Journal of Globalization and

Development
Volume 1, Issue 2 2010 Article 5

Aid, Growth, and Development: Have We

Come Full Circle?

Channing Arndt, University of Copenhagen

Sam Jones, University of Copenhagen
Finn Tarp, University of Copenhagen and UNU-WIDER

Recommended Citation:
Arndt, Channing; Jones, Sam; and Tarp, Finn (2010) "Aid, Growth, and Development: Have We
Come Full Circle?," Journal of Globalization and Development: Vol. 1: Iss. 2, Article 5.

DOI: 10.2202/1948-1837.1121
 Aid, Growth, and Development: Have We
Come Full Circle?
Channing Arndt, Sam Jones, and Finn Tarp

Abstract
The micro-macro paradox has been revived. Despite broadly positive evaluations at the
micro- and meso-levels, recent literature doubts the ability of foreign aid to foster economic
growth and development. This paper assesses the aid-growth literature and, taking inspiration from
the program evaluation literature, we re-examine key hypotheses. In our findings, aid has a
positive and statistically significant causal effect on growth over the long run, with confidence
intervals conforming to levels suggested by growth theory. Aid remains a key tool for enhancing
the development prospects of poor countries.

KEYWORDS: foreign aid, growth, aid effectiveness, causal effects

Author Notes: We have benefited greatly from editorial advice and a careful anonymous
referee report. We also thank Tony Addison, Ernest Aryeetey, Pranab Bardhan, Bruce Bolnick,
Imed Drine, Jan Willem Gunning, Gerry Helleiner, Paul Isenman, Homi Kharas, Dirk Krueger,
David Roodman, Erik Thorbecke, Alan Winters, and Adrian Wood for encouragement and most
valuable comments. The same goes for the participants at conferences and seminars held by the
African Economic Research Consortium (AERC), the Bergen Resource Center for International
Development at CMI Norway, the Brookings Institution, central ministries in Mozambique,
Tanzania and Vietnam, Cornell University, the European Union Development Network (EUDN)
(organized by CERDI, Clermont-Ferrand, France), the Helsinki Center for Economic Research
(HECER), NORAD Norway, OECD Paris, the United Nations HQ (organized by UNU-ONY), the
University of Copenhagen, the University of Ghana, the UNU Conference of Directors
(CONDIR), and UNU-WIDER. Thanks are also due to Tseday Jemaneh Mekasha for excellent
research assistance and to Raghuram G. Rajan and Arvind Subramanian for sharing their original
data and STATA files. The usual caveats apply.
 Arndt et al.: Aid, Growth, and Development: Have We Come Full Circle?

1. Introduction
The extent to which foreign aid can be a decisive factor in the economic
development of low-income countries remains controversial. In 1987, Paul
Mosley suggested that while aid seems to be effective at the microeconomic level,
any positive aggregate impact of aid is much harder to identify (Mosley, 1987).
He labeled this the “micro-macro” paradox. Now, after two decades of intense
analytical work using new theory, new data, and new empirical methodologies, it
would appear that the paradox has been revived. At the macro-level, Rajan and
Subramanian (2008) conclude “it is difficult to discern any systematic effect of
aid on growth”. Nevertheless, evaluations of aid effectiveness at the
microeconomic level continue to indicate positive rates of return (World Bank,
2008). Also, an increasing number of rigorous microeconomic impact evaluations
have demonstrated the potential for well-designed project interventions to
generate positive results (Banerjee and Duflo, 2009).
This paper has two objectives. First, we attempt to provide a balanced, up-
to-date assessment of the aid-growth literature. We observe that, while aid may
have very high returns in specific circumstances (Collier and Hoeffler, 2004),
expectations surrounding the average potency of aid have been excessive. Second,
we turn our attention to the fundamental empirical evaluation challenge:
identifying the counterfactual. Using observational data, there is no way of
identifying a plausible counterfactual without making assumptions that are bound
to be debatable, in theory and practice. Applying recent methodological advances
in the program evaluation literature to a macroeconomic question, we propose and
implement a number of empirical improvements. These include strengthening the
aid instrument, improving the model specification, and employing a new doubly
robust estimator. We show, contrary to other recent results, that there is no firm
basis on which to reject the prior that foreign aid exerts a modest long-run
positive effect on growth in developing countries.
The remainder of this paper is structured as follows. Section 2 provides a
literature review. Section 3 considers empirical strategies used to investigate the
aid-growth relationship. Section 4 implements our proposed enhancements and
presents results. A final section summarizes and concludes that there is no micro-
macro paradox.1

1
Please refer to Arndt et al. (2010) for an earlier working paper version of this paper for further
details on background, analytical methods, data, sensitivity tests, and results.

1
 Journal of Globalization and Development, Vol. 1 [2010], Iss. 2, Art. 5

2. Literature review
Scholarship on the relationship between aid, growth, and development is
voluminous. This section provides a brief survey of how the aid-growth debate
has evolved.

2.1. Earlier generations

Studies of the aid-growth relationship from the 1970s until recently have been
classified into three generations, each influenced by dominant theoretical
paradigms as well as available empirical tools (Hansen and Tarp, 2000). The first
two generations were inspired by simple models of the growth process, i.e. the
Harrod-Domar model and the two-gap Chenery-Strout extension. The underlying
idea behind the Harrod-Domar model is of a stable linear relationship between
growth and investment in physical capital. Assuming all aid is invested, it is
straightforward to calculate how much aid is required to achieve a target growth
rate. The impact of aid is positive and helps plug either a savings or a foreign
exchange gap. Empirical studies in this tradition consequently focused on the
extent to which aid increased savings and investment in recipient countries
(Papanek, 1972, 1973). Overall, first generation studies show that aid tends to
increase total savings, but not by as much as the aid flow. Quite reasonably, this
suggests a non-negligible proportion of aid is consumed rather than invested.
Retaining the focus on capital accumulation, a second generation of
literature explored the impact of aid on growth via investment. Using data for a
cross section of countries, a large number of studies of this kind were produced
during the 1980s and early 1990s. These studies consistently pointed to a positive
link between aid and investment. While a majority of the aid-growth studies of
this generation also suggested a positive impact, the result that captured attention
was Paul Mosley’s micro-macro paradox. An influential line of critique of the
Harrod-Domar and two-gap approach was the argument that growth is less related
to physical capital investment than often assumed (Easterly, 1999). If the
productive impact of aid depends more on incentives and relative prices, as well
as the policy environment more generally, then it becomes important to consider
these broader effects. The second generation of studies also introduced the
problem that poorly performing countries may receive more aid precisely because
of their poor growth performance. Empirical analyses that do not account for the
endogeneity of aid will not reveal aid’s causal impact. Most second generation
studies, however, did not deal with this issue.
From the early 1990s a third generation of more sophisticated econometric
studies came to dominate the discourse about aid. This was motivated by the
availability of panel data, allowing analysts to look at changes both across and

DOI: 10.2202/1948-1837.1121 2
 Arndt et al.: Aid, Growth, and Development: Have We Come Full Circle?

within countries over time. Insights from new theories of economic growth also
influenced the research agenda. Mindful of the weaknesses of previous studies,
the aid-growth relationship came to be perceived as (possibly) non-linear and the
endogeneity of aid was taken more seriously. Among the numerous studies of this
generation, the contribution by Burnside and Dollar (2000) came to exert a
significant influence on policy. These authors made an argument for conditional
aid effectiveness, specifically: “aid has a positive impact on growth in developing
countries with good fiscal, monetary and trade policies ... [but] ... in the presence
of poor policies, aid has no positive effect on growth” (2000: 847).
However, these results were subject to criticism. Hansen and Tarp (2001)
found that a story of diminishing returns to aid best captured the non-linear
relationship between aid and growth. In a later contribution, Easterly et al. (2004)
added that the Burnside-Dollar aid-policy result is fragile when the dataset is
expanded (by years and countries). Dalgaard et al. (2004) found that aid has been
less effective in tropical areas over the last 30 years and called for further research
to help disentangle the channels through which aid matters for productivity. In an
empirical review of these contributions, Roodman (2007) argued that the results
of this generation are extremely sensitive to methodological choices, concluding
that while some aid is likely to increase investment and growth, aid “is probably
not a fundamentally decisive factor for development” (2007: 275).

2.2. Recent studies

More recently, a fourth generation of literature has emerged. A distinctive aspect

of this generation is the view that aid’s aggregate impact on economic growth is
non-existent. A leading paper that appears to establish this result is Rajan and
Subramanian (2008). They find no systematic effect of aid on growth regardless
of the estimation approach, the time period. and the type of aid. Explanations for
non-positive aggregate effects of aid often refer to political economy dynamics.
For example, Djankov et al. (2008) argue that aid has effects that are analogous to
a natural resource curse. Similarly, Rajan and Subramanian (2007) find that the
rate of growth of value added by the manufacturing sector in developing countries
has been undermined by a detrimental effect of aid inflows on governance quality.
Fourth generation scholars have also become increasingly skeptical about
our ability to make valid causal inferences with respect to complex aggregate
phenomena, such as the determinants of economic growth. In particular, previous
methods used to deal with endogeneity have been subject to criticism. There is
increasing awareness that dynamic panel (system) GMM methods – frequently
employed in the third generation – are not a panacea. The concern that weak
instruments typically bias coefficient estimates towards their unadjusted
counterparts (e.g., OLS or panel fixed effects estimates) applies as much to panel
GMM as to cross-section estimators. Bun and Windmeijer (2010) show that the

3
 Journal of Globalization and Development, Vol. 1 [2010], Iss. 2, Art. 5

weak instrument problem (previously attributed mainly to the Arellano-Bond

estimator) may be equally problematic in the system approach. Also, for the
Blundell-Bond (system GMM) estimator to be valid, both country fixed effects
and omitted variables must be orthogonal to the lagged differences of the right-
hand side (RHS) variables which are used as instruments for the level equation.
This assumption cannot be tested and may be suspect given (i) the highly complex
nature of the growth process, and (ii) that country fixed effects are expected to
incorporate determinants of steady-state income levels that may correlate with
growth along individual countries’ steady-state transition paths.
In a Monte Carlo investigation of the robustness of different panel
estimators, Hauk and Wacziarg (2009) conclude that the principle issue for
system GMM is not one of strong or weak instruments but the validity of these
moment conditions. Roodman (2009) warns that the Blundell-Bond estimator may
give a false sense of certainty as a large number of internal instruments can over-
fit the endogenous variables and may weaken the power of Hansen/Sargan tests.
Finally, internal instruments do not prevent bias arising from systematic
measurement error in the endogenous regressors, which is an important limitation
in the context of aid-growth regressions.
In response to these concerns, alternative methods to assess aid
effectiveness have come to the fore. These often eschew cross-country
macroeconomic analysis in favor of specific micro- and meso-outcomes (Temple,
2010). Mishra and Newhouse (2007), for example, uncover a small but
statistically significant effect of health aid on infant mortality. Masud and
Yontcheva (2005) also find that aid helps reduce infant mortality, but this effect is
only significant for aid provided by non-governmental organizations (NGOs).
Alongside cautious optimism surrounding the potential efficacy of
microeconomic policy interventions, financed either directly by donors through
projects or indirectly via budget or sector-wide support, these findings give new
sustenance to the micro-macro paradox. Indeed, with few exceptions (e.g., Sachs,
2005), recent findings at the micro- and meso-levels have not been deployed to
argue that aid is effective on aggregate. This is despite increasing evidence that
meso-level outcomes can add up to substantial macroeconomic effects (Cohen
and Soto, 2007).

2.3. Rationale for continued macro-analysis

In light of the methodological concerns raised by recent scholarship, it is helpful

to reflect on the value of attempting to assess the macroeconomic impact of
foreign aid. In the first place, serious empirical challenges have not dissuaded
economists from investigating other complex questions (Angrist and Pischke,
2010). For example, there are numerous parallels between the problem of
estimating the causal impact of aid on growth and the causal impact of schooling

DOI: 10.2202/1948-1837.1121 4
 Arndt et al.: Aid, Growth, and Development: Have We Come Full Circle?

on earnings. Both problems are likely to be characterized by endogenous

selection, heterogeneous treatment responses, and mis-measurement of treatment
input (both in terms of quality and quantity). Considerable effort has been
expended in the analysis of large, high-quality, schooling datasets by some of the
most skilled econometricians in the profession. Even so, debate has persisted, at
least until recently, with respect to the net bias of ordinary least squares (OLS)
estimates of returns to education (Card, 2001).
If the profession has experienced serious difficulties estimating the causal
effect of schooling on earnings in developed countries, then it should not be
surprising that estimating the impact of aid on growth in developing countries is
contentious. However, it is difficult to deny that the aid-growth issue is both
compelling and relevant. In developed countries, policy-makers and the wider
public continue to ask whether aid is a cost effective use of taxpayer money on
aggregate. Today, the attention of both the aid community and decision-makers is
on “Dead Aid” (Moyo, 2009), which argues for a complete cessation of aid flows
to Africa. We note that the financial crisis of 2008/09 has highlighted the
importance of public spending to stabilize and stimulate economic activity. While
foreign aid has multiple objectives, economic growth is central among them. If
the economics profession as a whole were to abandon the question of aid’s impact
on growth, it would leave the issue open to speculative and potentially unhelpful
contributions.

2.4. Formulating an appropriate prior

A key aim of empirical analysis is to falsify or discriminate between competing

hypotheses. Consequently, it is necessary to make explicit the prior upon which
empirical testing is focused. With respect to the effect of foreign aid on economic
growth, relatively few studies address the issue of an appropriate prior. A recent
exception is Rajan and Subramanian (2008) who consider aid in a standard
neoclassical growth model. Assuming that aid only augments physical capital
investment and has no effect on productivity, they derive that a one percentage
point increase in the ratio of aid to GDP should be expected to raise the growth
rate of per capita GDP by around 0.16 percentage points on average. In practice,
at least some aid is directed towards consumption or non-growth enhancing
activities. As a result, Rajan and Subramanian place the expected growth return at
around 0.1 percentage point for each percentage point of aid in GDP. Thus, the
implied increase in the growth rate accruing from aid inflows at 10% of GDP
should be about 1%, which is considerably less than the predictions based on
Harrod-Domar models. In sum, growth theory points towards more modest
expectations with respect to the potency of aid (see also Dalgaard and Erickson,
2009).

5
 Journal of Globalization and Development, Vol. 1 [2010], Iss. 2, Art. 5

A related issue is the appropriate time-frame over which any growth

effects accruing from aid can be expected to materialize. Various factors may
exert a cumulative but not immediate impact on the rate of income growth. For
example, changes in education move only slowly at the aggregate level and have a
positive influence on economic growth with a substantial lag. This follows from
simple demographics whereby improvements in schooling indicators can take
many years to translate into noticeable increases in average education levels
among working-age adults. Changes in human capital due to improved health
indicators may take even longer to translate into more rapid economic growth.
Ashraf et al. (2008) and Acemoglu and Johnson (2007) find that the initial
economic impact of gains in life expectancy from the health interventions
introduced from the 1940s may be a reduction in per capita incomes due to the
increase in population and dependency ratios. The former authors find that it can
take 30 years or more for per capita incomes to return to pre-intervention levels.
They also find that significant increases in life expectancy at birth only begin to
have a modest positive effect on incomes after about a 35 year lag.2
Overall, a series of considerations indicate that the aid-growth relationship
is only likely to emerge over a long time-horizon. Many aid investments, such as
in education, health, and institution-building are long term in nature; and growth
theory indicates that the contribution of these investments to growth is likely to be
relatively modest. When these observations are combined with the volatility of
growth in most developing countries and the high degree of measurement error
inherent in nearly all the variables of interest, relatively long time-frames would
appear to be necessary to reliably detect the aid-growth relationship.

3. Empirical strategies
Following Temple’s (2010) recommendation to build explicitly on existing
empirical work, our starting point for developing an appropriate empirical
strategy is Rajan and Subramanian (2008) (henceforth RS08). RS08 provides a
thoughtful and highly influential contribution that is widely understood to have
established that aid has no impact on growth. However, a number of concerns
question this fundamental conclusion. This section addresses these concerns and
suggests suitable improvements. Specifically, Section 3.1 provides a brief
summary of RS08’s approach and main results. Section 3.2 presents a detailed
analysis of the validity of their instrumentation strategy. This motivates the
development of an amended instrument in Section 3.3, while Sections 3.4 and 3.5
propose modifications to the specification and regression estimators. Once these
2
Ashraf et al. (2008) focus on demographic trends as a result of disease eradication. Productivity
effects, demand effects, and complementary policies may speed the realization of growth benefits
from health gains.

DOI: 10.2202/1948-1837.1121 6
 Arndt et al.: Aid, Growth, and Development: Have We Come Full Circle?

modifications are incorporated, either individually or in conjunction, the empirical

aid-growth relationship is shown to conform to theoretical priors and the micro-
macro paradox disappears once again. These results are given in Section 4.

3.1. RS08

Before turning to RS08’s main results, it is helpful to examine simple OLS

estimates of the relationship between aid and growth. They are reported by RS08
and are replicated here in column I of Table 1. As expected, they show a negative
estimated coefficient on Aid/GDP. These estimates are biased downward
principally due to a simultaneity problem – because of their slow growth, slower
growing countries typically received more aid during the period. Thus, the
challenge is to find valid and relevant external instruments that explain variation
in aid receipts across developing countries, but which are unrelated to their
growth performance.
In their principal IV approach, RS08 investigate the aid-growth
relationship using long-run averages, turning the focus away from dynamic panel
methods. This is sensible. Dynamic panel methods are subject to doubt given the
expected cumulative effect of aid and corresponding concerns regarding the
validity of internal instruments in GMM estimators. Their long-run approach
echoes the average OLS estimator proposed by Mankiw, Romer and Weil (1992)
as well as the long-difference approach used by Acemoglu and Johnson (2007).
RS08 consider four periods separately: 1960-2000; 1970-2000; 1980-
2000; and 1990-2000. In each period, their instrument for aid is generated from a
“zero” stage regression estimated at the bilateral donor-recipient level using
supply-side factors. These include past colonial relations, relative population
sizes, and interaction terms. The predicted Aid/GDP ratio estimated from this
regression is aggregated across donors to give a fitted average ratio for each
recipient. This is then used as a single excluded instrument in a 2SLS estimation,
where average growth over the period is the dependent variable. Core results from
RS08 for the 1970-2000 period are replicated in column II of Table 1. The
generated instrument appears reasonably strong according to conventional
measures, such as the first stage partial F-statistic; also, the coefficient on
Aid/GDP is exactly in line with the prediction from their growth model but is not
significant.
RS08 conclude that there is no systematic (causal) effect of aid on growth,
and move on to show this holds for alternative sub-periods (RS08 Table 4),
alternative growth horizons (RS08 Table 6), non-linear effects (RS08 Table 7),
and different types of aid (RS08 Table 8). Similarly, the same basic result
emerges when the question is considered in a dynamic panel setting (RS08 Table
10).

7
 Journal of Globalization and Development, Vol. 1 [2010], Iss. 2, Art. 5

Table 1: Alternative models for aid and growth, 1970-2000

(I) (II) (III) (IV) (V) (VI) (VII) (VIII)
OLS 2SLS IV-LIML IV-LIML IV-LIML IV-LIML IV-LIML IV-LIML

Aid / GDP -0.08* 0.10 0.10 0.07 0.08 0.12 0.21* 0.19*
(0.04) (0.07) (0.08) (0.06) (0.06) (0.10) (0.11) (0.10)

Initial per cap. GDP -1.67*** -1.41*** -1.40*** -1.44*** -1.44*** -1.44*** -1.34*** -1.36***
(0.32) (0.43) (0.39) (0.37) (0.36) (0.30) (0.33) (0.33)
Initial level of policy 2.28*** 2.14*** 2.13*** 2.16*** 2.15*** 2.28*** 2.29*** 2.28***
(0.47) (0.62) (0.56) (0.52) (0.53) (0.46) (0.52) (0.51)
Initial life expectancy 0.02 0.08* 0.08** 0.07** 0.07** 0.04 0.05 0.05
(0.03) (0.04) (0.04) (0.03) (0.03) (0.04) (0.04) (0.04)
Geography 0.39** 0.61** 0.62** 0.58*** 0.58*** 0.25 0.29 0.28
(0.18) (0.26) (0.24) (0.22) (0.22) (0.22) (0.24) (0.23)

Excluded instruments 1 1 7 3 1 7 3 1
Regional dummies SSA, EA SSA, EA SSA, EA SSA, EA SSA, EA SSA, A, LA SSA, A, LA SSA, A, LA
N 78 78 78 78 78 78 78 78
R-squared 0.70 0.59 0.58 0.62 0.61 0.69 0.65 0.66
Weak identification stat. - 31.6 5.50 10.40 33.41 4.32 5.06 14.32
Stock-Wright LM S stat. - - 11.85 3.05 2.12 18.17 5.40 5.20
(probability) 0.158 0.384 0.145 0.020 0.145 0.023
Hansen J stat. - - 7.72 0.29 - 11.88 0.31 -
(probability) 0.358 0.865 0.104 0.857
significance level: * 10%; ** 5%; *** 1%
Notes: only selected variables reported; columns (I) and (II) replicate results in Rajan and Subramanian (2008; RS08);
column (III) employs a full set of aggregate instruments (see Table 2) in place of a single generated instrument; column (IV)
restricts the aggregate instrument set to the mean population ratio, the colony dummy and their interaction; column (V) uses
only log initial (recipient) population as the instrument; columns (VI) to (VIII) replicate columns (III) to (V) but use a
preferred set of conditioning variables (not shown, see Section 3.4); regional dummies are included as indicated, SSA = sub-
Saharan Africa, EA = East Asia, A = Asia, LA = Latin America & Caribbean; weak identification statistic is the first stage
partial-F statistic in col. (II), and the Kleibergen-Paap Wald F statistic elsewhere; initial policy refers to the Sachs-Warner
trade policy index; geography refers to the average of the number of frost days and tropical land area; standard errors, given
in parentheses, are robust to arbitrary heteroskedasticity; dependent variable is mean real growth rate; Aid/GDP is treated as
endogenous in all models except column (I); in columns (I) to (V) Aid /GDP is taken from RS08, in columns (VI) to (VIII)
it is re-estimated from OECD-DAC (2008) data treating possible missing values as zeroes (see Section 3.3).
Source: authors’ estimates.

3.2. Instrument validity

The supply-side approach to instrumentation developed by RS08 represents the

state of the art in the aid-growth literature. Nevertheless, it has been subject to
criticism. Clemens and Bazzi (2009) note that different authors have used the
same variables as exogenous instruments for a wide range of endogenous
variables. This raises the possibility that these exogenous instruments are
correlated with other omitted variables, thereby invalidating the exclusion
restriction on which valid causal inference depends. They direct specific attention
to the reliance of the RS08 (fitted) instrument on the natural logarithm of the aid
recipient’s population size. They find that log population has a “statistically

DOI: 10.2202/1948-1837.1121 8
 Arndt et al.: Aid, Growth, and Development: Have We Come Full Circle?

significant partial relationship with several variables that are plausible growth
determinants” (2009: 11) and that are omitted from RS08’s specification.
While the existence of a partial correlation between omitted explanatory
variables and the chosen instrument indicates that the coefficients in a regression
specification may be biased, the extent of bias is, ultimately, an empirical matter.
This is recognized by Clemens and Bazzi (2009), leading them to advocate
application of a range of empirical tests for instrument validity. On the face of it,
straightforward validity checks of the RS08 (generated) instrument based on
Sargan or Hansen tests are not possible because their IV model is just identified –
i.e., the number of excluded instruments equals the number of endogenous
variables. Nonetheless, recalling that the zero stage of the RS08 approach
generates a single instrument as a linear combination of variables, it is possible to
use modified versions of these same variables as excluded instruments directly in
the aggregate aid-growth regressions. This provides for a large number of
potential instruments, and therefore permits Hansen tests to be run either on the
full set or on specific sub-sets of instruments.
Following this logic, we collapse the bilateral aid dataset along the donor
dimension, thereby transforming the explanatory variables used in the bilateral
zero stage regressions for use at a more aggregate level. For continuous zero stage
regressors, such as the donor-recipient population ratio, the corresponding
“aggregate” instrument is the mean of the population ratio for each recipient
across all donors. For dummy regressors, such as the specific colonizer, it is more
appropriate to take the maximum value of the dummy for a given recipient (again,
across all donors). Ignoring relatively minor variables, such as currently being a
colony and the population-colony interaction terms employed in RS08, this yields
a set of eight possible instruments as per the rows of Table 2.
Column I of Table 2 verifies whether these aggregate instruments are
adequate proxies for the fitted instruments generated from the zero stage
regressions. As expected, the explanatory power is high. Moreover, underlining
the contention of Clemens and Bazzi (2009), a driving force behind the fitted aid
instruments appears to be the population ratio term. Thus, a fundamental issue for
the RS08 instrumentation strategy is the validity of the exclusion restriction as it
applies to the population-based instruments. Nevertheless, the results from
column I of Table 2 indicate that other variables make some (albeit smaller)
contribution to the overall fitted instrument. Thus, to further test instrument
validity, we re-estimate the RS08 model employing the full set of eight aggregate
instruments. The results, reported in column III of Table 1, closely replicate
column II; and the Hansen J test reports a probability of 0.358, which fails to
reject the validity of the exclusion restriction assumption.

9
 Journal of Globalization and Development, Vol. 1 [2010], Iss. 2, Art. 5

Table 2: Instrument validity checks, 1970-2000

Fitted coefficients Residual coefficients RS08 model AJT model
RS08 AJT RS08 AJT C stat. Prob. C stat. Prob.
(I) (II) (III) (IV) (V) (VI) (VII) (VIII)

Population ratio 0.85*** 0.59*** -0.18 -0.03 0.40 0.53 0.24 0.63
Colony (ever) 0.10** 0.01 -0.19 -0.25 1.11 0.29 1.96 0.16
Pop. ratio × colony 0.11 0.32*** 0.10 0.09 0.48 0.49 0.06 0.81
Common language 0.04 0.21*** 0.25* 0.26* 2.36 0.12 0.27 0.60
Spanish colony -0.08 0.03 0.26 0.18 0.29 0.59 1.36 0.24
Portuguese colony 0.03 0.04* 0.27** 0.08 3.28 0.07 1.77 0.18
French colony -0.01 0.11*** 0.53** 0.71*** 2.14 0.14 8.13 0.00
UK colony -0.25*** 0.11** 0.35 0.42** 0.02 0.89 1.02 0.31
R-squared 0.95 0.99 0.12 0.23 - - - -
significance level: * 10%; ** 5%; *** 1%
Notes: columns (I) and (II) report standardized OLS regression coefficients in which the dependent variable is the fitted aid
instrument taken from Table 3 columns (I) and (V) respectively; columns (III) and (IV) report standardized OLS regression
coefficients from regressions of residuals saved from columns (IV) and (VI) of Table 1 against the row variables; columns
(V) to (VIII) report individual difference-in-Hansen C statistics and corresponding probabilities associated with each
individual row instrument (relative to the full instrument set); all OLS regression specifications incorporate the relevant set
of included instruments as additional controls (not reported); significance from OLS regressions are based on robust
standard errors.
Source: authors’ estimates.

Nevertheless, we take one further step to investigate the exclusion

restriction. Following the intuition of Sargan-type tests, we save the residuals and
regress them against the set of excluded instruments. This provides initial insight
as to which variables in the instrument set may be suspect. Standardized
coefficients from these regressions are given in column III of Table 2. They show
that neither the population ratio term nor its interaction with the (ever being a)
colony dummy is significantly correlated with the unexplained components of the
growth models. In contrast, the Portuguese and French colonizer effects are
significant. In our alternative specification (denoted AJT) shown in column IV of
Table 2, the French and UK colonizer terms are both significant.
Following this simple but intuitive OLS approach, columns V to VIII of
Table 2 report formal tests of the orthogonality of each of the individual aggregate
instruments to the growth regression errors. Specifically, they report the
difference-in-Hansen C statistic associated with excluding each row instrument
(individually) from the full set. For example, in the first row the C statistic
corresponds to the reduction in the overall Hansen J test statistic when the
population ratio term is excluded from the instrument set; the corresponding
probability is also shown. These findings corroborate the residual-based OLS
results. The most suspicion falls on the colonizer terms; however, the population

DOI: 10.2202/1948-1837.1121 10
 Arndt et al.: Aid, Growth, and Development: Have We Come Full Circle?

ratio variables do not give cause for concern, providing comfort as to their
suitability as exogenous instruments in these models.

3.3. Improved instrumentation strategy

The results of Section 3.2 indicate that the RS08 instrumentation approach is
broadly convincing but is weakened by inclusion of suspect variables in the zero
stage. From a theoretical point of view, the validity of using colonizer dummies
and their interactions as instruments is questionable. The institutional transplants
and broader colonizing strategies pursued by imperial powers were not alike, and
they may have a persistent effect on income levels to the present day. This notion
is at the heart of the debate concerning the effect of different legal origins (La
Porta et al., 2008), historical events (e.g., Nunn, 2008), and other institutional
forms on contemporary economic outcomes. Put simply, the colonial relations
variables are not orthogonal to growth and therefore should not be included in the
zero stage regression explaining aid.
As a first step towards improving the RS08 instrument, we re-run their
aid-growth model using a smaller and “less suspect” sub-set of the aggregate
instruments used in Section 3.2. These are the population ratio, a dummy for ever
having been a colony and their interaction. Results are given in column IV of
Table 1, showing that the Hansen J test is now passed with a high level of
confidence. Nevertheless, compared against column II, the results also suggest a
trade-off between efficiency and transparency in instrument selection. While the
use of multiple aggregate instruments is more transparent, it does not exploit the
full information about bilateral aid flows contained in the zero stage. This may be
one reason why the weak identification statistics are considerably lower in
column IV versus column II of Table 1. In fact, as shown in column V of Table 1,
even if only one aggregate instrument is employed, namely the log of the
recipient’s initial population, the strength of the instrument returns to similar
values to those in column II and all coefficients are essentially unchanged.3
Consequently, using a single instrument is likely to be more efficient but there are
also potential information gains from employing a zero stage, especially in small
aggregate samples such as those used in (static) cross-country regressions.
Thus, to strengthen the instrumentation approach, we return to the zero
stage regressions. Aside from removal of suspect terms, additional concerns
motivate further modifications. First, there are errors in the calculation of average
Aid/GDP in all stages of RS08’s regressions. The OECD-DAC aid dataset used
for bilateral aid flows includes numerous missing values. While in some cases
these genuinely refer to absent data, in most cases they represent unreported null

3
This result further underlines the reliance of the RS08 instrumentation strategy on population
size (also see Clemens and Bazzi, 2009).

11
 Journal of Globalization and Development, Vol. 1 [2010], Iss. 2, Art. 5

values.4 RS08 incorrectly treat these as missing. This is material because it

distorts estimates for average bilateral aid flows over time. Consequently, it is
necessary to re-estimate the bilateral aid variables and calculate period averages
for Aid/GDP and aid per capita, setting missing entries to zero. This affects the
dependent variable employed in the zero stage regression as well as the
endogenous aid variable used in the IV estimations.
Second, in the RS08 strategy, recipient GDP occurs in the denominator of
the dependent variable in the zero stage regressions. Following Kronmal (1993),
inappropriate use of ratio variables may lead to substantial misinterpretation (or
bias) in least squares regressions. This may arise if the denominator of the
dependent variable is correlated with the RHS variables independently of the
numerator of the dependent variable. In the present case, this could arise if donor
decision rules do not target the Aid/GDP ratio, and/or if there is a direct
association between recipient GDP levels and population size or past colonial
experiences.
Third, it is apparent that individual donor countries exhibit distinct
attitudes to giving foreign aid (Alesina and Dollar, 2000), which reflect cultural
and historical factors. These time-invariant influences can be understood as fixed
effects and may be included as RHS variables in the zero stage regression.
Notably, and unlike the RS08 explanatory variables, these fixed effects may
explain a part of the variation in aid allocations that is unrelated to purely strategic
or political motives. As such, their inclusion may strengthen the overall validity
and interpretation of the generated instrument.
To address these concerns, we modify the RS08 specification of the zero
stage regression. In place of Aid/GDP, we use aid per capita (Aid/POP) as the
dependent variable which accords closely with the explicit aid allocation rules
used by donors, such as the World Bank (see Annex 1 of IDA15, 2008).5 We drop
the colonizer-specific variables (and interactions) and only include a dummy for
whether a country was ever a colony (COLONY). Adding donor-specific fixed
effects (DONOR), our zero stage regression emerges as follows:

(1)

where the subscripts d and r represent donors and recipients respectively;

CURCOL indicates whether the recipient is a current colony of the donor.

4
Confirmed in correspondence with the OECD DAC Secretariat.
5
Note that in all subsequent regression stages the endogenous variable of interest remains
Aid/GDP.

DOI: 10.2202/1948-1837.1121 12
 Arndt et al.: Aid, Growth, and Development: Have We Come Full Circle?

Table 3: Alternative zero stage regressions, 1970-2000

(I) (II) (III) (IV) (V)
OLS OLS OLS OLS Heckman

Colonial relationship (dummy) 1.65*** 2.09*** 11.95*** -0.55 -0.88

(0.24) (0.19) (1.62) (2.08) (2.20)

Currently a colony (dummy) -0.97* 0.63 9.88*** 14.14 24.48

(0.56) (0.45) (3.81) (21.15) (36.71)

Common language (dummy) 0.07* 0.09*** 1.36*** 1.30** 1.30*

(0.04) (0.03) (0.27) (0.60) (0.67)

Ratio of (initial) log population 0.09*** 0.05*** 0.40*** 0.32*** 0.45***

(0.01) (0.00) (0.04) (0.06) (0.08)

Ratio of log population x colony 0.62*** 0.77*** 7.16*** 3.32*** 3.36***

(0.11) (0.08) (0.69) (0.72) (0.77)

Dependent variable Aid/GDP Aid/GDP Aid p.c. Aid p.c. Aid p.c.
Treatment of “missing” aid values Unknown Zero Zero Zero Zero
Metropole fixed effects & interactions Yes Yes Yes No No
Donor fixed effects No No No Yes Yes
Outcome and selection independence - - - - 9.56***
Number of obs. 3288 3286 3328 3328 3328
R-squared 0.42 0.31 0.26 0.21 -
F statistic 185.93 113.55 90.49 10.65 -
significance level: * 10%; ** 5%; *** 1%
Notes: column (I) replicates Rajan and Subramanian’s zero stage regression (2008, Table 4); columns (II) and (III) retain
the same RHS specification, but alter the dependent variable (denoted in the table); column (IV) revises the specification,
dropping metropole (colony-specific) fixed effects & interactions (coefficients not shown); column (V) implements a
Heckman correction, based on the specification in column (IV); Heckman estimator uses full information maximum
likelihood (FIML); the Heckman selection equation (not shown) includes all outcome covariates and a dummy for the
number of colonial relationships experienced by the recipient; test for independence of outcome and selection equations
refers to a Wald test that the correlation (rho) between the residuals in the two equations is equal to zero; intercept not
shown; standard errors are robust to arbitrary heteroskedasticity and intra-group correlation between aid recipients (except
for columns I to III where standard errors assume homoskedasticity in order to replicate Rajan and Subramanian, 2008).
Source: authors’ estimates.

Results from these modifications are given in Table 3. Column I replicates

the RS08 specification (only selected coefficients shown); column II employs the
revised dependent variable in which missing Aid/GDP values are set to zero. This
change has a moderate impact and the pair-wise correlation between the fitted
values from these two models is 0.83. Column III also retains the original RHS
specification, but introduces aid per capita as the dependent variable (with
missing aid values set to zero). All core coefficients retain the same sign and
significance, but there is a minor fall in explanatory power, indicating there may
have been some unwanted independent correlation between GDP in the dependent
variable and the RHS variables. Column IV employs the new RHS specification,

13
 Journal of Globalization and Development, Vol. 1 [2010], Iss. 2, Art. 5

as per equation (1). Again, there is a small loss of explanatory power, but the
population ratio and its interaction with the colony dummy remain highly
significant. Also, the donor fixed effects (coefficients not shown in the table) vary
in sign and many are significant. Overall, the RHS variables continue to explain a
reasonable share of observed aid allocations.
The existence of zero-value aid inflows points to a final possible
weakness. In principle, the decision by a donor to provide aid involves at least
two distinct choices (Tarp et al., 1999): (i) which recipients should receive aid;
and (ii) how much to supply – i.e., the distribution of bilateral aid flows reflects
an unobserved selection process. In the absence of an explicit model, one way to
address potential bias from unobserved selection effects is to use Heckman’s
correction (Heckman, 1979). Column V of Table 3 employs a Heckman selection
model (estimated by full information maximum likelihood) to the specification in
column IV, where the existence of zero or non-zero aid flows is used as the binary
selection variable. Despite these changes, the direction of the results and their
interpretation are largely unchanged. However, we reject the hypothesis that there
is no selection bias. We therefore retain the Heckman estimator employed in
Column V as our preferred zero stage regression.

3.4. Improved specification

Before presenting the results of the aid-growth IV regressions using the improved
instrument, it is appropriate to discuss additional areas where the RS08 approach
can be strengthened. The first of these is the choice of covariates. Given the
relatively small sample available in the aggregate regressions (78 countries),
inclusion of redundant variables may lead to a loss of efficiency and/or contribute
to undesirable multicollinearity. In the case of RS08, we note that the three
macroeconomic initial conditions (inflation, money supply, and budget balance)
as well as ethnic fractionalization are insignificant in RS08’s cross-section
outcome regressions for all periods. In addition, and as Wooldridge (2005)
clarifies, inclusion of contemporaneous outcome variables – i.e., variables which
may also be affected by the level of treatment – can invalidate the
unconfoundedness assumption required for valid causal inference (Angrist and
Pischke, 2008). This is pertinent as RS08’s chosen specification includes two
variables that capture average outcomes during the period of analysis –
institutional quality and the number of forced changes in the top government elite,
labelled “revolutions”. Inclusion of these variables is puzzling in light of the
literature which examines the effects of aid on growth through institutional
performance. Controlling for such outcomes blocks potential channels through
which aid may affect growth and thereby restricts the estimated coefficient on aid
to a partial as opposed to a general effect. Such variables may also introduce
unwanted reverse causality.

DOI: 10.2202/1948-1837.1121 14
 Arndt et al.: Aid, Growth, and Development: Have We Come Full Circle?

It is also helpful to consider the appropriate role of regional fixed effects.

In RS08’s specification, only East Asia and sub-Saharan Africa are included as
regional dummy variables. This appears to be an ex post choice in the sense that
prior to the 1980s there was no particular reason to identify these as “special”.
Including regional dummy variables helps absorb intra-regional correlations and
captures omitted spatial fixed effects such as those arising from geography, shared
historical experiences, and trade relationships. A priori, a more plausible approach
is to include a fuller set of regional dummies. Finally, it is appropriate to include
additional variables that reflect initial socioeconomic conditions such as education
and health indicators, as well as additional geographic characteristics such as
trading distances. These variables are frequently seen as important determinants
of growth and may also proxy for initial conditions; as such, they may explain
some of the variation in the expected growth returns to aid.
Consequently, we propose a revised covariates specification (denoted
AJT). This involves dropping contemporaneous outcome covariates and
redundant variables, adding an alternative set of regional dummies and including
additional controls. These are selected following Sala-i-Martin et al. (2004) who
undertake comprehensive Bayesian averaging of long-run growth estimates. We
include variables identified by these authors that are among those with the highest
posterior probability of inclusion and refer to initial conditions. To this, we add
civil liberties in 1972 and distance to major ports. The first of these captures
additional dimensions of initial institutional quality, including the ability of
citizens to bring the government to account, which is often deemed relevant for
aid effectiveness. Air distance is associated with export transaction costs, and ease
of access to developed markets and has recently been identified by Moral-Benito
(2009) as a robust correlate of growth.

3.5. Alternative estimators

Another area that can be strengthened concerns the choice of IV estimator. In

light of the expected complexity of the growth process as well as the different
properties of alternative estimators, it is valid to investigate whether or not
empirical results hold across different estimators. While RS08 employ a 2SLS
estimator, this is not the only option. Other possible estimators, which offer
moderate differences, include LIML (limited information maximum likelihood),
Fuller’s modified LIML (with alpha = 1), and a continuously updated GMM
estimator (GMM-CU).
In the program evaluation literature, the “doubly robust” estimators of
Robins and Rotznitzky (1995) are attractive. Imbens and Wooldridge describe
these estimators as “best practice” (2009: 25). Various doubly robust estimators
have been proposed (see Imbens, 2004); however, none of these can be applied
straightforwardly to the current aid-growth problem. They assume a binary

15
 Journal of Globalization and Development, Vol. 1 [2010], Iss. 2, Art. 5

treatment/control framework in which receipt of the treatment is conditionally

independent of potential outcomes. Indeed, Imbens and Wooldridge (2009) note
that, compared with the binary case, much less is known about settings with
continuous treatment variables even though such settings are common in practice.
Nevertheless, some simple extensions enable doubly robust estimators to be
extended to the instrumental variables context (see Arndt et al., 2010).
Principally, this involves dichotomization of the (generated) instrument, whilst
maintaining the endogenous variable in continuous form. This is our preferred
estimator, and we denote this IV-IPWLS (IV inverse probability weighted least
squares).

4. Empirical results
Section 3 motivated and proposed three main improvements to RS08’s empirical
approach. These refer to their instrument, their specification, and their chosen
estimator. Tables 4 and 5 present results for different combinations of these
modifications. We focus on the 1970-2000 period, allowing the effects of aid to
be considered over a generation of elapsed time. The shorter periods (1980-2000
and 1990-2000) may not allow sufficient time for the aid-growth relationship to
emerge. With respect to 1960-2000, many countries had not attained
independence by 1960, particularly those in Africa. Further, even though the
majority of French colonies achieved independence in 1960, the shift to
independent administration was very gradual in most cases. In contrast, by 1970
the large majority of developing countries had achieved independence and had
operated independently for at least a few years, with Portuguese colonies being
the prominent exception.
Column I of Table 4 reports regression results using the RS08
specification, the new preferred instrument (based on the zero stage regression in
column V of Table 3), and the IV-LIML estimator. Column II introduces the
doubly robust IV-IPWLS estimator, while columns III and IV replicate columns I
and II with our new specification. Columns V and VI continue with the same
model, but respectively use the GMM-CU and Fuller estimators. Note that in all
columns the improved instrument is employed.

DOI: 10.2202/1948-1837.1121 16
 Arndt et al.: Aid, Growth, and Development: Have We Come Full Circle?

Table 4: Modified IV regressions, 1970-2000

(I) (II) (III) (IV) (V) (VI)
IV-LIML IV-IPWLS IV-LIML IV-IPWLS GMM-CU Fuller

Aid / GDP 0.22* 0.21* 0.25** 0.13*** 0.25** 0.24**

(0.12) (0.13) (0.12) (0.05) (0.12) (0.12)

Initial per cap. GDP -1.34*** -1.92*** -1.03*** -1.33*** -1.03*** -1.05***
(0.40) (0.39) (0.38) (0.27) (0.38) (0.37)
Initial level of policy 2.14*** 2.58*** 2.12*** 2.44*** 2.12*** 2.12***
(0.60) (0.62) (0.54) (0.46) (0.54) (0.53)
Initial life expectancy 0.09** 0.05 0.04 0.03 0.04 0.03
(0.04) (0.03) (0.04) (0.04) (0.04) (0.04)
Geography 0.63** 0.48** 0.29 0.25 0.29 0.29
(0.25) (0.24) (0.26) (0.21) (0.26) (0.25)
Coastal pop. density in 1965 0.00** 0.00*** 0.00** 0.00**
(0.00) (0.00) (0.00) (0.00)
Primary schooling in 1960 2.58** 2.26** 2.58** 2.56**
(1.15) (0.88) (1.15) (1.13)
Malaria risk in 1966 -1.50* -1.06* -1.50* -1.49*
(0.85) (0.58) (0.85) (0.83)
Invest. goods price, 1960-64 -0.01 -0.01 -0.01 -0.01
(0.00) (0.00) (0.00) (0.00)
Civil liberties in 1972 -1.28* -0.98* -1.28* -1.24*
(0.70) (0.50) (0.70) (0.68)
Air distance (log) 0.09 -0.03 0.09 0.09
(0.38) (0.33) (0.38) (0.38)

Specification RS08 RS08 AJT AJT AJT AJT

Scale of excluded instrument Continuous Binary Continuous Binary Continuous Continuous
Regional dummies SSA, EA SSA, EA SSA, A, LA SSA, A, LA SSA, A, LA SSA, A, LA

N 78 78 78 78 78 78
R-squared 0.57 0.70 0.59 0.77 0.59 0.60
Kleibergen-Paap Wald F stat. 29.48 24.42 17.28 39.78 17.28 17.28
Stock-Wright LM S stat. 4.33 3.53 5.77 6.49 5.77 5.77
(probability) 0.037 0.060 0.016 0.011 0.016 0.016
significance level: * 10%; ** 5%; *** 1%
Notes: the endogenous variable is Aid/GDP, re-estimated from OECD-DAC (2008) data treating possible missing values as
zeroes; in columns (I) and (II) the specification follows Rajan and Subramanian (2008) (only selected covariates shown); all
remaining columns use a modified specification, removing contemporaneous and redundant covariates and adding
additional initial conditions; chosen estimator is in the column title; initial policy refers to the Sachs-Warner trade policy
index; geography refers to the average of the number of frost days and tropical land area; intercept not shown; standard
errors, given in parentheses, are robust to arbitrary heteroskedasticity; dependent variable is the average real growth rate.
Source: authors’ estimates.

With respect to the implementation of the IV-IPWLS estimator (columns

III and IV), a binary instrument is required. This is derived by taking the fitted
instrument from RS08’s zero stage regression, sorting countries in ascending

17
 Journal of Globalization and Development, Vol. 1 [2010], Iss. 2, Art. 5

order (from lowest to highest predicted aid shares), and then selecting the first 30
for the “control” and the rest for the “treatment” group. The motivation for this
choice is to identify a sub-sample of countries with the smallest possible average
value for predicted aid inflows while still maintaining statistical viability. Thus, in
practice, the control group approximately corresponds to all countries falling
below the 40th percentile. Besides permitting application of doubly robust
techniques, dichotomization of the instrument represents a useful robustness
check. If results arising from the binary instrument were not comparable with its
continuous counterpart, this might indicate that the latter findings were driven by
peculiarities in the distribution of the instrument. It also relaxes the assumption of
a constant linear relationship between aid and growth, instead placing emphasis
on the average difference between treatment and control groups regardless of the
shape of growth’s response to aid. Consequently, possible non-linear effects due
to diminishing returns to aid are addressed by this dichotomization. Finally, as the
instrument is derived from a zero stage regression, dichotomization provides a
check against measurement error or misspecification in the zero stage.
Turning to results, the range of test statistics reported in Table 4 indicates
that the new instrument continues to perform strongly across different
specifications and estimators. Under-identification tests (not shown), which can
be interpreted as testing the null hypothesis of a zero correlation between the
instruments and the endogenous regressors, are all rejected. The weak
identification test (the Kleibergen-Paap Wald F statistic, which uses a finite-
sample adjustment of the standard F-statistic to assess the strength of the partial
correlation between the excluded instruments and the endogenous variables in
first-stage regressions) not only exceeds critical values in all cases but is
comparable to the levels achieved using RS08’s original approach (Table 1,
column II). Perhaps more importantly, the Stock-Wright S statistic, which is
based on the reduced form regression and is robust to the presence of weak
instruments (see Baum et al., 2007), finds a significant (partial) correlation
between the instrument and dependent variable in all cases.
Moving across the columns of Table 4, we note that the treatment effect –
i.e., the coefficient on the endogenous aid variable – is consistently positive,
significant, and in a domain that is consistent with the RS08 prior (Section 2.4).
The main effect of using the new and strengthened instrument (column I) is that
the treatment effect estimate edges upwards (from 0.l0 to 0.22). The doubly-
robust estimator leaves this result almost unchanged, but enhances the overall
explanatory power of the model. According to the Kleibergen-Paap Wald F
statistic, switching to the modified specification (column III onwards) slightly
reduces the strength of the instrument in the IV-LIML first stage. However, by
placing greater emphasis on the most informative observations, the strength of the
instrument is considerably improved for the IV-IPWLS estimator. Finally, the

DOI: 10.2202/1948-1837.1121 18
 Arndt et al.: Aid, Growth, and Development: Have We Come Full Circle?

alternative IV estimators (columns V and VI) are virtually identical to the results
of column III.
The preferred estimate presented in Column IV of Table 4 represents a
new estimator, a new specification, and a new instrumentation strategy.6 In
contrast to RS08, we find robust statistically significant empirical support for a
positive aid-growth relationship for the 1970-2000 period. Robustness and
sensitivity tests, documented in Arndt et al. (2010), include application of a
flexible doubly robust estimator, exclusion of (possibly) influential observations,
as well as implementation of sample restrictions and additional covariates. They
show that, despite some sensitivity to different choices, there is no firm basis on
which to reject our main results.

Table 5: Summary of results from model modifications, 1970-2000

Estimator
Instrument Specification
2SLS / IV-LIML IV-IPWLS

RS08 0.10 0.15*

RS08
AJT 0.10 0.10**

RS08 0.22* 0.21*

AJT
AJT 0.25** 0.13***

significance level: * 10%; ** 5%; *** 1%

Notes: AJT refers to our preferred instrument (Table 3, Column V) or specification (Table 4 Column III). Cells show the
estimated coefficient on Aid/GDP from IV regressions involving different combinations of specifications, instruments, and
estimators; column (I) provides estimates from standard IV estimators (2SLS or IV-LIML); column (II) employs the IV-
IPWLS estimator, as described in the text; standard errors on which statistical inference is based are robust to arbitrary
heteroskedasticity; dependent variable is the average real growth rate.

Source: authors’ estimates.

To get a better sense of the individual and joint impact of alternative

combinations of our three main modifications, Table 5 provides a summary of the
various models. Each cell reports the estimated impact of aid on growth over the
1970-2000 period for a given combination of modifications. Reinforcing the
previous results, we find that the RS08 result is not robust. With the exception of
employing the AJT specification alone, all other modifications (either individually
or jointly) yield a significant aid-growth relationship. Moreover, the probability of
falsely rejecting the null hypothesis falls with the number of modifications

6
Note that adding the three initial macroeconomic conditions employed by RS08, which had been
excluded for redundancy, to the models estimated in columns III and IV of Table 4 leaves all
results essentially unchanged; moreover, these three variables continue to be redundant.

19
 Journal of Globalization and Development, Vol. 1 [2010], Iss. 2, Art. 5

employed. Thus, when all three modifications are incorporated, the estimated
coefficient on Aid/GDP of 0.13 becomes significant at the 1% level.
Given our preferred approach employs a single generated instrument, it
remains to be established whether the underlying instrumentation strategy remains
valid in the context of the new set of conditioning variables. Thus, maintaining
the improved specification and IV-LIML estimator, we replace the generated
instrument with different sets of aggregate instruments as per Sections 3.2 and
3.3. These results are reported in columns VI to VIII of Table 1, which
respectively employ a full set of aggregate instruments (used in the RS08 zero
stage), a preferred sub-set of three “least suspect” instruments, and initial log
population size only. The most important result is that the Hansen J statistic is
considerably strengthened when three as opposed to eight aggregate instruments
are used, passing all conventional test thresholds. Moreover, in the last two
columns, the coefficient on Aid/GDP remains positive and significant, once again
supporting our principal results.
We now return to the other periods analyzed in RS08 (1960-2000, 1980-
2000, and 1990-2000), using our preferred specification and instrumentation
strategy as well as both the IV-LIML and IV-IPWLS estimators. Results for each
alternative period are presented in Table 6. For the 1960-2000 period (columns I
and II) both the point estimate and variance of the estimated treatment effect are
squarely in the domain found in Table 4 (columns III and IV) for the 1970-2000
period. The long-run impact of foreign aid comes across as well established. With
respect to the shorter run effects of aid, given in columns III to VI of Table 6, we
cannot reject the hypothesis that the treatment effect is zero. This is confirmed by
the (very weak) relation in the reduced form given by the Stock-Wright S statistic.
The plausible range for the treatment effect is much wider for these periods,
reflected by larger standard errors on the treatment effect. This is most apparent
for 1990-2000 where the standard error on the FDR treatment effect estimate is
almost five times larger than that for the 1960-2000 period. As suggested in
Section 2, a meaningful and robust average short-run effect of aid on growth may
be very difficult to discern from the available empirical data.
At this point, it is helpful to reflect on what exactly the estimated
regression parameters represent. As has been established in the literature,
instrumental variables estimators typically cannot be interpreted as average
treatment effects. This is only appropriate under strong additional assumptions –
in this case, homogeneity across countries in their response to aid. Rather, what is
actually recovered depends on the instruments chosen as well as the extent of
heterogeneity in responses to changes in these instruments. In the case of a single
binary instrument and endogenous response variable, IV estimators often recover
a local average treatment effect (LATE), defined as the average treatment effect
for the sub-population that switches from the control to the treatment group on

DOI: 10.2202/1948-1837.1121 20
 Arndt et al.: Aid, Growth, and Development: Have We Come Full Circle?

account of the switch in the instrument. This can be understood as a weighted

average of the marginal treatment effects for the sub-population that is responsive
to the instrument (Heckman, 2001).

Table 6: Modified regressions, alternative periods

1960-2000 1980-2000 1990-2000
(I) (II) (III) (IV) (V) (VI)
IV-LIML IV-IPWLS IV-LIML IV-IPWLS IV-LIML IV-IPWLS

Aid / GDP 0.16* 0.09** 0.02 0.05 -0.11 0.11

(0.08) (0.04) (0.14) (0.10) (0.19) (0.14)

Initial per cap. GDP -0.67** -0.83*** -1.41*** -1.36*** -0.72 -0.12
(0.31) (0.24) (0.42) (0.35) (0.74) (0.57)
Initial level of policy 1.88*** 2.33*** 2.10*** 1.51 0.65 0.99*
(0.45) (0.42) (0.71) (1.01) (0.53) (0.52)
Initial life expectancy 0.01 0.02 0.06 0.10** 0.12 0.13**
(0.02) (0.03) (0.04) (0.04) (0.07) (0.06)
Geography 0.26 0.21 0.48** 0.33 0.13 0.23
(0.19) (0.17) (0.21) (0.23) (0.41) (0.38)
Coastal pop. density in 1965 0.00*** 0.00*** 0.00* 0.00 0.00 0.01***
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
Primary schooling in 1960 2.76*** 2.13*** 1.45 1.15 -1.38 -2.08
(0.91) (0.74) (1.00) (1.04) (1.93) (1.68)
Malaria risk in 1966 -1.20** -1.03** -0.99 -1.09 -2.36** -2.48**
(0.57) (0.40) (0.79) (0.77) (0.96) (1.02)
Invest. goods price, 1960-64 -0.01** -0.01** 0.00 -0.00 -0.01 -0.01
(0.00) (0.00) (0.00) (0.00) (0.01) (0.01)
Civil liberties in 1972 -0.33 -0.40 -0.34 -0.19 0.67 0.74
(0.26) (0.27) (0.31) (0.35) (0.64) (0.62)
Air distance (log) 0.31 0.28 1.19* 1.27 2.11** 1.78**
(0.42) (0.39) (0.69) (0.82) (0.90) (0.75)

Scale of excluded instrument Binary Continuous Binary Continuous Binary Continuous

N 74 74 75 75 70 69
R-squared 0.71 0.84 0.64 0.60 0.52 0.51
Kleibergen-Paap Wald F stat. 15.82 42.80 19.31 22.01 17.58 20.04
Stock-Wright LM S stat. 4.30 4.30 0.03 0.20 0.30 0.63
(probability) 0.038 0.038 0.869 0.654 0.584 0.428

Notes: the endogenous variable is Aid/GDP, instrumented as per the models in Table 3 and re-estimated from the relevant
bilateral data and time period; initial policy refers to the Sachs-Warner trade policy index; geography refers to the average
of the number of frost days and tropical land area; covariate values also vary according to the period chosen, however this
is not the case with respect to the variables from Sala-i-Martin et al. (2004) as alternative initial values are not available;
intercept and regional dummies not shown; standard errors, given in parentheses, are robust to arbitrary heteroskedasticity;
dependent variable is the average real growth rate.

Source: authors’ estimates.

21
 Journal of Globalization and Development, Vol. 1 [2010], Iss. 2, Art. 5

In the present case, the complex mapping from multiple instruments to a

continuous endogenous variable significantly complicates a LATE-type
interpretation of the estimates. Furthermore, given the small sample sizes
involved, distinguishing between key sub-populations and important observations
is exceedingly difficult operationally. Nevertheless, in light of the instruments
used, a possible interpretation is that the estimates reflect the growth response to
aid for countries whose total aid inflows have been most influenced by differences
in relative population sizes or ever having been a colony. These drivers typically
would be associated with political rather than altruistic motivations. If politically
motivated aid is less effective, this may bias downward the estimated aid
coefficient. Finally, one notes that moving from the continuous instrument to the
binary instrument (used in the doubly robust estimations) has a relatively small
effect on the estimated parameter of interest. This is consistent with the
observation made in Arndt et al. (2010) that the results are not dependent on the
specific threshold chosen to define the binary instrument.
Lastly, we consider how this evidence stacks up against theory. In all the
regression specifications reported so far, tests have been made against a null
hypothesis of a zero relationship between the explanatory and dependent
variables. Nevertheless, it is straightforward to calculate the t-statistic for testing
whether the estimates differ from the theoretical prior that the point estimate for
the long-run effect of foreign aid on growth is around 0.1 (see Section 2.4). With
the exception of the OLS regression reported in column I of Table 1, all point
estimates for the aid growth relationship across all regressions reported in Tables
1, 4, 5, and 6 are not significantly different from 0.1.7 To put it differently, there is
no basis on which to reject the theoretical prior that aid has a positive long-run
effect on growth.

5. Conclusion
To conclude, we respond to the question posed in the title to this paper: has the
aid, growth, and development literature gone full circle? Our answer is “no”.
While in the most recent literature the pendulum has swung to deep skepticism
concerning the ability of aid to contribute to economic growth, a series of
important points of consensus have emerged. First, methodological advances in
the program evaluation literature have improved the profession’s capacity to
identify causal effects in economic phenomena. These advances are beginning to
be applied at the more aggregate level, as pursued here. Second, methodological
advances also highlight the serious challenges that must be surmounted in order to
7
For these regressions, the null hypothesis that the coefficient of the estimated aid-growth
parameter is equal to 0.1 is, in fact, never close to being rejected, including in RS08’s original IV
specification (Table 1, Column II).

DOI: 10.2202/1948-1837.1121 22
 Arndt et al.: Aid, Growth, and Development: Have We Come Full Circle?

derive robust causal conclusions from non-experimental data. In many important

areas of inquiry, long-standing debates with respect to causal impacts persist
despite improved methods and improved data availability. Third, the formation of
reasonable expectations about the likely returns to foreign assistance has been
greatly facilitated by the application of growth theory. Finally, there is increasing
recognition that many of the key interventions pursued by foreign aid will only
result in positive growth outcomes over long time horizons.
In line with Temple (2010), we started by replicating RS08. Based on a
detailed analysis of their approach, we subsequently developed a better
instrumentation strategy, an improved specification, and a preferred estimator.
The improved specification contains a fuller set of regional fixed effects and
indicators of initial human capital and geographic conditions. These were drawn
from theory and previous research. They included primary schooling, coastal
population density, and malaria risks. Consistent with best practice in the program
evaluation literature, we excluded covariates, such as revolutions and institutional
performance, which represent potential channels through which aid affects
growth. With respect to the zero stage instrumentation, we (i) excluded suspect
variables; (ii) corrected errors in the implementation of the RS08 instrumentation
strategy; (iii) employed aid per capita in place of Aid/GDP to preclude spurious
correlation with the chosen instruments; (iv) introduced donor-specific fixed
effects; and (v) accounted for selection bias through a Heckman correction.
Finally, we deployed robust regression estimators which adjust for heterogeneity
across countries. This involved introducing a new doubly robust estimator that
can be used in instrumental variable contexts. A variety of robustness and validity
checks, including of the underlying instrument, provide support to our approach.
Overall, we believe our study represents the most carefully developed
empirical strategy employed in the aid-growth literature to date. Our results
provide solid support for the view that the effect of aid on growth is positive in
both the 1970-2000 and 1960-2000 periods. The preferred doubly robust
estimator places the point estimate of the long run elasticity of growth with
respect to the share of aid in recipient GDP at 0.13 (IV-IPWLS). This is below the
unweighted point estimate of 0.25 (IV-LIML). In sum, our findings suggest that
an inflow on the order of 10 percent of GDP spurs the per capita growth rate by
more than one percentage point per annum in the long run. These estimates are
consistent with the view that foreign aid stimulates aggregate investment and may
also contribute to productivity growth, despite some fraction of aid being
allocated to consumption. The 95% confidence interval around our estimates lies
in the strictly positive domain and contains the prior, suggested in RS08 that the
long-run elasticity of growth to foreign aid should be around 0.1. In the shorter
term, our analysis indicates that the impact of aid is difficult to discern.
Nevertheless, when the longer run macro evidence is combined with the evidence

23
 Journal of Globalization and Development, Vol. 1 [2010], Iss. 2, Art. 5

at the micro- and meso-levels, a consistent case for aid effectiveness emerges.
There is no micro-macro paradox.
We find ourselves in a similar position to Winters (2004) in his review of
the implications of trade liberalization for growth. While he concludes that trade
liberalization stimulates growth over the long run and on average, he adds that
“For a variety of reasons, the level of proof remains a little less than one might
wish but the preponderance of evidence certainly favors that conclusion” (2004:
F18). Similarly, we conclude that the bleak pessimism of much of the recent aid-
growth literature is unjustified and the associated policy implications drawn from
this literature are often inappropriate and unhelpful. Aid has been and remains an
important tool for enhancing the development prospects of poor nations.
Finally, the complex and idiosyncratic process of managing aid to spark
and sustain growth is subject to considerable learning. Nearly all participants in
the aid-growth debate, not least these authors, recognize the potential for aid to do
better, particularly in fostering productivity growth. Abolishing foreign aid, or
drastically cutting it back, would be a mistake and is not warranted by any
reasonable interpretation of the evidence. The challenge is to improve foreign
assistance effectiveness so that living standards in poor countries are substantially
advanced over the next three decades.

References

Acemoglu, D. and S. Johnson. 2007. “Disease and Development: The Effect of

Life Expectancy on Economic Growth.” Journal of Political Economy 115
(6): 925-985.
Alesina, A., and D. Dollar. 2000. “Who Gives Foreign Aid to Whom and Why?”
Journal of Economic Growth 5 (1): 33-63.
Angrist, J.D., and J.-S. Pischke. 2010. “The credibility revolution in empirical
economics: How better research design is taking the con out of
econometrics.” Journal of Economic Perspectives 24(2): 3-30.
Angrist, J.D., and J.-S. Pischke. 2008. Mostly Harmless Econometrics: An
Empiricist’s Companion. Princeton: Princeton University Press.
Arndt, C., S. Jones and F. Tarp. 2010. “Aid, Growth, and Development: Have We
Come Full Circle?”. UNU-WIDER Working Paper 96.
Ashraf, Q.H., A. Lester, and D.N. Weil. 2008. “When Does Improving Health
Raise GDP?”. NBER Working Paper 14449, National Bureau of Economic
Research, Inc.
Banerjee, A.V. and E. Duflo. 2009. “The Experimental Approach to Development
Economics”, Annual Review of Economics 1: 151-178.

DOI: 10.2202/1948-1837.1121 24
 Arndt et al.: Aid, Growth, and Development: Have We Come Full Circle?

Baum, C.F., M.E. Schaffer, and S. Stillman. 2007. “Enhanced Routines for
Instrumental Variables/generalized Method of Moments Estimation and
Testing.” Stata Journal 7 (4): 465-506.
Bun, M., and F. Windmeijer. 2010. “The Weak Instrument Problem of the System
GMM Estimator in Dynamic Panel Data Models.” Econometrics Journal
13(1): 95-126.
Burnside, C., and D. Dollar. 2000. “Aid, Policies, and Growth.” American
Economic Review 90 (4): 847-868.
Card, D. 2001. “Estimating the Return to Schooling: Progress on Some Persistent
Econometric Problems.” Econometrica 69 (5): 1127-60.
Clemens, M., and S. Bazzi. 2009. “Blunt Instruments: On Establishing the Causes
of Economic Growth.” Center for Global Development Working Paper
171.
Cohen, D. and M. Soto. 2007. “Growth and Human Capital: Good Data, Good
Results.” Journal of Economic Growth 12: 51-76.
Collier, P., and A. Hoeffler. 2004. “Aid, Policy and Growth in Post-conflict
Societies.” European Economic Review 48 (5): 1125-1145.
Dalgaard, C.-J,. and L. Erickson. 2009. “Reasonable Expectations and the First
Millennium Development Goal: How Much Can Aid Achieve?” World
Development 37 (7): 1170-1181.
Dalgaard, C.-J., H. Hansen, and F. Tarp. 2004: “On The Empirics of Foreign Aid
and Growth.” Economic Journal 114 (496): F191-F216.
Djankov, S., J. Montalvo, and M. Reynal-Querol. 2008. “The Curse of Aid.”
Journal of Economic Growth 13 (3): 169-194.
Easterly, W. 1999. “The Ghost of Financing Gap: Testing the Growth Model
Used in the International Financial Institutions.” Journal of Development
Economics 60 (2): 423-438.
Easterly, W., R. Levine, and D. Roodman. 2004. “Aid, Policies, and Growth:
Comment.” American Economic Review 94 (3): 774-780.
Hansen, H., and F. Tarp. 2000. “Aid Effectiveness Disputed.” Journal of
International Development 12 (3): 375-398.
Hansen, H., and F. Tarp. 2001. “Aid and Growth Regressions.” Journal of
Development Economics 64 (2): 547-570.
Hauk, W., and R. Wacziarg, R. 2009. “A Monte Carlo Study of Growth
Regressions.” Journal of Economic Growth 14 (2): 103-147.
Heckman, J.J. 1979. “Sample Selection Bias as a Specification Error.”
Econometrica 47 (1), 153-61.
Heckman, J.J. 2001. “Micro Data, Heterogeneity, and the Evaluation of Public
Policy: Nobel Lecture." Journal of Political Economy 109(4): 673-748
IDA15. 2008. “Additions to IDA Resources: Fifteenth Replenishment.” Technical
report, International Development Association.

25
 Journal of Globalization and Development, Vol. 1 [2010], Iss. 2, Art. 5

Imbens, G.W. 2004. “Nonparametric Estimation of Average Treatment Effects

Under Exogeneity: A Review.” The Review of Economics and Statistics 86
(1): 4-29.
Imbens, G.W., and J.M. Wooldridge. 2009. “Recent Developments in the
Econometrics of Program Evaluation.” Journal of Economic Literature 47
(1): 5-86.
Kronmal, R.A. 1993. “Spurious Correlation and the Fallacy of the Ratio Standard
Revisited.” Journal of the Royal Statistical Society. Series A (Statistics in
Society) 156 (3): 379-392.
La Porta, R., F.L. de Silanes, and A. Shleifer, A. 2008. “The Economic
Consequences of Legal Origins.” Journal of Economic Literature 46 (2):
285-332.
Mankiw, N.G., D. Romer, and D.N. Weil. 1992. “A Contribution to the Empirics
of Economic Growth.” The Quarterly Journal of Economics 107 (2): 407-
37.
Masud, N. and B. Yontcheva. 2005. “Does Foreign Aid Reduce Poverty?
Empirical Evidence from Nongovernmental and Bilateral Aid.”
International Monetary Fund Working Paper 05/100, International
Monetary Fund.
Mishra, P., and D.L. Newhouse. 2007. “Health Aid and Infant Mortality.”
International Monetary Fund Working Paper 07/100, International
Monetary Fund.
Moral-Benito, E. 2009. "Determinants of Economic Growth: A Bayesian Panel
Data Approach." Policy Research Working Paper Series 4830, The World
Bank.
Mosley, P. 1987. Overseas Aid: Its Defence and Reform. Brighton: Wheatshead
Books.
Moyo, D. 2009. Dead Aid: Why Aid Is Not Working and How There Is a Better
Way for Africa. London: Allen Lane.
Nunn, N. 2008. “The Long-Term Effects of Africa's Slave Trades.” The Quarterly
Journal of Economics 123 (1): 139-176.
Papanek, G.F. 1972. “The Effect of Aid and Other Resource Transfers on Savings
and Growth in Less Developed Countries.” Economic Journal 82 (327):
935-50.
Papanek, G.F. 1973. “Aid, Foreign Private Investment, Savings, and Growth in
Less Developed Countries.” Journal of Political Economy 81 (1): 120-30.
Rajan, R., and A. Subramanian. 2007. “Does Aid Affect Governance?” AEA
Papers and Proceedings, 97(2): 322-327.
Rajan, R.G., and A. Subramanian. 2008. “Aid and Growth: What Does the Cross-
Country Evidence Really Show?” The Review of Economics and Statistics
90 (4): 643-665.

DOI: 10.2202/1948-1837.1121 26
 Arndt et al.: Aid, Growth, and Development: Have We Come Full Circle?

Robins, J.M., and A. Rotnitzky, A. 1995. “Semiparametric Efficiency in

Multivariate Regression Models with Missing Data.” Journal of the
American Statistical Association 90 (429): 122--129.
Roodman, D. 2007. “The Anarchy of Numbers: Aid, Development, and Cross-
Country Empirics.” World Bank Economic Review 21 (2): 255-277.
Roodman, D., and J. Morduch. 2009. “The Impact of Microcredit on the Poor in
Bangladesh: Revisiting the Evidence.” Center for Global Development
Working Paper 174
Roodman, D. 2009. “A Note on the Theme of Too Many Instruments.” Oxford
Bulletin of Economics and Statistics 71 (1): 135-158.
Sachs, J. 2005. The End of Poverty: Economic Possibilities for Our Time, Penguin
Press: New York.
Sala-i-Martin, X., G. Doppelhofer, and R.I. Miller. 2004. “Determinants of Long-
Term Growth: A Bayesian Averaging of Classical Estimates (BACE)
Approach.” American Economic Review 94 (4): 813-835.
Tarp, F., C.F. Bach, H. Hansen, and S. Baunsgaard. 1999. “Danish Aid Policy:
Theory and Empirical Evidence”. Chapter 9 (pp. 149-69) in Kanhaya
Gupta (ed.) Foreign Aid: New Perspectives. Norwell, MA: Kluwer
Academic Publishers.
Temple, J.R.W. 2010. “Aid and Conditionality.” In Dani Rodrik and Mark
Rosenzweig (eds.), Handbook of Development Economics. 5: 4415-4523.
Winters, L.A. 2004. “Trade Liberalisation and Economic Performance: An
Overview.” Economic Journal 114 (493): F4-F21.
Wooldridge, J.M. 2005. “Violating Ignorability of Treatment By Controlling For
Too Many Factors.” Econometric Theory 21 (05): 1026-1028.
World Bank. 2008. Annual Review of Development Effectiveness 2008: Shared
Global Challenges, Independent Evaluation Group, available at:
http://go.worldbank.org/U2T30HQKG0

27