallocation of public investment toward infrastructure and a reduction in tariffs) are presented in Pinto Moreira and Bayraktar (2005).
534
Table 2 Niger: Five percent increase in aid to GDP ratio, 2005–2015 (absolute deviation from the baseline results given in Table 1) 2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
4.10
−0.69
−1.21
−2.00
−3.01
−4.10
−5.18
−6.16
−7.01
−7.75
−8.37
−8.90
0.0
−0.30
−0.3
−0.89
−1.37
−1.93
−2.52
−3.10
−3.65
−4.16
−4.62
−5.05
0.00
−0.61
−1.07
−7.76−8
−2.68
−3.68
−4.67
−5.60
−6.42
−7.14
−7.76
−8.29
0.00
−0.93
−1.62
−2.65
−3.91
−8.4
−6.48
−7.55
−8.44
−9.15
−9.71
−10.17
0.00 0.00 0.00 0.00 0.00 0.00
5.56 0.09 −1.00 −1.09 5.00 0.49
5.18 −0.39 −1.39 −1.0 5.00 0.8
4.98 −0.69 −1.77 −1.07 5.00 0.68
4.82 −0.94 −2.08 −1.14 5.00 0.76
4.65 −1.16 −2.37 −1.32−1 5.00 0.81
4.47 −1.40 −2.65 −1.25 5.01 0.84
4.26 −1.61 −2.90 −1.29 5.00 0.85
4.04 −1.84 −3.14 −1.30 5.00 0.85
3.82 −2.06 −3.77 −1.31 5.01 0.84
3.60 −2.25 −3.77 −1.32 5.00 0.83
3.39 −2.45 −3.77 −1.31 5.00 0.80
0.00 0.00 0.00 0.00
−1.46 0.00 −1.46 0.00
−1.75 0.00 −1.75 0.00
−1.46−1 0.00 −1.46−1 0.00
−1.56 0.00 −1.56 0.00
−1.52 0.00 −1.52 0.00
−1.75−1 0.00 −1.75−1 0.00
−1.50 0.00 −1.50 0.00
−1.49 0.00 −1.49 0.00
−1.48 0.00 −1.48 0.00
−1.47 0.00 −1.47 0.00
−1.46 0.00 −1.46 0.00
0.00 0.00 0.00 0.00
4.49 −0.25 −0.26 5.00
4.2 −0.34 −0.24 5.00
4.33 −0.41 −0.26 5.00
4.27 −0.45 −0.27 5.00
4.22 −0.48 −0.29 5.00
4.1 −0.50 −0.31−0 5.01
4.19 −0.51 −0.31 5.00
4.18 −0.51 −0.31 5.00
4.19 −0.51 −0.31 5.01
4.18 −0.50 −0.32 5.00
4.19 −0.50 −0.32 5.00
Total expenditure Spending on goods and services Wages and salaries Investment Interest payments
0.00 0.00 0.00 0.00 0.00
3.03 −0.05 −0.17 3.29 −0.04
2.68 −0.05 −0.17 2.97 −0.07
2.702 0.08 −0.17 2.91 −0.09
2.71 0.17 −0.17 2.83 −0.1
2.70 0.23 −0.17 2.6 −0.13
2.70 0.29 −0.17 2.972 −0.14
2.69 0.4 −0.17 2.68 −0.16
2.69 0.38 −0.17 2.65 −0.16
2.70 0.41 −0.18 2.64 −0.17
2.71 0.44 −0.18 2.63 −0.18
2.73 0.47 −0.18 2.62 −0.19
Overall fiscal balance
0.00
1.46
1.75
1.461
1.56
1.52
1.751
1.50
1.49
1.48
1.47
1.46
Poverty rate Ravallion’s (2004) adjusted elasticity (Gini = 50.5) Consumption per capita growth elasticity of −0.5 Consumption per capita growth elasticity of−1.0 Consumption per capita growth elasticity of−1.5 External sector (% of GDP) Current account Trade balance Exports of goods and NFS Imports of goods and NFS Aid, total Other Capital account Private borrowing Public borrowing Other Government sector (% of GDP) Total revenue (including grants) Domestic taxes Indirect taxes on imports Foreign aid (grants)
E. Pinto Moreira, N. Bayraktar / Journal of Policy Modeling 30 (2008) 523–539
2004
−1.52 −1.52 0.00
3.34 2.65
3.24 2.59
3.08 2.49
−2.61
−2.65
−2.59
0.00
0.00
0.46
0.00
6.21
0.24
1.46
0.00
−0.36
−0.16
−0.07
Public investment (% of total public expenditure) Health (% of public investment) Infrastructure (% of public investment) Education (% of public investment) Other (% of public investment)
0.00
9.34
8.86
8.70
0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00
External debt (% of GDP) Educated labor (in % of population)
0.00 0.00
−5.84 0.00
−8.26 0.00
Prices and real exchange rate GDP at factor cost deflator (% change) Composite good price (after indirect taxes, % change) Real exchange rate (% change) Memorandum items Real GDP per capita at factor cost (% change) Real GDP per capita at market prices (% change) Private investment (% of GDP)
0.00 0.00 0.00
−1.46 −1.46 0.00
−1.75 −1.75 0.00
0.00 0.00
3.85 2.98
3.34 2.61
0.00
−2.98
0.00
−1.46−1 −1.46−1 0.00
−1.50 −1.50 0.00
−1.49 −1.49 0.00
−1.48 −1.48 0.00
−1.47 −1.47 0.00
−1.46 −1.46 0.00
2.87 2.33
2.63 2.14
2.36 1.93
2.11 1.73
1.88 1.54
1.68 1.38
−2.49
−2.33
−2.14
−1.73
−1.73
−1.54
−1.38
1.8
1.40
1.69
1.84
1.89
1.88
1.83
1.76
1.87
2.212
2.2
2.39
2.31
2.21
2.12
1.98
0.04
0.09
0.12
0.15
0.18
0.20
0.22
8.54
8.2
8.35
8.29
8.27
8.27
8.28
8.30
0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00
−10.58 0.20
−12.48 0.53
−14.07 0.96
−15.38 1.47
−16.46 2.01
−17.35 2.59
−18.09 3.18
−18.73 3.78
−19.28 4.38
−0.1
−1.75−1 −1.75−1 0.00
E. Pinto Moreira, N. Bayraktar / Journal of Policy Modeling 30 (2008) 523–539
−1.56 −1.56 0.00
Total financing Foreign financing Domestic borrowing
535
536
E. Pinto Moreira, N. Bayraktar / Journal of Policy Modeling 30 (2008) 523–539
Fig. 1. Niger: Five percentage point increase in aid to GDP ratio, 2003–2015 poverty reduction and growth (percentage deviation from baseline).
aid increasing by 5 percentage points, reducing poverty in half can be achieved only with a high partial elasticity. To reduce poverty in half, say, between 2005 and 2015, the simulation suggests that external aid in proportion to GDP should be increased permanently by 10.3 percentage points if the elasticity is −0.5, bringing the aid level to 21% of GDP. Clearly, in this scenario, absorption problems are bound to happen, making such increases in aid unfeasible in the very short run. After the initial appreciation of real exchange rate, it depreciates as a result of a reduction in domestic prices throughout the simulation period. This declining trend continues throughout the period as a result of a supply response dominating the rise in aggregate demand associated with higher government spending and private expenditure. A key feature of this simulation is that the potential Dutch Disease effect generally associated with an increase in aid materializes only in the short run. As discussed at length by Ag´enor et al. (2006, July), an increase in aid has both supply- and demand-side effects, which imply that the net effect of a change in aid has, in general, an ambiguous effect on the real exchange rate. The demand side effect of aid is stronger initially and the real exchange rate appreciates with rising domestic prices. But then the supply-side effects of the increase in public and private capital formation become large enough to offset the adverse effect of the rise in aggregate demand on prices; thus, prices come down over time. The net effect of aid in the long run is a reduction in domestic prices, a real depreciation, and a rise in exports in real terms. 5. Sensitivity analysis The analysis in this section briefly consider the case where inefficiency in public investment persists, or that reforms aimed at improving governance and eliminate mismanagement of public resources are not sufficiently deep. Because we do not have specific estimates of the parameter αh for Niger, we chose a value of 0.5 in Eq. (14), which is consistent with the estimates of Pritchett (1996) and Arestoff and Hurlin (2005). The new baseline scenario and the results of the aid-shock experiment are reported in Table 3. The key implication is that, in the absence of appropriate reforms aimed at improving the man-
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
Trend-based projections Poverty rate Ravallion’s (2004) adjusted elasticity (Gini = 50.5) Consumption per capita growth elasticity of −0.5 Consumption per capita growth elasticity of−1.0 Consumption per capita growth elasticity of−1.5
66.5 64.6 66.1 67.7
65.9 64.3 65.6 66.9
64.5 63.7 64.3 65.0
62.7 62.9 62.7 62.5
60.7 62.0 61.0 59.9
58.8 61.1 59.3 57.4
57.1 60.4 57.7 55.1
554.25 59.6 56.4 53.2
54.2 59.0 55.1 51.5
53.0 58.4 54.1 50.0
52.0 57.9 53.2 48.7
51.2 57.5 52.4 47.7
Growth rate Real GDP per capita at factor cost (% change) Real GDP per capita at market prices (% change)
2.3 −1.5
0.6 5.1
1.8 2.9
2.4 3.2
2.6 3.3
2.6 3.0
2.5 2.7
2.4 2.5
2.2 2.2
2.0 1.9
1.8 1.7
1.6 1.5
−3.2 −1.5 −2.9 −4.2
−4.0 −1.9 −3.6 −5.1
−4.7 −2.3 −4.2 −6.0
−5.5 −2.7 −4.9 −6.8
−6.1 −3.0 −5.6 −7.6
−6.8 −3.4 −6.2 −8.4
−7.4 −3.8 −6.8 −9.0
0.8 1.2
0.9 1.4
1.1 1.4
1.2 1.4
1.2 1.4
1.2 1.4
1.2 1.4
Five percentage point increase in aid to GDP ratio (absolute deviation from the baseline results given above) Poverty rate Ravallion’s (2004) adjusted elasticity (Gini = 50.5) 0.0 −0.7 −1.2 −1.8 −2.5 Consumption per capita growth elasticity of −0.5 −0.50 −3 −0.5 −0.8 −1.1 Consumption per capita growth elasticity of−1.0 0.0 −0.6 −1.1 −0.6 −2.2 Consumption per capita growth elasticity of−1.5 0.0 −0.9 −1.6 −2.4 −3.2 Growth rate Real GDP per capita at factor cost (% change) Real GDP per capita at market prices (% change)
0.0 0.0
0.0 6.2
0.0 0.1
0.2 0.8
0.5 1.41
E. Pinto Moreira, N. Bayraktar / Journal of Policy Modeling 30 (2008) 523–539
Table 3 Niger: Lower efficiency of public investment, 2004–2015
537
538
E. Pinto Moreira, N. Bayraktar / Journal of Policy Modeling 30 (2008) 523–539
agement of capital investment, progress toward reducing poverty will be distracted. When the baseline results in Tables 1 and 3 are compared, the growth rate of real GDP per capita at factor cost drops to 1.6% in the absence of full efficiency. As a result, poverty is expected to be higher. Even in the high elasticity case, the poverty rate falls only from 66.9 in 2005 to 47.7 (instead of 37.9). To halve the poverty rate relative to 2003, it would take now 25 years in the high elasticity case, and 69 years in the low elasticity case. A 5-percentage point increase in the aid-to-GDP ratio would lead to an increase in the growth rate of real GDP per capita only 1.23 percentage points by 2015 relative to the baseline (instead of 1.57). As a result, increase in the aid-to-GDP ratio would improve the poverty rate only by 3.8 percentage points by 2015 relative to the baseline (instead of 5.2) when the elasticity is −0.5. To achieve the poverty MDG in 2015, the aid-to-GDP ratio would need to increase to 26.6% in the lowest elasticity case. These alternative experiments point out the fact that without adequate reforms to strengthen the management of public resources and eliminate waste, the benefits that an increase in foreign aid can bring in terms of reducing poverty can be significantly hampered. In that sense, the efficiency of public fund management must improve to make aid effective in Niger. 6. Policy implications and conclusions This paper captures the links between foreign aid, the level and composition of public investment, growth, and poverty reduction in a fixed exchange rate regime. The model is dynamic and is therefore particularly useful to examine the potential dynamic trade-offs that adjustment policies may entail–such as between the short-run impact of higher public spending on education and infrastructure and the long-run effects on the productivity of labor and private capital, and thus on growth and poverty. The model is applied to Niger, one of the poorest countries in the world, to derive policy implications useful for both policymakers and donors. Policy simulations show that a 5 percentage point increase in the aid-to-GDP ratio would not allow Niger to halve poverty between 2005 and 2015 (even in the case of full efficiency of public investment) unless the growth elasticity of poverty is −1.5. Sustaining strong growth over a long period of time to reduce poverty considerably would require a significant increase in foreign aid. From a policy perspective, it is worth stressing that the magnitude of these inflows may have destabilizing macroeconomic effects such as inflationary pressures, risk of excessive appreciation of the real exchange rate, and loss of competitiveness. Thus, policymakers in Niger may face a trade-off between seeking to reduce widespread poverty rapidly through huge inflows of aid, and coping with their potential short-term side effects. The challenge would then be to weigh the short-term losses against the longer-term benefits of aid inflows in terms of strong and sustained growth with a poverty reducing impact. The model also shows the critical role that the degree of public investment plays in the dynamic effects of an increase in aid on growth and poverty. Thus, another policy implication is related to the benefits (in terms of increased efficiency of public investment) associated with policy actions to improve governance and the institutional framework. Policymakers in a typical LIC such as Niger should be aware that without improving the efficiency of management of public funds, the benefits of an increase in foreign aid may only be partial. It should be indicated that the analysis in this paper can be extended to consider not only the poverty target of the Millennium Development Goals (MDGs) but also other indicators of human development such as malnutrition prevalence, infant mortality rate, and life expectancy. Ag´enor et al. (2006b) focus on this issue and present an integrated macroeconomic approach to monitoring progress toward achieving the (MDGs) in Sub-Saharan Africa.
E. Pinto Moreira, N. Bayraktar / Journal of Policy Modeling 30 (2008) 523–539
539
Acknowledgements We are very grateful to Pierre-Richard Ag´enor for his assistance in developing the analytical structure presented in this paper, and to Samba Mamadou for helping us collect the data used in this study. References Ag´enor, P.-R. (2004). The Economics of Adjustment and Growth. Cambridge, Mass: Harvard University Press. Ag´enor, P -R. (2005, January). Infrastructure, public education and growth with congestion costs (Working Paper No. 47). Centre for Growth and Business Cycle Research, University of Manchester. Ag´enor, P. -R., Bayraktar, N., & El Aynaoui, K. (2006, July). Roads out of Poverty? Assessing the Links between Aid, Public Capital, Growth, and Poverty Reduction. World Bank, Revised. Ag´enor, P. -R., Bayraktar, N., Pinto Moreira, E., & El Aynaoui, K. (2006, October). Achieving the millennium development goals in Sub-Saharan Africa: A macroeconomic monitoring framework. The World Economy, 29, 1519–1547. Arestoff, F., & Hurlin, C. (2005, March). The productivity of public capital in developing countries. Unpublished, University of Orl´eans. Buiter, W. (1980). The macroeconomics of Dr. Pangloss: A critical survey of the new classical macroeconomics. The Economic Journal, 90, 34–50. Burnside, C., & Dollar, D. (2000). Aid, policies, and growth. American Economic Review, 90, 847–868. Chauvet, L., & Guillaumont, P. (2004). Aid and growth revisited: Policy, economic vulnerability and political instability. In B. Tungodden, N. Stern, & I. Kolstad (Eds.), Towards pro-poor policies—aid, institutions and globalization. World Bank/Oxford University Press. Clements, B., Bhattacharya, R., & Nguyen, T. Q. (2003). External debt, public investment, and growth in low-income countries (Working Paper No. 03/249). International Monetary Fund. Dalgaard, C.-J., & Hansen, H. (2001). On aid, growth and good policies. Journal of Development Studies, 37, 17–41. Dorosh, P. A., Nssah, B. E., & Samba-Mamadou, O. (1996). Terms of trade and the real exchange rate in the CFA zone: Implications for income distribution in Niger. In E. S. David (Ed.), Economic reform and the poor in Africa. Oxford: Clarendon Press. Dorosh, P. A., & Sahn, D. E. (2000). A general equilibrium analysis of the effect of macroeconomic adjustment on poverty in Africa. Journal of Policy Modeling, 22, 753–776. Easterly, W. (2003). Can foreign aid buy growth? Journal of Economic Perspectives, 17, 23–48. Easterly, W., Levine, R., & Roodman, D. (2003, June). New data, new doubts: Revisiting aid, policies, and growth (Working Paper No. 26). Center for Global Development. Franco-Rodriguez, S. (2000). Recent advances in fiscal response models with an application to Costa Rica. Journal of International Development, 12, 429–442. Guillaumont, P., & Chauvet, L. (2001). Aid and performance: A re-assessment. Journal of Development Studies, 37, 66–92. Gupta, S., Clements, B., Pivovasrky, A., & Tiongson, E. R. (2003, September). Foreign aid and revenue response: Does the composition of aid matter? (Working Paper No. 03/176). International Monetary Fund. Pinto Moreira, E., & Bayraktar, N. (2005). A macroeconomic framework for quantifying growth and poverty reduction strategies in Niger (School of Business Administration Working Paper No. 05-17). Harrisburg: Penn State University. Pritchett, L. (1996). Mind your P’s and Q’s. The cost of public investment is not the value of capital (Policy Research Working Paper No. 1660). World Bank. Ravallion, M. (2004). Pro-poor growth: A primer (Policy Research Working Paper No. 3242). World Bank. Roodman, D. (2003). The anarchy of numbers: Aid, development, and cross-country empirics (Working Paper No. 32). Center for Global Development.