## _wp05100 - References

## Source details

**Canonical URL:** [_wp05100 - References](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp05100.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp05100.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp05100.pdf.json)

---

### I. Introduction and research question
- Research question: Assess whether foreign aid improves human development indicators (HDIs) and helps recipient countries reach selected Millennium Development Goals (MDGs).
- Focus: Two HDIs — infant mortality and education (adult illiteracy).
- Aid measures used:
  - Official bilateral aid (standard measure).
  - Aid projects led by international non-governmental organizations (NGOs).
- Key motivation for NGO measure:
  - NGOs channel a growing share of development assistance and operate at the grassroots level.
  - NGO aid may avoid misallocation and misuse commonly attributed to official bilateral aid.
- Main reported results:
  - NGO aid reduces infant mortality and does so more effectively than official bilateral aid.
  - The impact on illiteracy is less significant.
  - Evidence of a substitution effect between bilateral aid and public social sector expenditures; NGO aid does not affect social spending in the recipient country.

### II. Literature review on aid effectiveness
- Conventional literature primarily measures aid impact on GDP growth, savings, and investment; results are mixed and methodologically disputed.
- Two-gap model (Chenery and Strout, 1966) underpins many aid-growth studies; it views aid as filling savings/export constraints to promote investment and growth.
- Selected empirical findings summarized:
  - Early studies: aid had no significant impact on growth, savings, or investment; aid increased unproductive public consumption (Mosley and others, 1992).
  - Boone (1996) and Reichel (1995): negative relationship between savings and aid (substitution effect).
  - Hadjimichael (1995): relationship negative in most countries, but positive for “good adjusters.”
  - Burnside and Dollar (2000): aid can be effective when policies are good; their results have been criticized as data dependent.
- Three explanatory arguments for disappointing macro results:
  1. Misallocation: donors give aid for strategic reasons to the wrong recipients.
  2. Misuse: recipient governments pursue non-developmental agendas; donor-recipient objectives may conflict.
  3. Wrong outcome measure: GDP growth may not capture poverty-reducing effects of aid; aid could increase consumption or provision of services that reduce poverty without boosting growth.
- Boone (1996) proposed measuring aid impact on human development indicators (infant mortality, primary schooling, life expectancy) rather than GDP.

### III. NGOs in the aid landscape
- Growth and role:
  - NGO participation in foreign aid intensified over the last two decades; number and funding of NGOs have increased substantially.
  - Perceived differentiators: advocacy for vulnerable populations; grassroots operations; private-sector levels of cost control and efficiency (Rose-Ackermann, 1996).
- Critiques and concerns:
  - Donor-state and NGO relationships can compromise NGO work (Edwards and Hulme, 1997).
  - NGOs risk becoming implementers of donors' policies (Robinson, 1997).
  - NGOs may allocate aid influenced by donor funding concerns (Gauri and Fruttero, 2003).
- Empirical gap:
  - Prior to this paper, no macro-level empirical study on NGO aid effectiveness existed; existing evaluations were project-level.

### IV. MDGs, selected HDIs, and explanatory variables
- MDGs: Adopted as part of the Millennium Declaration (Appendix D lists Goals 1–8 and Targets 1–18).
- Selected HDIs: infant mortality and adult illiteracy.
- Rationale:
  - Health and education indicators are more concrete measures than poverty.
  - Infant mortality is a “flash indicator” responding quickly to improved health services (Boone, 1996).
- Controls tested for infant mortality:
  - Per capita GDP; Poverty headcount; Agricultural value added per worker; Female illiteracy; Per capita health expenditure; Governance index (Freedom House policy indicator, ICRG); Degree of urbanization; others following Caldwell (1986), Dreze and Sen (1989), and Gupta and others (2003).

### V. Descriptive comparisons and stylized facts (selected exact statistics)
- Infant mortality descriptive table (means for 87 countries):
  - Mean for 10 countries with highest infant mortality:
    - 135.12          74.95          49.89             265.39 222.27 58.98 28.69 2.61           0.21          30.87 4.70
  - Overall mean:
    - 66.13 38.00 61.50 1693.37 1822.96 39.82 42.81 2.22 0.19 36.63 56.34
  - Mean for 10 countries with lowest infant mortality:
    - 13.02            8.70            71.37            5329.10 7074.57 21.41 66.85 1.47            0.15               55.33 247.05
- Illiteracy descriptive statistics (sample: 76 countries, 1990–2001):
  - Mean for 10 countries with highest illiteracy rates:
    - Illiteracy: 70.03; ICRG: 56.16; Agricultural value added per worker: 295.863; Real GDP per capita: 18.01; Poverty headcount: 48.57; Urbanization: 28.76; Population growth rate: 2.66; NGO aid per capita: 0.29; Bilateral aid per capita: 30.45; Education expenditure per capita: 10.65
  - Overall mean:
    - Illiteracy: 31.06; ICRG: 61.66; Agricultural value added per worker: 2124.01; Real GDP per capita: 1733.10; Poverty headcount: 39.46; Urbanization: 44.34; Population growth rate: 2.16; NGO aid per capita: 0.20; Bilateral aid per capita: 35.73; Education expenditure per capita: 81.96
  - Mean for 10 countries with lowest illiteracy rates:
    - Illiteracy: 4.14; ICRG: 69.27; Agricultural value added per worker: 5063.095; Real GDP per capita: 5182.00; Poverty headcount: 25.33; Urbanization: 65.68; Population growth rate: 1.47; NGO aid per capita: 0.16; Bilateral aid per capita: 51.32; Education expenditure per capita: 258.77
- Allocation patterns (Figures 1a and 1b findings):
  - Infant mortality (20-country sub-sample): Countries with lower infant mortality receive relatively less NGO aid but relatively more bilateral aid; countries with higher infant mortality receive relatively more NGO aid and less-than-the-mean bilateral aid.
  - Illiteracy (20-country sub-sample): NGO aid is allocated relatively more to countries with high illiteracy and relatively less to countries with better HDIs. "No such pattern emerges from bilateral aid allocation."

### VI. Data sources, sample periods, and methodology issues
- NGO aid data: limited to European NGOs projects cofinanced by the European Union (EU); data converted from current euros to constant dollars; sample period 1990–2001.
- Bilateral aid: OECD data.
- Government effort proxies: per capita health spending and per capita education spending (IMF staff estimates and World Bank databases).
- Exogenous variables: female illiteracy, ICRG Index, agricultural value added per worker, real GDP per capita, poverty headcount, population growth rate (sources include World Development Indicators).
- Endogeneity and estimation choices:
  - Potentially endogenous regressors: NGO aid, bilateral aid, social spending.
  - Davidson and MacKinnon test (1993) used to check exogeneity and guide use of instrumental variables / fixed effects vs other estimators.
  - Lags used as instruments where indicated.

### VII. Infant mortality regression — methodology and main results (exact coefficients and p-values preserved)
- Regression specification:
  - ln(HDI)it = β0 + β1 ln(GEpc)it + β2 ln(NGOAid)it + β3 ln(BAid)it + βz' zit + μi + εit
  - HDIj ∈ {Infant Mortality, Illiteracy}; GEit proxied by per capita health or education spending.
- Sample: unbalanced panel of 58 countries, 1990–2001, 233 observations.
- Estimation: one-way error component static panel model; least squares; random effects when Hausman test does not reject; fixed effect when it does.
- Main empirical findings (selected coefficients from Table 3 with p-values in brackets):
  - Health expenditure per capita coefficients:
    - Column I: -0.0527 [0.034]**
    - Column II: -0.0607 [0.021]**
    - Column III: -0.0145 [0.597]
    - Column IV: omitted
  - NGO aid per capita coefficients:
    - Column I: -0.0152 [0.027]**
    - Column II: -0.0126 [0.086]*
    - Column III: -0.0102 [0.147]
    - Column IV: -0.0131 [0.037]**
  - Bilateral aid per capita coefficients:
    - Column I: 0.0063 [0.648]
    - Column II: 0.0086 [0.514]
    - Column III: 0.0064 [0.611]
    - Column IV: 0.0171 [0.160]
  - GDP per capita elasticities:
    - Column III: -0.3083 [0.000]***
    - Column IV: -0.2915 [0.000]***
    - Reported interpretation: If GDP per capita increases by 1 percent, infant mortality decreases by 0.3 percent (reported elasticity).
  - Agricultural value added per worker:
    - Column II: -0.162 [0.001]***
  - Female illiteracy coefficients:
    - I: 0.5272 [0.000]***
    - II: 0.3314 [0.000]***
    - III: 0.3167 [0.000]***
    - IV: 0.336 [0.000]***
  - Poverty headcount reported significant coefficients:
    - 0.2694 [0.001]***; 0.1921 [0.023]**; 0.1696 [0.045]**
  - ICRG governance coefficients where significant:
    - -0.0027 [0.030]**; -0.0022 [0.067]*; -0.0027 [0.012]**
  - Population growth, urbanization, IMF program dummy: insignificant in most specifications.
- Source interpretation:
  - NGO aid per capita reduces infant mortality; bilateral aid per capita shows no significant effect.
  - Possible reasons: NGOs target higher mortality countries and may be more efficient at reaching the poor; bilateral aid appears fungible or diverted, consistent with Boone (1996) concerns.

### VIII. Adult illiteracy regression — methodology and main results (exact coefficients and p-values)
- Sample: unbalanced panel of 76 countries, 1990–2001, 420 observations.
- Estimation: Two-stage least squares (2SLS) / instrumental variable regression; random effect model specification using instrumental variables; instruments include lag of education expenditure and lag of past NGO aid per capita.
- Table 4 selected coefficients with p-values:
  - Education expenditure per capita:
    - I: -0.055 [0.001]***
    - II: -0.031 [0.067]*
    - III: -0.029 [0.117]
    - IV: -0.146 [0.002]***
  - Bilateral aid per capita:
    - I: 0.0456 [0.597]
    - II: 0.077 [0.307]
    - III: 0.072 [0.378]
    - IV: 0.173 [0.384]
  - Real NGO aid per capita:
    - I: 0.001 [0.891]
    - II: 0.004 [0.470]
    - III: 0.000 [0.956]
    - IV: -0.006 [0.658]
  - Urbanization:
    - I: -1.035 [0.000]***
    - II: -1.036 [0.000]***
    - III: -0.898 [0.000]***
    - IV: -0.486 [0.303]
  - Population growth rate:
    - I: 0.175 [0.000]***
    - II: 0.242 [0.000]***
    - III: 0.153 [0.000]***
    - IV: 0.210 [0.001]***
  - IMF dummy:
    - I: -0.018 [0.161]
    - II: -0.002 [0.884]
    - III: -0.020 [0.110]
    - IV: -0.021 [0.280]
  - GDP per capita (column IV only): -0.151 [0.010]***
- Main findings reported:
  - Education expenditure per capita significantly and negatively associated with adult illiteracy.
  - No significant impact of NGO aid per capita and total bilateral aid per capita on adult illiteracy.
  - Urbanization and population growth show expected associations; IMF-supported programs do not appear detrimental to literacy.
  - Governance (ICRG) and average poverty level not significant in these regressions.

### IX. Government effort and aid fungibility — methods and exact results
- Research question: Do foreign aid flows (NGO and bilateral) affect government effort measured as share of health or education expenditure in total public spending?
- Dynamic model specification (as used in source):
  - ln(GEexit) = β0 + β1 ln(GEex_{i,t-1}) + β2 ln(Mil) + β3 ln(BAidex) + β4 ln(NGOex) + β5 z_it + β6 IMF + μi + μt + εit
  - GEex measured as government education/health expenditure as percentage of total government expenditure; NGOAid and BAidex expressed as percentage of expenditure where noted.
- Estimation: Dynamic panel estimated via first-difference and system GMM; preferred estimator GMM-SYS when lagged dependent variable coefficient close to 1; lags used as instruments; diagnostics reported (Hansen J, Arellano-Bond AR(1)/AR(2)).
- Health expenditure share regressions (sample: 50 countries, 1990–2001, 305 observations):
  - Lagged health expenditure coefficients (GMM-SYS):
    - Column 1: 0.958 [0.000]***
    - Column 2: 0.938 [0.000]***
  - Current revenue coefficients:
    - 0.149 [0.003]*** and 0.111 [0.015]**
  - IMF dummy:
    - 0.070 [0.045]** and 0.062 [0.031]**
  - Bilateral aid and NGO aid:
    - Bilateral aid: 0.012 [0.441], 0.019 [0.180] (not significant)
    - NGO aid: -0.006 [0.407], -0.005 [0.458] (not significant)
  - ICRG: system result not significant [0.989]; first-difference shows 0.006 [0.072]* (mild significance)
  - Diagnostics: Hansen test reported as [1.000]; AR(1) and AR(2) p-values reported indicating validity of instruments.
  - Interpretation: Past share strongly predicts current share; no evidence that bilateral aid or NGO aid changes the share of spending on health; IMF programs associated with higher health share; current revenue positively associated.
- Education expenditure share regressions (sample: 51 countries, 1990–2001, 313 observations):
  - Lagged education expenditure coefficients (GMM-SYS):
    - Column 1: 0.836 [0.000]***
    - Column 2: 0.803 [0.000]***
  - Current revenue:
    - 0.184 [0.004]*** and 0.244 [0.000]***
  - IMF dummy:
    - 0.070 [0.048]** and 0.056 [0.064]*
  - Bilateral aid:
    - System results: 0.006 [0.586], 0.009 [0.439] (not significant)
    - First-difference: 0.026 [0.058]* (mild significance; potential bias)
  - NGO aid:
    - System results: -0.009 [0.259], -0.005 [0.490] (not significant)
  - GDP per capita, ICRG: not significant in system results.
  - Interpretation: Lagged education share strongly predicts current share; no robust evidence that NGO aid affects education share; no clear evidence bilateral aid causes substitution away from education.

### X. Conclusion, interpretation, and policy implications (exact phrasings and numeric note)
- Main objective: Reassess effectiveness of foreign aid in terms of its impact on selected human development indicators (infant mortality and illiteracy).
- Measures used: bilateral aid flows to governments and non-governmental aid (NGO aid).
- Key empirical conclusions:
  - NGO aid significantly reduces infant mortality.
  - Bilateral aid does not significantly reduce infant mortality.
  - Explanations offered:
    - NGOs more effective at reaching poor and vulnerable populations; NGOs allocate more to high infant mortality countries while bilateral aid favors countries with lower infant mortality.
    - Bilateral aid appears fungible: increases in bilateral aid don’t seem to be reflected in health expenditures (lack of additionality); Boone’s (1996) diversion-to-elite hypothesis consistent with observed patterns.
  - Illiteracy results less significant:
    - Only government education expenditures shown to reduce illiteracy.
    - Possible explanation: a 10-year period is too short for aid to affect illiteracy; improvements take longer to record.
  - Government effort:
    - NGO aid does not reduce recipient governments’ efforts.
- Policy implications and open questions:
  - If infant mortality is a “flash indicator” of living conditions of the poor (Boone, 1996), NGO aid appears more effective in reaching the poor.
  - Donors who channel aid through NGOs have made an effective choice for reaching vulnerable populations.
  - This paper is the first empirical macro-level study of NGO aid effectiveness and supports the legitimacy of NGO actions.
  - Open question: whether NGO aid can be easily scaled up — recommended for future research.
- Supporting empirical note (statistical indication preserved):
  - "0.878 whereas the within groups estimate is 0.412, suggesting that there is indeed a significant downward bias in the estimates obtained by the first difference estimates."

### XI. Appendices and data documentation (items listed in source)
- Appendix A: Data description and sources for variables including NGO AID (European Commission, budget line B7-6000), Bilateral Aid (OECD), Infant mortality rate (World Development Indicators), Total adult illiteracy rate (World Development Indicators), Female adult illiteracy rate (World Development Indicators), Public education expenditure (World Development Indicators and IMF staff estimates), Public health expenditure (IMF staff estimates), Health expenditure per capita (World Development Indicators and staff estimates), Education expenditure per capita (World Development Indicators and staff estimates), Agriculture value added per worker (constant U.S. 1995 dollars) (World Development Indicators), GDP per capita (World Development Indicators, constant 1995 U.S. dollars), ICRG (International Country Risk Guide/ Political Risk Services), IMF program dummy, Urbanization (World Development Indicators), Population growth rate (World Development Indicators), Military expenditure (World Development Indicators), National poverty (World Development Indicators).
- Appendix B: Level of government relevant for public expenditure data (country-specific listings).
- Appendix C: Country samples for stylized facts and regressions on infant mortality, adult illiteracy, government effort in health, and government effort in education (extensive lists).
- Appendix D: Millennium Development Goals and Targets (Goals 1–8 with Targets 1–18, including Target 1: Halve, between 1990 and 2015, the proportion of people whose income is less than $1 a day; Target 5: Reduce by two-thirds, between 1990 and 2015, the under-five mortality rate; Target 6: Reduce by three-quarters, between 1990 and 2015, the maternal mortality ratio; and other targets).

*Source: _wp05100 - References*

### References..............................................................................................................

### _wp05100 - References

### I. Introduction
- Research question: Assess whether foreign aid improves human development indicators (HDIs) and helps recipient countries reach selected Millennium Development Goals (MDGs).
- Focus: Two HDIs — infant mortality and education (adult illiteracy).
- Aid measures used:
  - Official bilateral aid (standard measure).
  - Aid projects led by international non-governmental organizations (NGOs).
- Key motivation for NGO measure:
  - NGOs channel a growing share of development assistance and operate at the grassroots level.
  - NGO aid may avoid misallocation and misuse commonly attributed to official bilateral aid.
- Main reported results (from the paper):
  - NGO aid reduces infant mortality and does so more effectively than official bilateral aid.
  - The impact on illiteracy is less significant.
  - Evidence of a substitution effect between bilateral aid and public social sector expenditures; NGO aid does not affect social spending in the recipient country.

### II. Literature review on aid effectiveness
- Conventional literature primarily measures aid impact on GDP growth, savings, and investment; results are mixed and methodologically disputed.
- Two-gap model (Chenery and Strout, 1966) underpins many aid-growth studies; it views aid as filling savings/export constraints to promote investment and growth.
- Empirical findings summarized:
  - Early studies: aid had no significant impact on growth, savings, or investment; aid increased unproductive public consumption (Mosley and others, 1992).
  - Boone (1996) and Reichel (1995): negative relationship between savings and aid (substitution effect).
  - Hadjimichael (1995): relationship negative in most countries, but positive for “good adjusters.”
  - Burnside and Dollar (2000): aid can be effective when policies are good; their results have been criticized as data dependent.
- Three explanatory arguments for disappointing macro results:
  1. Misallocation: donors give aid for strategic reasons to the wrong recipients.
  2. Misuse: recipient governments pursue non-developmental agendas; donor-recipient objectives may conflict.
  3. Wrong outcome measure: GDP growth may not capture poverty-reducing effects of aid; aid could increase consumption or provision of services that reduce poverty without boosting growth.
- Boone (1996) proposed measuring aid impact on human development indicators (infant mortality, primary schooling, life expectancy) rather than GDP.

### III. NGOs in the aid landscape
- Growth: NGO participation in foreign aid intensified over the last two decades; number and funding of NGOs have increased substantially.
- Perceived differentiators of NGOs:
  - Advocacy for vulnerable populations; motivation perceived as mainly altruistic.
  - Grassroots operations with private-sector levels of cost control and efficiency (Rose-Ackermann, 1996).
- Critiques and concerns:
  - Donor-state and NGO relationships can compromise NGO work (Edwards and Hulme, 1997).
  - NGOs risk becoming implementers of donors' policies (Robinson, 1997).
  - NGOs may allocate aid influenced by donor funding concerns (Gauri and Fruttero, 2003).
- Empirical gap: No macro-level empirical study had been undertaken on NGO aid effectiveness prior to this paper; existing evaluations were project-level.

### IV. MDGs, selected HDIs, and explanatory variables
- MDGs: Adopted as part of the Millennium Declaration; Appendix D lists all MDGs.
- Selected HDIs for analysis: infant mortality and adult illiteracy.
- Rationale for indicator choice:
  - Health and education indicators are more concrete measures than poverty.
  - Infant mortality is a “flash indicator” responding quickly to improved health services (Boone, 1996).
- Controls and determinants tested for infant mortality:
  - Per capita GDP (level of development).
  - Poverty headcount.
  - Rural development proxy: agricultural value added per worker.
  - Female illiteracy.
  - Government effort: per capita health expenditure.
  - Institutional variables: governance index (Freedom House policy indicator, ICRG).
  - Degree of urbanization.
- The paper follows Caldwell (1986), Dreze and Sen (1989), and Gupta and others (2003) for determinants of HDIs.

### V. Descriptive comparison (Table 1: Infant Mortality)
- Table 1 presents average levels of indicators and aid flows for 87 countries, comparing:
  - Mean for 10 countries with highest infant mortality.
  - Overall mean.
  - Mean for 10 countries with lowest infant mortality.
- Key comparative observations (textual):
  - Higher female illiteracy associated with higher infant mortality.
  - Countries with high infant mortality tend to be less urbanized and have lower rural development (agricultural value added).
  - Poverty headcount higher than average in countries with higher infant mortality.
  - NGO aid per capita higher on average in countries with higher infant mortality; bilateral aid per capita is lower than average in high infant mortality countries and higher than average in low infant mortality countries.
  - Government health expenditure per capita is far lower than average in countries with low levels of infant mortality and much higher than average in countries with low levels of infant mortality.
- Table 1 numeric rows (preserved exactly as in source):
  - Mean for 10 countries with highest infant mortality
    - 135.12          74.95          49.89             265.39 222.27 58.98 28.69 2.61           0.21          30.87 4.70
  - Overall mean
    - 66.13 38.00 61.50 1693.37 1822.96 39.82 42.81 2.22 0.19 36.63 56.34
  - Mean for 10 countries with lowest infant mortality
    - 13.02            8.70            71.37            5329.10 7074.57 21.41 66.85 1.47            0.15               55.33 247.05

### VI. Contributions and empirical approach
- Two main contributions claimed by the paper:
  1. Assessing the impact of foreign aid flows on social indicators (infant mortality and adult illiteracy) rather than on GDP growth.
  2. Providing an empirical evaluation of NGO aid at the macro level, distinguishing NGO aid from bilateral aid.
- Additional analysis: tests whether foreign aid reduces recipient government effort in fighting illiteracy and infant mortality by assessing the impact of aid on the share of social spending.

*Content unit: _wp05100 - References; source PDF: _wp05100 - References..............................................................................................................*

### Appendix C lists the countries included in this sample.

### _wp05100 - Appendix C lists the countries included in this sample.

### Illiteracy: sample and descriptive comparisons
- Sample: 76 countries (1990–2001); refer to Appendix C for the sample.
- Table 2 key statistics (means reported in source):
  - Mean for 10 countries with highest illiteracy rates:
    - Illiteracy: 70.03
    - ICRG: 56.16
    - Agricultural value added per worker: 295.863
    - Real GDP per capita: 18.01
    - Poverty headcount: 48.57
    - Urbanization: 28.76
    - Population growth rate: 2.66
    - NGO aid per capita: 0.29
    - Bilateral aid per capita: 30.45
    - Education expenditure per capita: 10.65
  - Overall mean:
    - Illiteracy: 31.06
    - ICRG: 61.66
    - Agricultural value added per worker: 2124.01
    - Real GDP per capita: 1733.10
    - Poverty headcount: 39.46
    - Urbanization: 44.34
    - Population growth rate: 2.16
    - NGO aid per capita: 0.20
    - Bilateral aid per capita: 35.73
    - Education expenditure per capita: 81.96
  - Mean for 10 countries with lowest illiteracy rates:
    - Illiteracy: 4.14
    - ICRG: 69.27
    - Agricultural value added per worker: 5063.095
    - Real GDP per capita: 5182.00
    - Poverty headcount: 25.33
    - Urbanization: 65.68
    - Population growth rate: 1.47
    - NGO aid per capita: 0.16
    - Bilateral aid per capita: 51.32
    - Education expenditure per capita: 258.77
- Descriptive associations highlighted in source:
  - Higher illiteracy associated positively with higher poverty headcount.
  - Higher illiteracy associated negatively with urbanization and rural development.
  - NGO aid per capita is higher-than-average in countries with high illiteracy; lower-than-average in countries with low illiteracy.
  - Bilateral aid per capita is lower-than-average in countries with high illiteracy and higher in countries with low illiteracy.
  - Education expenditure per capita is far lower-than-average in countries with highest illiteracy and much higher-than-average in countries with low illiteracy.

### Figures: NGO vs bilateral aid allocation patterns
- Figure 1a (20-country sub-sample sorted by infant mortality):
  - Finding: Countries with lower infant mortality receive relatively less NGO aid but relatively more bilateral aid; countries with higher infant mortality receive relatively more NGO aid and less-than-the-mean bilateral aid.
- Figure 1b (20-country sub-sample sorted by illiteracy):
  - Finding: NGO aid is allocated relatively more to countries with high illiteracy and relatively less to countries with better HDIs. "No such pattern emerges from bilateral aid allocation" (source wording).

### Data sources and methodology issues
- NGO aid data: limited to European NGOs projects cofinanced by the European Union (EU); data in current euros converted to constant dollars; sample period 1990–2001.
- Bilateral aid: OECD data.
- Government effort proxies: per capita health spending and per capita education spending (IMF staff estimates and World Bank databases used for expenditures).
- Exogenous variables included: female illiteracy, ICRG Index, agricultural value added per worker, real GDP per capita, poverty headcount, population growth rate (sources include World Development Indicators).
- Endogeneity concerns:
  - Potentially endogenous regressors: NGO aid, bilateral aid, social spending.
  - Test used: Davidson and MacKinnon test (1993) (similar to Durbin-Wu-Hausman) to check exogeneity and to determine use of instrumental variables / fixed effects vs other estimators.
  - Instrument strategy referenced in regressions (lags used as instruments where indicated).

### IV. Infant mortality regression — methodology and main results
- Regression framework (source equation reproduced):
  - ln(HDI)it = β0 + β1 ln(GEpc)it + β2 ln(NGOAid)it + β3 ln(BAid)it + βz' zit + μi + εit
  - HDIj ∈ {Infant Mortality, Illiteracy}; GEit proxied by per capita health or education spending.
- Sample for infant mortality regressions:
  - Unbalanced panel of 58 countries from 1990 to 2001 with 233 observations.
- Estimation choices:
  - One-way error component static panel model; least squares estimation.
  - Random effects used when Hausman test does not reject; fixed effect used when it does.
- Main empirical findings (source language and elasticities reported):
  - Increased health expenditure per capita reduces infant mortality.
    - Table 3 coefficient on Health expenditure per capita:
      - Column I: -0.0527 [0.034]**
      - Column II: -0.0607 [0.021]**
      - Column III: -0.0145 [0.597]
      - Column IV: not applicable (omitted in IV)
  - Greater NGO aid per capita reduces infant mortality.
    - NGO aid per capita coefficients:
      - Column I: -0.0152 [0.027]**
      - Column II: -0.0126 [0.086]*
      - Column III: -0.0102 [0.147]
      - Column IV: -0.0131 [0.037]**
  - No significant impact of total bilateral aid per capita on infant mortality.
    - Bilateral aid per capita coefficients:
      - Column I: 0.0063 [0.648]
      - Column II: 0.0086 [0.514]
      - Column III: 0.0064 [0.611]
      - Column IV: 0.0171 [0.160]
  - Development, rural development, and female illiteracy effects:
    - If GDP per capita increases by 1 percent, infant mortality decreases by 0.3 percent (reported elasticity).
      - GDP per capita coefficients:
        - Column III: -0.3083 [0.000]***
        - Column IV: -0.2915 [0.000]***
    - Agricultural value added per worker sometimes significant:
      - Column II: -0.162 [0.001]***
    - Female illiteracy has a strong positive association with infant mortality:
      - Coefficients: 0.5272 [0.000]*** (I), 0.3314 [0.000]*** (II), 0.3167 [0.000]*** (III), 0.336 [0.000]*** (IV)
  - Poverty headcount increases infant mortality; elasticity ranges from 0.16 to 0.26 in regressions.
    - Poverty headcount coefficients reported where significant: 0.2694 [0.001]***; 0.1921 [0.023]**; 0.1696 [0.045]**
  - ICRG (governance) reduces infant mortality with lower elasticities where significant:
    - ICRG coefficients (columns where shown): -0.0027 [0.030]**; -0.0022 [0.067]*; -0.0027 [0.012]**
  - Population growth rate, urbanization, and IMF program dummy found insignificant for infant mortality in most specifications.
- Interpretation and discussion in source:
  - Potential reasons NGO aid appears effective while bilateral aid does not:
    - NGOs and bilateral donors allocate aid differently (figures show NGOs target higher mortality/illiteracy more).
    - NGOs may have direct links to poor and vulnerable populations and be more efficient.
    - Aid transiting through recipient governments may be diverted to elites (cites Boone (1996) concept).

### V.B. Adult illiteracy regression — methodology and main results
- Sample: unbalanced panel of 76 countries from 1990 to 2001 with 420 observations.
- Estimation method: Two-stage least squares (2SLS) / instrumental variable regression using Davidson MacKinnon tests for exogeneity; specification modeled as a random effect model using instrumental variable regression. Instruments: lag of education expenditure for current education expenditure; lag of past NGO aid per capita as instrument for bilateral aid per capita (source description).
- Table 4 key regression coefficients and p-values (columns I–IV reproduced as in source):
  - Education expenditure per capita:
    - I: -0.055 [0.001]***
    - II: -0.031 [0.067]*
    - III: -0.029 [0.117]
    - IV: -0.146 [0.002]***
  - Bilateral aid per capita:
    - I: 0.0456 [0.597]
    - II: 0.077 [0.307]
    - III: 0.072 [0.378]
    - IV: 0.173 [0.384]
  - Real NGO aid per capita:
    - I: 0.001 [0.891]
    - II: 0.004 [0.470]
    - III: 0.000 [0.956]
    - IV: -0.006 [0.658]
  - Urbanization:
    - I: -1.035 [0.000]***
    - II: -1.036 [0.000]***
    - III: -0.898 [0.000]***
    - IV: -0.486 [0.303]
  - Population growth rate:
    - I: 0.175 [0.000]***
    - II: 0.242 [0.000]***
    - III: 0.153 [0.000]***
    - IV: 0.210 [0.001]***
  - IMF dummy:
    - I: -0.018 [0.161]
    - II: -0.002 [0.884]
    - III: -0.020 [0.110]
    - IV: -0.021 [0.280]
  - GDP per capita (column IV only): -0.151 [0.010]***
  - Constant terms and sample sizes as in source table.
- Main findings reported in source:
  - Significant and negative impact of education expenditure on adult illiteracy.
  - No significant impact of NGO aid per capita and total bilateral aid per capita on adult illiteracy.
  - Urbanization negatively associated with illiteracy (significant in three out of four estimations).
  - Higher population growth associated with higher illiteracy.
  - IMF-supported structural adjustment program does not appear detrimental to literacy (no significant adverse effect).
  - No significant impact of governance level (ICRG) or average poverty level on illiteracy in these regressions.

### VI. Government efforts and foreign aid — methodology and main results
- Research question: How do foreign aid flows (NGO and bilateral) affect government effort measured as share of health or education expenditure in total public spending?
- Model specification used (dynamic panel, source equation reproduced):
  - ln(GEexit) = β0 + β1 ln(GEex_{i,t-1}) + β2 ln(Mil) + β3 ln(BAidex) + β4 ln(NGOex) + β5 z_it + β6 IMF + μi + μt + εit
  - GEex measured as government education/health expenditure as percentage of total government expenditure; NGOAid and BAidex expressed as percentage of expenditure where noted.
- Estimation approach:
  - Dynamic specification estimated via GMM (first difference and system GMM).
  - Preferred estimator: system GMM (GMM-SYS) when coefficient on lagged dependent variable close to 1.
  - Instrument sets: lags of dependent variable (t-2 and earlier) and lags of NGO aid; predetermined variables included.
  - Diagnostics: Hansen J test reported; Arellano-Bond AR(1) and AR(2) tests reported.
- Health expenditure share regressions (Table 5):
  - Sample: unbalanced panel of 50 countries, 1990–2001, 305 observations (base for health regressions: 50 countries; observations vary by specification).
  - Key coefficients (GMM-SYS preferred):
    - Health expenditure (t-1):
      - GMM-SYS column 1: 0.958 [0.000]***
      - GMM-SYS column 2: 0.938 [0.000]***
    - Current revenue:
      - 0.149 [0.003]*** and 0.111 [0.015]**
    - IMF dummy:
      - 0.070 [0.045]** and 0.062 [0.031]**
    - Bilateral aid and NGO aid:
      - Bilateral aid: 0.012 [0.441], 0.019 [0.180] (not significant)
      - NGO aid: -0.006 [0.407], -0.005 [0.458] (not significant)
    - ICRG:
      - System result shows ICRG not significant [0.989]; first-difference shows 0.006 [0.072]* (mild significance)
  - Diagnostics:
    - Hansen test: [1.000] in reported columns (in-source formatting).
    - AR(1) p-values: [0.013], [0.004] etc.; AR(2) p-values indicate no evidence of second-order serial correlation.
  - Interpretation in source:
    - Past share of health spending strongly predicts current share.
    - No evidence that bilateral aid or NGO aid changes the share of spending on health (no additionality for bilateral aid; NGO aid does not negatively affect government health share).
    - IMF-supported programs associated with higher share of expenditure allocated to health; current revenue positively associated.
    - Possible fungibility of bilateral aid (no change in government health share) could help explain lack of bilateral aid effect on infant mortality.
- Education expenditure share regressions (Table 6):
  - Sample: unbalanced panel of 51 countries, 1990–2001, 313 observations (base for education regressions).
  - Key coefficients (GMM-SYS preferred):
    - Education expenditure (t-1):
      - GMM-SYS column 1: 0.836 [0.000]***
      - GMM-SYS column 2: 0.803 [0.000]***
    - Current revenue:
      - 0.184 [0.004]*** and 0.244 [0.000]***
    - IMF dummy:
      - 0.070 [0.048]** and 0.056 [0.064]* (system results)
    - Total bilateral aid:
      - System results: 0.006 [0.586], 0.009 [0.439] (not significant)
      - First-difference shows mildly significant positive 0.026 [0.058]* (but first-difference likely biased downward; source concludes no clear substitution effect)
    - NGO aid:
      - System results: -0.009 [0.259], -0.005 [0.490] (not significant)
    - GDP per capita, ICRG: not significant in system results.
  - Diagnostics:
    - Hansen test: 1.00 reported.
    - Arellano-Bond AR(1) test: 0.00 reported (in-source formatting).
  - Interpretation in source:
    - Lagged education share strongly predicts current share.
    - No robust evidence that NGO aid affects the share of education spending in total expenditure.
    - No clear evidence of bilateral aid causing substitution away from education (system results show no significant effect; difference results mildly positive but potentially biased).
    - IMF presence and current revenue positively associated with education share.

*Source: _wp05100 - Appendix C lists the countries included in this sample.*

### 0.878  whereas  the  within  groups  estimate  is  0.412,  suggesting  that  there  is  indeed  a  significant

### VII. CONCLUSION

### Main objective
- Reassess the effectiveness of foreign aid in terms of its impact on selected human development indicators.
- Focused evaluation on how aid affects infant mortality and illiteracy, which are official objectives of all donors who have adopted the Millennium Development Goals.

### Measures of aid used
- Two separate measures of aid: bilateral aid flows to the governments and non-governmental aid (NGO aid).
- Rationale: official aid and NGO aid flows differ in allocation pattern and implementation.

### Key empirical findings
- NGO aid significantly reduces infant mortality.
- Bilateral aid does not significantly reduce infant mortality.
- Possible explanations for differing effectiveness:
  - NGO aid may be more effective at reaching the poor and operating at the grassroots level, which is important for reducing infant mortality.
  - NGO aid is allocated more toward countries with high infant mortality while bilateral aid favors countries with already lower infant mortality.
  - Bilateral aid seems fungible: increases in bilateral aid don’t seem to be reflected in health expenditures, implying a lack of additionality where bilateral aid increases lead to declines in non-aid-financed expenditures.
  - The measure of official aid used (total bilateral aid per country) may not be the appropriate indicator as it covers all types of projects and programs.19
  - Boone’s (1996) claim that recipient governments divert aid to benefit a wealthy elite may also explain the lack of bilateral aid effectiveness.

- Concerning illiteracy:
  - Results are less significant.
  - Apart from usual control variables that behave as expected, only government education expenditures are shown to reduce illiteracy.
  - A possible explanation: a 10-year period is too short for aid to have an impact on illiteracy, as improvements in illiteracy take much more time to be recorded. Research should be conducted again when longer time series are available.

- On government effort:
  - NGO aid does not reduce recipient governments’ efforts.

### Policy implications and interpretation
- If infant mortality is taken as a “flash indicator” of the living conditions of the poor (Boone, 1996), NGO aid appears more effective in reaching poor and vulnerable populations.
- Donors who channel aid through NGOs have made the right choice in terms of reaching those populations.
- This paper is the first empirical study on the effectiveness of NGO aid at the macro level and confirms the legitimacy of NGO actions.
- Open question: whether NGO aid can be easily scaled up — this should be addressed in future research.

### Supporting empirical note (statistical indication)
- 0.878 whereas the within groups estimate is 0.412, suggesting that there is indeed a significant downward bias in the estimates obtained by the first difference estimates.

### Appendices (data sources and samples referenced)
- Appendix A: Data description and sources for variables including NGO AID (European Commission, budget line B7-6000), Bilateral Aid (OECD), Infant mortality rate (World Development Indicators), Total adult illiteracy rate (World Development Indicators), Female adult illiteracy rate (World Development Indicators), Public education expenditure (World Development Indicators and IMF staff estimates), Public health expenditure (IMF staff estimates), Health expenditure per capita (World Development Indicators and staff estimates), Education expenditure per capita (World Development Indicators and staff estimates), Agriculture value added per worker (constant U.S. 1995 dollars) (World Development Indicators), GDP per capita (World Development Indicators, constant 1995 U.S. dollars), ICRG (International Country Risk Guide/ Political Risk Services), IMF program dummy, Urbanization (World Development Indicators), Population growth rate (World Development Indicators), Military expenditure (World Development Indicators), National poverty (World Development Indicators).
- Appendix B: Level of government relevant for public expenditure data (country-specific listings).
- Appendix C: Country samples for stylized facts and regressions on infant mortality, adult illiteracy, government effort in health, and government effort in education (extensive country lists).
- Appendix D: Millennium Development Goals and Targets (Goals 1–8 with Targets 1–18, including Target 1: Halve, between 1990 and 2015, the proportion of people whose income is less than $1 a day; Target 5: Reduce by two-thirds, between 1990 and 2015, the under-five mortality rate; Target 6: Reduce by three-quarters, between 1990 and 2015, the maternal mortality ratio; and other targets).

*Source: _wp05100 - 0.878  whereas  the  within  groups  estimate  is  0.412,  suggesting  that  there  is  indeed  a  significant*

### REFERENCES

### REFERENCES

### Empirical studies on aid and growth
- Alesina, Alberto, and David Dollar, 2000, “Who Gives Foreign Aid to Whom and Why?” Journal of Economic Growth, Vol. 5 (March), pp. 33–63.
- Burnside, Craig, and David Dollar, 1998, “Aid, the Incentive Regime, and Poverty Reduction,” Policy Research Working Paper 1937 (Washington: World Bank).
- Burnside, Craig, and David Dollar, 2000, “Aid, Policies and Growth,” American Economic Review, Vol. 90 (September), pp. 847–67.
- Burnside, Craig, and David Dollar, 2004, “Aid, Policies and Growth: Revisiting the Evidence,” Policy Development Working Paper 3251 (Washington: World Bank).
- Clemens, Michael A., Steven Radelet, and Rikhil Bhavnani, 2004, “Counting Chickens When They Hatch: the Short-Term Effect of Aid on Growth,” Center for Global Development Working Paper 44 (Washington).
- Collier, Paul, and Jan Dehn, 2001, “Aid, Shocks, and Growth,” Policy Research Working Paper 2688 (Washington: World Bank).
- Collier, Paul, and David Dollar, 2002, “Aid Allocation and Poverty Reduction,” European Economic Review, Vol. 46 (September), pp. 1475–1500.
- Dalgaard, Carl-Johan, Henrik Hansen, and Finn Tarp, 2004, “On the Empirics of Foreign Aid and Growth,” Economic Journal, Vol. 114 (June), pp. F191–F216.
- Dalgaard, Carl-Johan and Henrik Hansen, 2001, “On Aid, Growth and Good Policies,” Journal of Development Studies, Vol. 37 (August), pp. 17–41.
- Easterly, William, 1999, “The Ghost of Financing Gap: Testing the Growth Model Used in the International Financial Institutions,” Journal of Development Economics, Vol. 60 (December), pp. 423–38.
- Easterly, William, 2003, “Can Foreign Aid Buy Growth?” Journal of Economic Perspectives, Vol. 17 (Summer), pp. 23–48.
- Easterly, William, Ross Levine, and David Roodman, 2003, “New Data, New Doubts: A Comment on Burnside and Dollar’s ‘Aid, Policies and Growth 2000’,” American Economic Review.
- Guillaumont, Patrick, and Lisa Chauvet, 2001, “Aid and Performance: A Reassessment,” Journal of Development Studies, Vol. 37 (August) pp. 66–92.
- Hansen, Henrik, and Finn Tarp, 2000, ”Aid Effectiveness Disputed,” Journal of International Development Vol. 12 (April), pp. 375–98.
- Hansen, Henrik, and Finn Tarp, 2001, “Aid and Growth Regressions,” Journal of Development Economics, Vol. 64 (April), pp. 547–570.
- Lensink, Robert, and Howard White, 2000, “Aid Allocation, Poverty Reduction and the Assessing Aid Report,” Journal of International Development, Vol. 12 (April), pp. 399–412.
- Lensink, Robert, and Howard White, 2001, “Are There Negative Returns to Aid?” Journal of Development Studies, Vol. 37 (August) pp. 42–65.
- Mosley, Paul, John Hudson, and Arjan Vershoor, 2004, “Aid, Poverty Reduction and the New Conditionality,” The Economic Journal, Vol. 114 (June), pp. F217–F243.
- Reichel, Richard, 1995, “Development Aid, Savings and Growth in the 1980s: A Cross-Section Analysis,” Savings and Development, Vol. 19, No. 3, pp. 279–96.
- Robinson, 1997, “Privatising the Voluntary Sector: NGOs as Public Sector Contractors,” in NGOs, States and Donors: Too Close for Comfort?, ed. by Michael Edwards and David Hulme (London: Pergamon Press).

### Aid fungibility, public spending, and fiscal behavior
- Feyzioglu, Tarhan, Vinaya Swaroop, and Min Zhu, 1996, “Foreign Aid’s Impact on Public Spending,” Policy Research Working Paper 1610 (Washington: World Bank).
- Feyzioglu, Tarhan, Vinaya Swaroop, and Min Zhu, 1998, “A Panel Data Analysis of the Fungibility of Foreign Aid,” World Bank Economic Review, Vol. 12 (January), pp. 29–58.
- Devarajan, Shantayanan, and Vinaya Swaroop, 1998, “The Implications of Foreign Aid Fungibility for Development Assistance,” Policy Research Working Paper 2022 (Washington: World Bank).
- Pack, Howard, and Janet Rothenberg Pack, 1993, “Foreign Aid and the Question of Fungibility,” Review of Economics and Statistics, Vol. 75 (May), pp. 258–66.
- Petersson, J., 2004, “Foreign Aid Fungibility, Growth, and Poverty Reduction,” Working Paper (unpublished; Stockholm: Stockholm University).
- McGillivray, Mark, and Oliver Morrisey, 2001, “Aid Illusion and Public Sector Fiscal Behaviour,” Journal of Development Studies, Vol. 37 (August) pp. 118–36.
- Gupta, Sanjeev, and others, 2003, “Foreign Aid and Revenue Response: Does the Composition of Aid Matter?” IMF Working Paper 03/176 (Washington: International Monetary Fund).

### Aid allocation, donor motives, and politics
- Boone, P., 1996, “Politics and the Effectiveness of Foreign Aid,” European Economic Review, Vol. 40 (February), pp. 289–329.
- Snyder, Donald W., 1993, “Donor Bias Towards Small Countries: An Overlooked Factor in the Analysis of Foreign Aid and Economic Growth,” Applied Economics, Vol. 25 (April), pp. 481–88.
- Svensson, Jakob, 2000, “Foreign Aid and Rent-Seeking,” Journal of International Economics, Vol. 51 (August), pp. 437–61.
- Tavares José, 2003, “Does Foreign Aid Corrupt?” Economics Letters, Vol. 79 (April), pp. 99–106.
- Murshed, S. Mansoob, and Somnath Sen, 1995, “Aid Conditionality and Military Expenditure Reduction in Developing Countries: Models of Asymmetric Information,” Economic Journal, Vol. 105 (March), pp. 498–509.
- Collier, Paul, and David Dollar, 2002, “Aid Allocation and Poverty Reduction,” European Economic Review, Vol. 46 (September), pp. 1475–1500.
- Alesina, Alberto, and David Dollar, 2000, “Who Gives Foreign Aid to Whom and Why?” Journal of Economic Growth, Vol. 5 (March), pp. 33–63.

### NGOs, voluntary sector, and civil society roles
- Clark, John, 1995, “The State, Popular Participation, and the Voluntary Sector,” World Development, Vol. 23 (April), pp. 593–601.
- Cox, Aidan, and Antonique Koning, 1997, “EC Cooperation With NGO: An Evaluation” (Brussels: European Commission).
- Edwards, Michael, and David Hulme, 1997, “Too Close for Comfort? The Impact of Official Aid on NGOs,” in NGOs, States and Donors: Too Close for Comfort? ed. by Edwards and Hulme (London: Pergamon Press).
- Gauri, Varun, and Anna Fruttero, 2003, “Location Decisions and Nongovernmental Organization Motivation: Evidence from Rural Bangladesh,” Policy Research Working Paper 3176 (Washington: World Bank).
- Meyer, Carrie A., 1995, “Opportunism and NGOs: Entrepreneurship and Green North-South Transfers,” World Development, 23 (August), pp. 1277–89.
- Robinson, 1997, “Privatising the Voluntary Sector: NGOs as Public Sector Contractors,” in NGOs, States and Donors: Too Close for Comfort?, ed. by Michael Edwards and David Hulme (London: Pergamon Press).
- Rose-Ackerman, Susan, 1996, “Altruism, Nonprofits and Economic Theory,” Journal of Economic Literature, Vol. 34 (June), pp. 701–28.
- Edwards, Michael, and David Hulme, 1997, “Too Close for Comfort? The Impact of Official Aid on NGOs,” in NGOs, States and Donors: Too Close for Comfort? ed. by Edwards and Hulme (London: Pergamon Press).
- OECD and UNDP, 1999, Improving the Effectiveness of Aid Systems: The Case of Mali, (Paris: Organization for Economic Cooperation and Development).

### Methodology, econometrics, and data issues
- Alonso-Borrego, Cesar, and Manuel Arellano, 1996, “Symmetrically Normalized Instrumental Variable Estimation Using Panel Data,” CEMFI Working Paper No. 9612 (Madrid: Centro de Estudios Monetarios y Financieros).
- Blundell, Richard, and Stephen Bond, 1999, “GMM Estimation with Persistent Panel Data: An Application to Production Functions,” IFS Working Paper W99/04 (London: Institute for Fiscal Studies).
- Blundell, Richard, and Stephen Bond, 1998, “Initial Conditions and Moment Restrictions in Dynamic Panel Data Models,” Journal of Econometrics, Vol. 87 (August), pp. 115–43.
- Bond, Stephen, 2002, “Dynamic Panel Data Models: A Guide to Micro Data Methods and Practice,” CeMAPP Working Paper CWP09/02 (London: Institute for Fiscal Studies).
- Bond, Stephen, and Frank Windmeijer, 2002, “Finite Sample Inference for GMM Estimators in Linear Panel Data Models,” CeMAPP Working Paper CWP04/02 (London: Institute for Fiscal Studies).
- Davidson, Russell, and James G., MacKinnon, 1993, Estimation and Inference in Econometrics (New York: Oxford University Press).
- Easterly, William, Ross Levine, and David Roodman, 2003, “New Data, New Doubts: A Comment on Burnside and Dollar’s ‘Aid, Policies and Growth 2000’,” American Economic Review.
- Feyzioglu, Tarhan, Vinaya Swaroop, and Min Zhu, 1998, “A Panel Data Analysis of the Fungibility of Foreign Aid,” World Bank Economic Review, Vol. 12 (January), pp. 29–58.
- Tsikata, Tsidi M., 1998, “Aid Effectiveness: A Survey of the Recent Empirical Literature” IMF Paper on Policy Analysis and Assessment 98/1 (Washington: International Monetary Fund).

### Sectoral, demographic, and development-focused studies
- Caldwell, J. C., 1986, “Routes to Low Mortality in Poor Countries, ” Population and Development Review, pp. 171-220.
- Chenery, Hollis B., and Alan M. Strout, 1966, “Foreign Assistance and Economic Development,” American Economic Review, vol. 56 (September), pp. 679–733.
- Dreze, J. and A. Sen (1989). Hunger and Public Action. Oxford, Clarendon Press.
- Hadjimichael, Michael T., and others, 1995, Sub-Saharan Africa: Growth, Saving, and Investment, 1986–93, IMF Occasional Paper No. 118 (Washington: International Monetary Fund).
- Mosley, Paul, John Hudson, and Sara Horrell, 1992, “Aid, The Public Sector and The Market in Less Developed Countries: A Return to The Scene of Crime, “Journal of International Development, Vol. 4 (March-April), pp. 139–50.
- Petersson, J., 2004, “Foreign Aid Fungibility, Growth, and Poverty Reduction,” Working Paper (unpublished; Stockholm: Stockholm University).
- Gauri, Varun, and Anna Fruttero, 2003, “Location Decisions and Nongovernmental Organization Motivation: Evidence from Rural Bangladesh,” Policy Research Working Paper 3176 (Washington: World Bank).

### Conditionality, donor–recipient interactions, and policy implications
- White, Howard, and Oliver Morrissey, 1997, “Conditionality When Donor and Recipient Preferences Vary,” Journal of International Development, Vol. 9 (June), pp. 497–505.
- Devarajan, Shantayanan, and Vinaya Swaroop, 1998, “The Implications of Foreign Aid Fungibility for Development Assistance,” Policy Research Working Paper 2022 (Washington: World Bank).
- Mosley, Paul, John Hudson, and Arjan Vershoor, 2004, “Aid, Poverty Reduction and the New Conditionality,” The Economic Journal, Vol. 114 (June), pp. F217–F243.
- Murshed, S. Mansoob, and Somnath Sen, 1995, “Aid Conditionality and Military Expenditure Reduction in Developing Countries: Models of Asymmetric Information,” Economic Journal, Vol. 105 (March), pp. 498–509.
- Collier, Paul, and Jan Dehn, 2001, “Aid, Shocks, and Growth,” Policy Research Working Paper 2688 (Washington: World Bank).

*Source: _wp05100 - REFERENCES*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp05100.pdf_