Foreign aid fiscal policy: theory and evidence.
Asongu, Simplice ; Jellal, Mohamed
INTRODUCTION
The issue of whether development assistance improves growth in
recipient countries can be traced back to the two-gap model (Chenery and
Strout, 1966), which to the best of our knowledge remains the most
influential theoretical underpinning in the literature on aid
effectiveness. According to the narrative, developing countries face
serious constraints in savings and 'export earnings' that are
not conducive for the growth of investment. In spite of severe
criticisms since its inception, this model has provided a background for
early aid policies (Easterly, 1999) and empirical specifications in many
studies (Masud and Yontcheva, 2005). Accordingly, both the Harrod-Domar
and Solow growth models that constitute the principal theoretical
underpinnings in the foreign aid literature are based on the need for
substantial aid-driven investment, with the purpose of reducing the
poverty gap between developed and poor countries.
The effect of development assistance on private sector investment
has long been an important issue of debate. Many economists have adopted
the position that aid stimulates private investment in least developed
countries (LDCs) by improving macroeconomic savings, while others have
contended that aid has a negative effect on private investment because,
inter alia, it: is often wasted or counterproductive; generates the
Dutch-disease and enables the central government to drain resources from
the private sector (Snyder, 1996). (1) However, recent empirical
evidence suggests that donors are concerned about how their aid is used,
especially how it affects the fiscal behavior of recipient governments
(Morrissey, 2012). Morrissey has reviewed the effects of aid and
concluded that aid significantly affects government spending and tax
effort in LDCs.
Our main contribution to the literature is twofold. On the one
hand, we propose an endogenous theory of aid and on the other hand
provide empirical validity for the proposed theory. The model we propose
postulates that the positive effect of aid reduces that burden of the
taxation system on the private sector, which ultimately leads to
economic growth in poor countries, especially when the amount of aid is
high and the public sector is less effective. In essence, the goal of
the study is to examine how aid affects private investment through
fiscal policy channels. We postulate that the effects of aid on tax
effort and government spending as suggested by Morrissey (2012) could
provide incentives for private investments and fixed capital formation,
which are essential for economic prosperity.
In addition to the above contributions, the paper has policy
implications in a number of areas. First, the global economic downturn
has resurfaced issues about donors': continued willingness 'to
give' and commitment to development assistance (Ahmed et al, 2011).
Therefore, investigating the effect of aid on investment could provide
additional insights into the ongoing debate. (2)
Second, a corollary of the first contribution is the shifting of
policy-space to foreign aid alternatives from East Asia. Accordingly,
the ability to learn from the East Asian success stories has been
substantially hampered by an asymmetric bargaining power between Africa
and her Western development partners. (3) Third, there have been
considerable shifts in the objectives announced by the donor community,
which have evolved from intensive industrialization programs advocated
in the 1950s to more recent poverty-reduction objectives such the
Millennium Development Goals (MDGs). Fourth, by using comparatively more
recent data (1996-2010) from 53 countries, we provide an updated account
of the nexuses. Moreover, the richness of our dataset also avails room
for more policy implications. Accordingly, in order to add subtlety to
the analysis, we disaggregate the dataset into fundamental
characteristics of investment (legal origins, petroleum-exporting
quality, political instability/conflicts, regional proximity,
income-levels, religious-domination and openness to sea).
The rest of the paper is organized as follows. The next section
presents controversial views in the literature before proposing the
endogenous theory. The data and methodology are discussed in after that.
The empirical analysis is covered by the subsequent section. The last
section concludes.
FOREIGN AID AND DEVELOPMENT
Consistent with Asongu (2014a), the Official Development Assistance
(ODA) programs that were instituted over five decades ago have led to
widely debated and unsolved issues surrounding aid effectiveness. In
2005, Western countries devoted substantial efforts to save Africa. In
July of this year, the Group of Eight (G8) agreed to double development
assistance to Africa from $25 billion a year to $50 billion in order to
finance the 'Big push', as well as cancel Africa's
aid-loans contracted during previous attempts at a 'Big push'.
According to most estimates, prior to this effort, Africa was already
the most aid-intensive region in the world. World leaders gathered at
the United Nations in September 2005 to further discuss progress towards
mitigating poverty on the continent. As far as we have reviewed,
Easterly (2005a) best highlights some frustrating statistics.
Accordingly, sub-Saharan Africa (SSA) contains more than 11 % of the
world's population but only accounts for 1 % of the world's
Gross Domestic Product (GDP). In the median African country, 43% of the
population survives on less than $1 per day. On the list of the World
Food Program, of the 23 countries with more than 35% of malnourished
population, 17 (73%) are in Africa. Poverty has been sustained by the
long and brutal civil wars in many countries (Angola, Chad, Sierra
Leone, Somalia, Liberia ... etc), Rwanda's genocide and recent
carnages in Darfur-Sudan: with the Democratic Republic of Congo
registering the world's highest casualties since World War 2. To
put these stylized facts into greater perspective, eight of the 11
recent cases of total societal breakdown into anarchy have been in
Africa, namely in: Angola, Burundi, Liberia, Sudan, Sierra Leone,
Somalia, the Democratic Republic of Congo and Libya (beside Afghanistan,
Iraq and Syria). As a means of reconstructing these war-torn countries,
foreign aid would obviously be considered as a 'Big economic
push'.
Controversial views in the literature
While development assistance is necessary in the short-run owing to
associated precarious circumstances (eg humanitarian concerns), there
has been a heated debate on the effectiveness of aid on the one hand and
linkages between aid, conditionality (4) and economic policies in
recipient countries on the other hand. In international policy
coordination, one of the most debated and controversial issue is foreign
aid. A strand of protagonists has engaged the debate with a mixture of
alleged altruism, economic interests, geo-strategic considerations and
historical ties (Alam, 2004). The post-decolonization period has been
characterized by substantial increase in grants and soft loans from
Western donor agencies and governments (Oya, 2006). In essence, the Cold
war and the battle for geopolitical control of Africa between
superpowers are considered by many scholars as the most important
determinants of foreign aid, which increased sharply in the 1980s
(Degnbol-Martinussen and Engberg-Pedersen, 2003). The debate has also
been extended to policies by the International Monetary Fund (IMF). (5)
We will now discuss the major strands of the debate on the
development outcomes of foreign aid. A substantial bulk of the
literature has been devoted to the macroeconomic consequences of
development assistance.
However, mixed results have been reported and studies that have
concluded on a significant and positive effect have faced heavy
methodological criticisms. Inconclusive results with recently refined
methodologies, heavy reliance on empirical evidence and the absence of
analytical frameworks (Masud and Yontcheva, 2005), have left much room
for debate on the aid-development nexus. Table 1 summarizes the debate
in two main strands. Whereas the first strand acknowledges the positive
sides of development assistance, the second sustains the negative
consequences of aid.
Among studies in the first strand, we shall highlight that of
Burnside and Dollar (2000), which has concluded that aid could be
effective when policies are appealing (conducive). The Burnside and
Dollar study has received abundant comments from scholars and policy
makers (Guillaumont and Chauvet, 2001; Collier and Dehn, 2001; Easterly
et al, 2003) with some claiming that corresponding findings are
extremely data-dependent (Clemens et al, 2004). Whereas Clemens et al.
(2004) have established that aid is beneficial in the short term, Minou
and Reddy (2010) have recently found that the beneficial effect could
also be in the long term. Gomanee et al. (2003) have emphasized that
development assistance has both a direct effect on welfare and an
indirect impact through public spending on social services. The indirect
stance has been further consolidated by Mosley et al (2004) on
well-being and poverty in recipient countries. Development assistance
has also been found to promote institutions in terms of its role on
corruption (Okada and Samreth, 2012) and transition to democracy
(Resnick, 2012).
The second strand is research that finds an insignificant effect of
aid on investment, savings and institutions. For example, it concludes
that aid promotes unproductive public consumption (Mosley et al, 1992)
without a positive effect on investment. The latter stance has been
sustained by Reichel (1995) and Boone (1996). Whereas Ghura et al (1995)
has emphasized the negative impact of development assistance on domestic
savings, Pedersen (1996) has established that foreign aid distorts
development and leads to aid-dependency. In direct response to the Okada
and Samreth (2012) position on a negative aid-corruption nexus, recent
African aid literature has supported this second strand from an
institutional standpoint. Accordingly, Asongu (2012a, 2013a) has engaged
in a debate on the 'effect of foreign aid on corruption'. (6)
It is also important to devote space to engaging some studies on
the development outcomes of foreign aid that have established
alternative conclusions to the two main strands summarized in Table 1.
These include research on the impact of aid on growth and changes in
recipient policies. Hansen and Tarp (2001) conclude that whereas the
effect of development assistance on economic growth may not be
contingent on 'good policy', human capital could be the
driving factor behind economic prosperity. The narrative is in
accordance with a recent strand of literature on soft economics (ie the
human side of economic activities) (Kuada, 2015) and knowledge economy
(Asongu, 2015a; Tchamyou, 2015; Asongu and Tchamyou, 2016). Furthermore,
the emphasis on human capital is consistent with another recent stream
of African development literature on the benefits of foreign aid in
economic growth (Kargbo and Sen, 2014; Gyimah-Brempong and Racine, 2014)
because the corresponding positive impact on economic growth is more
apparent when development assistance is channeled through educational
mechanisms (Asiedu and Nandwa, 2007; Asiedu, 2014).
Easterly (2003) has criticized the Burnside and Dollar (2000) model
by establishing that while foreign aid may stimulate growth when correct
policies are implemented, the data show that the linkage between aid and
recipients' policies is weak. The issue of exclusive growth in
Africa, which has motivated a book by Kuada (2015), has also been the
motivation behind another book by Fosu (2015a, b), which is devoted to
elucidating: (i) myths behind Africa's recent growth resurgence and
(ii) the role of institutions in the underlying growth resurgence. The
concern about institutions is important because Brautigam and Knack
(2004) have concluded that high aid to Africa is linked to deteriorating
governance and tax levels. The conclusions of Brautigam and Knack (2004)
on weak governance and low tax income are, respectively, in accordance
with Asongu and Nwachukwu (2016) and Asongu (2015b) who have used more
updated data. According to Brautigam and Knack (2004), growth in GDP per
capita is more linked to improvements in governance, as opposed to
foreign aid.
Whereas the effect of development assistance is more straight
forward to some scholars (Ishfaq, 2004; Addison et al, 2005; Fielding et
al, 2006), (7) its impact on development outcomes may also be indirect.
We have highlighted in one of the strands above that aid promotes
unsound public consumption (Mosley et al, 1992) without a positive
effect on investment. We have also highlighted in the introduction that
aid affects development objectives through fiscal behavior channels
(Morrissey, 2012). Therefore 'aid effects' on tax effort and
government spending could provide incentives for the investment needed
for economic prosperity.
Theoretical proposition: fiscal behavior as a transmission
mechanism
Theoretical and empirical underpinnings
The theoretical underpinnings of the fiscal behavior channel in the
aid-development nexus are broadly consistent with the
'Big-Push' model, which maintains that Africa is poor because
it is stuck in a poverty trap (Easterly, 2005a). In order to emerge from
the poverty pit, it needs a substantial aid-driven investment policy: a
'Big Push'. Both the Harrod-Domar and Solow growth models have
been based on this intuition. Accordingly, the underlying assumption for
the intuition is that, the 'Big-Push' is destined to bridge
the saving-investment gap poor countries face (Rostow, 1960; Chenery and
Strout, 1966; Easterly, 2005a).
In light of the summary of the literature presented in the section
'Controversial views in the literature', our model will
incorporate two channels for the influence of foreign aid: the
investment destination of aid and the fiscal behavior mechanism as a
channel to the investment. Hence, the goal of the present study is to:
propose an endogenous theory of aid and test the empirical validity of
the proposed theory. In essence, we examine how aid affects private
investment (and gross fixed capital formation) through tax efforts and
government spending. The model is primarily based on the assumption that
private investment and/or gross fixed capital formation are relevant for
economic prosperity.
Theoretical proposition: extension of Barro (1990)
There is a wealth of literature substantiating that the taxation
system adopted by a developing country creates large distortions that
substantially affect the dynamics of the private sector and hence
economic growth and development (Manly et al, 2006; Feredeand and
Dahlby, 2012). As highlighted earlier, some of this vast literature has
focused on the channels via which foreign aid affects economic
prosperity in recipient countries (see Table 1). In the same vein,
recent endogenous growth literature has elucidated the positive role of
public spending, notably, in: education, health and infrastructure for
economic growth (Alexiou, 2009). The underlying literature substantially
draws from the Barro (1990) model.
In essence, Barro determines the optimal size of the State: public
expenditure that maximizes the rate of economic growth. The simple
growth model does not take into account the issue of budget deficit
allocated to public spending. Hence, it is intuitively relevant to
propose a model that incorporates development assistance destined to
financing productive public expenditure. Therefore, the idea here is to
extend Barro's simple growth model while taking into consideration
the effect of foreign aid on private investment through the fiscal
behavior of the State. From Barro's theoretical underpinnings, we
suppose that productive investments may either be private investments or
gross fixed capital formations that ultimately have positive effects on
economic growth.
We consider a model similar to Barro (1990). The economy is
characterized by the decision of a household representative agent who is
a consumer and a producer with the following production function:
y = [Ak.sup.1-[alpha]][g.sup.[alpha]] (1)
where k is physical capital, g the amount of composite productive
public expenditure including: education, infrastructure and health. This
public expenditure is financed by taxes and an allocation to foreign
aid. That is:
g = [tau]y + A (2)
where A is the amount of international aid that is indexed on
national income and we suppose that it is determined in an exogenous
manner.
For the purpose of simplicity, we further assume that the budget of
the State is at equilibrium at every moment. Accordingly, the problem of
our representative agent is to solve the dynamic program of
decentralized economic growth given by:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (3)
Subject to :
[??](t) = (1 - [tau])y(f)-c(t)
g(t) = [tau]y(t) + A(t)
A(t) = ay(t)
k(0)>0.
where c(*) represents per capita consumption, [sigma] is the
constant inter-temporal elasticity of substitution, [rho] is the
constant rate of time preference, and a is the indexation rate of
foreign aid allocated to the production of social infrastructure g(*).
This rate is exogenous, fixed and considered as 'given' by
national economic agents.
We have already seen that a substantial bulk of the literature has
focused on the effect of aid on growth and development. The theoretical
and empirical relevance of aid to public spending has also been shown.
Now we suppose that the objective of donor(s) vis-a-vis poor countries
is the development of the private sector (liberal aspect of the
contract). Hence, its (their) aid is supposed to be entirely and
observably allocated directly to the financing of productive public
spending, which can be lacking in poor countries. Hence, the role of aid
is to provide socio-economic infrastructure that improves private sector
effectiveness. Within this framework, it can be established that the
equation for budget equilibrium is given this time by:
[tau]y(t) + ay(t) = g(t) [??]
g(t) = ([tau]+a)y(t) (4)
In the presence of foreign aid allocated for private sector
promotion, while acknowledging that aid as an exogenous factor, public
decision makers should therefore implement an endogenous economic growth
program by the optimal choice of the income-related direct tax
structure. Hence, taking the government's decisions as given, the
representative agent chooses consumption c, and capital k, to maximize
his/her welfare:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. (5)
Subject to :
[??](t) =(1 - [tau])y(t)-c(t) g(t) = ([tau]+a)y(t) [kappa](0)>0.
Proposition 1: In the presence of foreign aid:
(i) the economic growth rate is given by the following rate:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. (6)
(ii) The tax rate that maximizes national economic growth is
therefore given by:
[alpha]-a(1-[alpha]) = [[tau].sup.*]>0 [??] [alpha] > a/1 + a
(7)
It is immediately observable that the positive effect of aid
reduces that burden of the taxation system on the private sector of poor
countries, especially when the amount of aid is high and the public
sector less effective.
Hence, it is apparent that aid granted to developing countries
directly benefits them in terms of private sector dynamism, which
ultimately leads to economic growth while reducing the size of the
national public sector (Remmer, 2004; Payne and Kumazawa, 2005).
DATA AND METHODOLOGY
Data
We examine a panel of 53 African countries with data from African
Development Indicators (ADI) of the World Bank for the period 1996-2010.
Limitation to the time span is motivated by the interest of providing
results with updated and more focused policy implications. Moreover, the
focus on Africa and the time span enable follow-up of a recent foreign
aid debate that has had some influence in academic and policy making
circles. (8) The dependent variables are private investment and gross
fixed capital formation. While the former is used in baseline
regressions, the latter is employed for robustness checks.
Determination of fundamental characteristics
It is important to discuss the determination of fundamental
characteristics that are crucial for the relevance of the empirics.
Macroeconomic characteristics have the limitation of being time-dynamic.
Thus, the same non-dummy threshold may not be consistent over time. This
justification is even more relevant when short-run (business cycle)
disturbances loom substantially. Hence, we are consistent with recent
comparative literature in categorizing countries in terms of
conflict-affected (or political instability), petroleum-exporting, legal
origins, income-levels, regional proximity, religious-domination and
landlockedness (Weeks, 2012; Asongu, 2014b). From intuition, foreign
aid, private investment and fiscal policy substantially depend on the
above categories.
It is difficult to establish an objective definition of a conflict
country. Since few countries on the continent are completely
conflict-free, the distinction is made on the basis of degree of
significance of conflict-span, relative to the period of study. Based on
the information (53 countries over the period 1996-2010), two categories
emerge; civil wars and political strife. With respect to the first
category on civil wars, few would object to the inclusion of Angola
(1975-2002), Burundi (1993-2005), Chad (2005-2010), the Central African
Republic (series of failed coup d'etats between 1996-2003 and the
2004-2007 Bush War), the Democratic Republic of Congo, Cote
d'Ivoire (1999 coup d'etat, 2002-2007 civil war, rekindled in
2011), Liberia (1999-2003), Sierra Leone (1991-2002), Somalia and Sudan.
For the second category, in spite of the absence of some formal
characteristics of civil war, we also include Nigeria and Zimbabwe due
to the severity of their internal strife.
Second, on how to determine petroleum countries, a critical
categorical objection arises because some petroleum countries also
clearly qualify as conflict-affected (Angola and Sudan for instance). In
this study a country may fall into many categories if it has the
relevant categorical characteristics. Another concern that emerges is
arbitrariness if a country qualifies for only part of the time period,
either because of: a recent discovery of oil fields or a substantial
decline in production. In the same vein, another objection could be that
some resource-rich countries (eg Botswana) display macroeconomic
features that are similar to those of petroleum-exporting countries
because of intensive extractive industries. We take a minimalistic
approach to the issue by strictly adhering to the petroleum category and
including only countries whose exports have been oil-dominated for over
a decade during the span 1996-2010. These include: Algeria, Angola,
Cameroon, Chad, Congo Republic, Equatorial Guinea, Gabon, Libya, Nigeria
and Sudan.
Third, the basis of legal origin is founded on the premise that
legal origins place different emphasis on private property rights
vis-a-vis State power (La Porta et al, 1998, 1999). According to this
narrative, English common law countries place more emphasis on private
property rights, whereas French civil law focuses more on State power.
The intuition for this category as discussed in prior work accords with
African institutional quality (Asongu, 2015d) and property rights
(Asongu, 2015c) literature. The underlying logic for this segmentation
is that the institutional web of formal rules, informal norms and
enforcement characteristics affect the climate of investment. The legal
origin classification is according to La Porta et al. (2008, p. 289).
Fourth, the basis for including income-levels to examine
wealth-effects is founded on two premises. On the one hand, economic
prosperity could be associated with higher levels of private investment.
On the other hand, recent African institutional literature has shown
that wealth-effects matter in institutional quality (Asongu, 2012b,
2013b) that ultimately determines investment. The choice of
income-levels is in accordance with the Financial Development and
Structure Database of the World Bank.
Fifth, there is an investment cost of being landlocked (Arvis et
al, 2007). Moreover, in order to add subtlety to the analysis for more
policy implications, we include: (i) religious dominations (Christianity
and Islam) in accordance with the Central Intelligence Agency's
(2011) World Fact book and (ii) regional proximity consisting of SSA and
North African countries.
Endogenous explaining, instrumental and control variables
The fiscal policy measures of government expenditure and tax
revenues are consistent with the discussed literature. The instrumental
variables (IVs) include: Total Net Official Development Assistance
(NODA), NODA from Multilateral Donors (MD), NODA from the Development
Assistance Committee (DAC) countries and Grants excluding technical
cooperation. We employ only two control variables due to constraints in
degrees of freedom required for the Sargan over-identifying restrictions
(OIR) test for instrument validity. (9) The control variables are
corruption and 'voice and accountability' and are included to
reduce the degree of identification when development assistance
instruments are not valid. The choice of the control variables from ADI
of the World Bank is consistent with recent African institutional
literature (Asongu, 2012a, 2013a). These institutional variables are
determinants of a country's investment climate.
Variable definitions and the summary statistics are provided in
Table 2, whereas the categorization of countries is disclosed in the
Appendix.
Methodology
The study uses a Two-Stage Least Squares (2SLS) IV estimation
strategy for a twofold reason: the empirical strategy is consistent with
the problem statement and also addresses the issue of endogeneity. The
adopted IV procedure is in accordance with recent foreign aid (Asongu
and Nwachukwu, 2016) and development (Tchamyou, 2015) literature. The
purpose of adopting an IV approach is to have some bite on endogeneity.
Moreover, the line of inquiry is consistent with an IV technique
essentially because, the study aims to assess how foreign aid
instruments affect investment through mechanisms of fiscal behavior. The
following steps are adopted in the estimation procedure.
First-stage regression:
[FB.sub.it] = [[gamma].sub.0] + [[gamma].sub.1]
[(Instruments).sub.it] + [v.sub.it] (8)
Second-stage regression:
[Investment.sub.it] = [[beta].sub.0] + [[beta].sub.1] [(FB).sub.it]
+ [[beta].sub.2] [X.sub.it[ + [[mu].sub.it] (9)
In equation 9, X is a set of control variables, which include:
Corruption and 'voice and accountability'. FB entails Fiscal
behavior, which consists of Government's final consumption
expenditure and Tax revenues. Investment denotes Private investment and
Fixed capital formation. IVs include: Total NODA, NODA from DAC
countries, NODA from MD and Grants. In equations 8 and 9, v and u,
respectively represent the error terms.
In the estimation process, three main steps are adopted. First, we
justify the choice of the IV procedure with a Hausman test for
endogeneity. Then, we verify that the instruments are exogenous to the
endogenous components of the independent variables (government
expenditure and tax revenues). Last, we ensure that the instruments are
valid and uncorrelated with error term in the equation of interest with
an OIR test. Further robustness checks are ensured with: (i) restricted
and unrestricted modeling; (ii) modeling with robust Heteroscedasticity
and Autocorrelation Consistent (HAC) standard errors and (iii) use of
two investment indicators.
As highlighted in the 'Data' section, we employ only two
control variables due to constraints in degrees of freedom required for
the Sargan OIR test for instrument validity. The Sargan OIR test is only
applicable in the presence of over-identification. In other words, the
instruments must be higher than the endogenous explaining variables by
at least one degree of freedom. In the cases of exact-identification
(instruments equal to endogenous explaining variables) and
under-identification (instruments less than endogenous explaining
variables) the OIR test is by definition impossible. Accordingly, we
have four foreign aid instruments and cannot model with more than three
endogenous explaining variables.
EMPIRICAL ANALYSIS
Presentation of results
In this section, we aim to assess two main issues: (i) the ability
of the exogenous components of fiscal behavior to explain private
investment and (ii) the ability of the instruments to explain private
investment through the proposed fiscal policy channels. Whereas the
first concern is addressed by the significances and signs of estimated
coefficients, the second issue is tackled with the Sargan OIR test. The
null hypothesis of this test is the stance that the aid instruments
explain private investment only through the fiscal policy channels.
Therefore, a rejection of the null hypothesis is a rejection of the
perspective that the foreign aid instruments do not explain private
investment beyond the proposed mechanisms. We also employ a Hausman test
to account for endogeneity and justify the choice of the 2SLS-IV
estimation strategy. The null hypothesis of this test is the position
that estimated coefficients by OLS are consistent and efficient. Thus,
failure to reject this null hypothesis does not justify the choice of
the estimation strategy since it undermines the concern of endogeneity.
In light of the problem statement and theoretical background, the
Hausman test is a necessary but not a sufficient condition for the
employment of the 2SLS-IV strategy. Therefore, even in the absence of
endogeneity (failure to reject the null of the Hausman test), we still
employ the IV procedure.
In Table 3, we report a summary of findings from Tables 4-5. While
Table 4 is the baseline assessment with private investment, Table 5 is a
robustness check with fixed capital formation. Modeling is restricted
(Panel A) and unrestricted (Panel B) in both tables. While Tables 4-5
examine both the first and second concerns highlighted above, Table 3 is
based on only the second concern. Accordingly, given the problem
statement, the second issue is more relevant than the first because it
is premised on evidence from the first concern. In other words, while
addressing the first issue does not guarantee the second can be tackled,
examining the second is feasible when the first has been confirmed.
Therefore, the summary in Table 3 is based on the following information
criteria, the : (i) estimated coefficient should be significant; (ii)
adjusted coefficient of determination ([R.sup.2]) should not be
negative; (iii) Fisher statistics should be significant; (iv) null
hypothesis of the Sargan OIR test for the validity of the foreign aid
instruments should not be rejected and (v) Hausman test has an
informational role and is not indispensible for the validity of the
2SLS-IV model specification.
From Table 3, the following broad conclusions could be established.
(1) Foreign aid overwhelmingly increases private investment and gross
capital formation through tax effort, which is consistent with
theoretical underpinnings of and propositions in the study. (2) While
the effect of foreign aid on the dependent variables through government
expenditure is a bit mixed, the weight of available evidence on the
second issue broadly supports its positive impacts on private investment
and gross fixed capital formation. (3) It could be further inferred that
while the 'tax effort effect' is consistent across fundamental
characteristics of investment, the 'government spending
impact' may change as one move from one fundamental characteristic
to another. Hence, whereas the homogeneity on the tax effort mechanism
strongly confirms our theoretical hypothesis, the heterogeneity of the
government spending channel indicates that generalization of the
findings with respect of the government expenditure mechanism should be
treated with caution. (4) Our findings are more relevant for restricted
than for unrestricted modeling. This is an indication that autonomous
investment is not a very valid channel through which foreign aid is
instrumental in private investment. (5) Given the overwhelming presence
of 'not applicable' (na) (10) and degree ([degrees]) (11)
signs, it is difficult to establish significant asymmetries in various
dimensions of common fundamental characteristics. Therefore, evidence of
wealth-effect, legal-origin-effect ... landlocked-effect cannot be
feasibly drawn. (6) But for a thin exception (conflict-affected
countries), most of the significant control variables have the expected
signs: 'voice and accountability' and corruption-control are
logical incentives for private investors because they improve the
climate of doing business.
Discussion of results, policy implications and caveats
Discussion of results
From the weight of available empirical evidence (summarized in
Table 3), we have found an overwhelming homogenous effect of tax effort
on investment. Since the results are consistent with the proposed
theory; the explanation for the positive nexus conditional on foreign
aid has already been substantially covered in the 'Foreign aid and
development' section. Hence, the instrumentality or relevance of
foreign aid in the positive nexus could be explained by the fact that
development assistance reduces the tax effort of the government, which
provides additional incentives for private investment (either in terms
of reinvested profits or improvements in the investment climate). The
explanation extends to the formation of fixed capital (Table 5). Another
explanation to the positive relationship is that Western donor agencies
could require tax institutions to be: (i) more accountable and (ii) not
corrupt. Hence, the previously siphoned funds by corrupt officials are
transferred to the private sector. A third elucidation to the positive
nexus could be traceable to a lower composition of loans in the
development assistance portfolio. This is especially the case with
countries under the Highly Indebted Poor Countries initiative.
We have also found that the findings for the government expenditure
channel are heterogeneous or not consistently positive across
fundamental characteristics of private investment. The key idea to
understanding this heterogeneity is that the degree by which corrupt
officials chose to spend money on goods whose true value is hard to
identify, may differ across fundamental characteristics. Hence, the
negative nexus could be traceable to funds that are used for those
expenditures that provide more lucrative opportunities for bribery
(Shleifer and Vishny, 1993). Accordingly, expenditure on military and
high technology goods are some examples by which corrupt officials are
provided with lucrative mismanagement opportunities. Corruption and
military spending have been found to be closely linked, especially in
military aircraft (Hines, 1995). (12) On the other hand, the positive
nexus could be attributed to expenditures that do not seem to provide
any opportunities at all for corrupt officials and ultimately create
favorable conditions for private investments. Expenditure in education
is a case in point. For example, it may be difficult for a government
official to collect bribes for the appointment of unqualified persons to
teaching positions. This explanation could be extended to health,
although it is also disputable that sophisticated hospital equipment
could give rise to opportunities of corruption. (13)
This explanation confirms findings that corruption is linked to low
spending on education and health in developing countries (Mauro, 1998;
De la Croix and Delavallade, 2007).
Since, the negative nexus of government expenditure is contrary to
the proposed theoretical background, it is relevant to devote space to
explaining the discussion in the preceding paragraph to elaborate
detail, with hard stylized facts. It is worthwhile noting that the
'project approach' to foreign aid has underestimated the
incentive problems with aid delivery. Hence, education and health
ministries in recipient countries must be motivated to get school inputs
and medicines, respectively, to citizens. Moreover, donor bureaucracies
themselves must have the incentives to make sophisticated
infrastructural projects successful.
First, with respect to education, whereas enrollments have expanded
rapidly, the quality of education has been hampered by missing inputs
like textbooks and other school materials, corruption in 'education
bureaucracies' and weak incentives for teachers (Filmer and
Pritchett, 1997).
Second, from a health standpoint, some of the initial progress in
Africa has slowed possibly due to the siphoning of funds (Easterly,
2005a, p. 8). Studies in Cameroon, Guinea, Tanzania and Uganda estimate
that 30%-70% of government drugs disappear before they get to patients
and complicated health issues cannot be solved in the absence of routine
methods (Filmer et al., 2000; Prichett and Woolcock, 2004).
Third, with regard to the bureaucracy of sophisticated projects,
there have been some alarming dysfunctional signs. For example, donors
have spent over $2 billion over the past 20 years on roads in Tanzania,
but the roads have not improved. The principal output has been aid
bureaucracy because about 2,400 reports have been provided by 1,000
donor missions and government experts each year (Asongu and Jellal,
2014). The situation in Tanzania should not be generalized because
'aid conditionality' is also a relevant issue. Accordingly,
aid institutions could request complex road-building specifications
based on models in developed countries. Hence, the underlying issue is
also traceable to standards and procedural requirements.
The three points above could be summarized with another example
from Swaziland. It is a good aid candidate that substantially relies on
foreign aid, allocates about 55% of its public spending to the wage
bill, loses nearly double the annual social service budget to
corruption, sells food aid and deposits the money in foreign bank
accounts ... etc. The above points have one common denominator: foreign
aid channeled through dubious government expenditure mechanisms (that
serve only the interests of corrupt officials) may not provide the right
incentives for the growth of private investment and fixed capital
formation needed for economic prosperity in recipient countries.
Stimulating private investment with foreign aid through constraints
on fiscal behavior
The main policy implication arising from this study is that donor
agencies can condition aid to improve the fiscal system and management
of aid-related government expenditure in order to facilitate the inflow
of private investment and accrual of fixed capital needed for economic
growth (Asongu and Jellal, 2014). Hence, we shall briefly discuss
'revenue side' and 'expenditure side' constraints on
which development assistance can be conditioned in order to improve the
fiscal behavior of recipient countries.
On the revenue dimension of fiscal management, the following
constraints are worth noting. First, a tax administration reform should
embody the implementation of important anti-corruption measures within
the tax administrations, which include: (i) updating and modernizing tax
agency procedures; (ii) restructuring of internal organization based on
function (identification, assessment, billing ... etc) instead of by
'type of tax'; (iii) reducing the number of clearances that
are needed from taxpayers to complete compliance processes (ie, the
number of certifications, signatures, forms ... etc); (iv) limiting the
discretionary power of tax officials; (v) tax liability self-assessment
and (vi) exploring the use of electronic filling. Second, semiautonomous
revenue authorities are also vital. In essence, when properly
implemented, this enclave dimension to tax administration reform will
augment the possibility of de-politicizing tax officials, increase wage
levels for tax officials and strengthen internal monitoring mechanisms.
Consistent with the literature (Taliercio, 2003; Bird, 2004;
Martinez-Vazquez et al, 2006), these semi-autonomous authorities have
already been introduced in countries as diverse as Bolivia, Malaysia,
New Zealand, Singapore, Guatemala, Ghana, Guyana, Kenya, Malawi, Mexico,
Peru, Rwanda, South Africa, Tanzania, Uganda, Venezuela and Zambia.
Third, reforms of the tax system can reduce lucrative opportunities for
tax officials. Simplification of the tax system by reducing the number
of discretionary tax incentives, deductions and exemptions, is also
worthwhile.
From the supply perspective of fiscal management, the following
constraints are advisable. First, a modern treasury system should be
installed in a bid to augment transparency in cash management and
disbursement of resources for items authorized in the budget, needed for
consistency between formulation and execution. It is also relevant for
the treasury to operate separately from spending agencies and the
discretionary power of treasury officials can be reduced by separating
departments responsible for each budget execution stage. Second,
financial management reforms should be requested by aid agencies in
order to solidify basic procedures on budget accounting, auditing and
reporting. In essence, the public expenditure management should make use
of the integrated financial management systems and information
technologies. Third, a procurement system reform should be required to
facilitate the establishment of standardized procurement processes,
ensure maximum exposure and competition of foreign and national bidders
as well as satisfy international procurement standards. On account of
the fact that procurement systems can be particularly useful if combined
with the necessary administrative capacity, independent audition of the
procurement procedures should be conducted regularly and reviewed by
parliament. Fourth, a public expenditure tracking system should be
developed to identify leaks in the budget implementation stage. Fifth,
civil service reform should be oriented towards key measures that
mitigate the probabilities of patronage and corruption such as:
reduction of turnover rates, merit-based recruitment,
professionalization and de-politicization of public servants. Sixth, a
comprehensive coverage of the budget should minimize extra-budgetary and
off-budget accounts in order to maximize transparency in the use of
public resources. Seventh, strategies that emphasize political
accountability and political representation are necessary since broad
political contestability decreases the opportunities of state capture.
It is also worthwhile for ordinary citizens to have access to relevant
information concerning public spending, including parliamentary debates
on the budget formulation.
In addition to imposing constraints to improve the fiscal behavior
of aid-recipient countries, from the revenue and expenditure sides,
donors should also require an intergovernmental fiscal structure that
favors the decentralization of spending responsibilities and revenue
sources. This will provide increased accountability to citizens and
provide local governments with greater autonomy, which can be
instrumental in mitigating corruption in aid-funded projects.
It is interesting to note that for the most part, tax reforms have
been weak and belated in most African countries [in spite of aid
conditionalities), essentially because citizens are less willing to
comply with their tax obligations in the absence of political
accountability. This is the case with countries on the continent because
the Somaliland hypothesis provided by Eubank (2012) has been empirically
verified by Asongu (2015b) in 53 African countries with data of the same
periodicity as in the present line of inquiry. It follows that in the
absence of foreign aid; the dependence of recipients' governments
on local tax income provides the leverage for enhanced political
governance. The implication for our results is that foreign aid
decreases tax burden and the forgone tax income can be reinvested into
the economy by the private sector. This is essentially because in the
absence of foreign aid, governments are more willing to improve
political accountability in exchange for more tax income, since citizens
are more willing to pay taxes only in exchange for greater political
accountability.
In light of the above, expenditure reforms in Africa may have been
working exclusively at headquarters, but not downwards in the value
chain because of, inter alia: corruption in the allocation of projects
and mismanagement in the implementation of corresponding projects. This
has led to suggestions for more fiscal decentralization in policy
circles (eg the cases of Ethiopia, South Sudan and Sudan).
Unfortunately, foreign aid conditional on fiscal decentralization is
expectedly not a 'welcomed policy' by the political elites
that are benefiting from corrupt and mismanagement practices linked to
government centralization, in spite of the documented benefits of such
policy reforms. For instance, Teko and Nkote (2014) have recently shown
with the Ugandan experience that with effective fiscal decentralization,
aid flows are better managed because capacity building is enhanced.
It is relevant to briefly engage Rwanda as an example of good
recipient countries where foreign aid has been spent productively. This
country has particularly done well because of its specific development
model that is based on substantial decentralization. Accordingly, Rwanda
is widely recognized as a success story in aid effectiveness and
economic development partly because of its good leadership and division
of labor in the implementation of aid programs. These are consistent
with: (i) decentralization and enhanced harmonization and (ii) alignment
of national priorities with donor conditionalities. As documented by
Abbott and Rwirahira (2012), the country's development strategy has
resulted in: (i) enhanced transformation and economic growth, (ii)
reduced aid dependency and (iii) boosted pro-poor growth.
Before we conclude, it is important to emphasize that the findings
are particularly relevant to African countries in the post-2015
development agenda because the April 2015 World Bank report on MDGs
extreme poverty targets has revealed that poverty has been decreasing in
all regions of the world with the exception of SSA, where 45 % of
countries in the sub-region are substantially off-track from the target
(World Bank, 2015). In essence private investment is a good source of
employment and growth for poverty mitigation.
CONCLUSION
The paper has provided theoretical and empirical justifications for
the instrumentality of foreign aid in stimulating private investment and
fixed capital formation through fiscal policy mechanisms. We have
proposed an endogenous growth theory based on an extension of Barro
(1990) by postulating that the positive effect of aid mitigates the
burden of the taxation system on the private sector of recipient
countries. The empirical validity is based on data from 53 African
countries for the period 1996-2010. While the findings on the tax effort
channel are overwhelmingly consistent with theory across specifications
and fundamental characteristics, those of the government expenditure
channel are a little heterogeneous but broadly in line with the
theoretical postulations. Justifications for the slight heterogeneity
and policy implications have been discussed.
We devote some space to caveats and future research directions. In
light of theoretical underpinnings of the paper, the study has not taken
two major elements into account. First, it would be interesting to
decompose government expenditure into its constituent elements in order
to understand which components favor private investment activities more.
This is essentially because corrupt officials would always device
mechanisms by which to channel 'aid funds' to those
expenditures that provide more lucrative opportunities for bribery and
mismanagement. Second, the distinction between concessional loans and
grants in the measurement of development assistance will enable a better
understanding of the instrumentality of foreign aid in the
investment-'fiscal policy' nexuses. For instance, the type of
foreign aid that augments/reduces the tax effort related to private
investments. Hence, interesting future research directions could include
the incorporation of above caveats in order to provide policy makers
with more specific findings. Moreover, future inquiries devoted to
assessing investment volumes and policies within the frameworks of
subsidies and tax privileges, would also enrich the extant literature.
Moreover, in cases of bad recipient countries, it would be
interesting to document what donors have done or are doing in terms of
'aid conditionality'. Elucidating the political economy of
these countries may be an important direction towards understanding how:
(i) to deal with kleptokracies in Africa and (ii) such bad cases fit
into the 'aid conditionality approach' based on fiscal
behavior.
Acknowledgements
The authors are highly indebted to Paul Wachtel and reviewers for
constructive comments.
REFERENCES
Abbott, P and Rwirahira, J. 2012: 'Aid Effectiveness in
Rwanda: Who Benefits?', Institute of Policy Analysis and
Research--Rwanda, http://reliefweb.int/sites/reliefweb.int/files/resources/Aid % 20effectiness%20in%20Rwanda%202012.pdf, accessed 26 October
2015.
Addison, T, Mavrotas, G and McGillivray, M. 2005: Development
assistance and development finance: Evidence and global policy agendas.
Journal of International Development 17(6): 819-836.
Ahmed, I, Marcoux, C, Russell, B and Tierney, MJ. 2011: Foreign aid
in hard times: The political economy of aid effort. Institute of Theory
and International Relations: Williamsburg
Alam, S. 2004: Colonialism and industrialization: A critique of
Lewis. Journal of Radical Political Economics 36(2): 217-240.
Alexiou, C. 2009: Government spending and economic growth: Economic
evidence from South Eastern Europe (SSE). Journal of Economic and Social
Research 11(1): 1-16.
Asiedu, E. 2014: Does foreign aid in education promote economic
growth? Evidence from Sub-Saharan Africa. Journal of African Development
16(1): 37-59.
Asiedu, E and Nandwa, B. 2007: On the impact of foreign aid in
education on growth: How relevant is heterogeneity of aid flows and
heterogeneity of aid recipients? Review of World Economics 143(4):
631-649.
Asongu, SA. 2012a: On the effect of foreign aid on corruption.
Economics Bulletin 32(3): 2174-2180.
Asongu, SA. 2012b: Government quality determinants of stock market
performance in African countries. Journal of African Business 13(2):
183-199.
Asongu, SA. 2013a: On the effectiveness of foreign aid in
institutional quality. European Economics Letters 2(1): 12-19.
Asongu, SA. 2013b: Fighting corruption when existing corruption
control levels count: What do wealth-effects tell us in Africa?
Institutions and Economies 5(3): 53-74.
Asongu, SA. 2014a: The questionable economics of development
assistance in Africa: Hot-fresh evidence, 1996-2010. The Review of Black
Political Economy 41 (4): 455-480.
Asongu, SA. 2014b: Fighting African capital flight: Empirics on
benchmarking policy harmonization. European Journal of Comparative
Economics 11(1): 93-122.
Asongu, SA. 2015a: 'Knowledge economy gaps, policy syndromes
and catch-up strategies: Fresh South Korean lessons to Africa',
Journal of the Knowledge Economy: http://link.springer.com/
article/10.1007/sl 3132-015-0321-0.
Asongu, SA. 2015b: On taxation, political accountability and
foreign aid: Empirics to a celebrated literature. South African Journal
of Economics 83(2): 180-198.
Asongu, SA. 2015c: Fighting software piracy in Africa: How do legal
origins and IPRs protection channels matter? Journal of the Knowledge
Economy 6(4): 682-703.
Asongu, SA. 2015d: Law, finance, economic growth and welfare: Why
does legal origin matter? Institutions and Economies 7(2): 30-55.
Asongu, SA and Jellal, M. 2014: International aid, corruption and
fiscal policy behavior. African Governance and Development Institute
Working paper no. 14/007, Yaounde.
Asongu, SA and Nwachukwu, JC. 2016: Foreign aid and governance in
Africa. International Review of Applied Economics 30(1): 69-88.
Asongu, SA and Ssozi, J. 2016: Sino-African relations: Some
solutions and strategies to the policy syndromes. Journal of African
Business 17(1): 35-51.
Asongu, SA and Tchamyou, SV. 2016: The impact of entrepreneurship
on knowledge economy in Africa. Journal of Entrepreneurship in Emerging
Economies 8(1): 101-131.
Arvis, J-F, Marteau, J-F and Raballand, G. 2007: The cost of being
landlocked: Logistics costs and supply chain reliability. Word Bank
Working paper series no. 4258.
Barro, RJ. 1990: Government spending in a simple model of
endogenous growth. Journal of Political Economy 98(S5): 103-125.
Bartels, FL, Alladina, SN and Lederer, S. 2009: Foreign direct
investment in Sub-Saharan Africa: Motivating factors and policy issues.
Journal of African Business 10(2): 141-162.
Bird, RM. 2004: Administrative dimensions of tax reform.
Asia-Pacific Tax Bulletin 10(3): 3-21.
Boone, P. 1996: Politics and effectiveness of foreign aid. European
Economic Review 40(2): 289-329.
Brautigam, DA and Knack, S. 2004: Foreign aid, institutions, and
governance in Sub-Saharan Africa. Economic Development and Cultural
Change 52(2): 255-285.
Burnside, C and Dollar, D. 2000: Aid, policies and growth. American
Economic Review 90(4): 847-868.
CIA. 2011: Central Intelligence Agency, The World Factbook 2011.
Chenery, HB and Strout, AM. 1966: Foreign assistance and economic
development. American Economic Review 56(4): 679-733.
Clemens, MA, Radelet, S and Bhavnani, R. 2004: Counting chickens
when they hatch: The short-term effect of aid on growth. Center for
Global Development Working paper no. 44, Washington DC.
Collier, P and Dehn, J. 2001: Aid, shocks and growth. World Bank
Policy Research Working paper no. 2001/11/06.
Collier, P and Dollar, D. 2001: Can the world cut poverty in half?
How policy reform and effective aid can meet international development
goals. World Development 29(111): 1787-1802.
Crotty, J and Lee, K. 2002: Is financial liberalization good for
developing countries? The case of South Korea in the 1990s. Review of
Radical Political Economics 34(3): 327-334.
Crotty, J and Lee, K. 2006: The effects of neoliberal reforms on
the postcrisis Korean economy. Review of Radical Political Economics
38(3): 381-387.
Crotty, J and Lee, K. 2009: Was IMF-imposed economic regime change
in Korea justified? The political economy of IMF intervention. Review of
Radical Political Economics 41(2): 2149-2169.
Degnbol-Martinussen, J and Engberg-Pedersen, P. 2003: Aid:
Understanding international development cooperation. Zed Books: London.
De la Croix, D and Delavallade, C. 2007: Corruption et allocation
optimale de l'investissement public. Revue Economique 58(3):
637-647.
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. 2003: Can foreign aid buy growth? Journal of Economic
Perspectives 17(3): 23-46.
Easterly, W. 2005a: Can foreign aid save Africa. Saint John's
University: New York.
Easterly, W, Levine, R and Roodman, D. 2003: New data, new doubts:
a comment on burnside and dollar's 'aid, policies and growth
2000'. Working Paper No. 9846, American Economic Review, NBER,
Cambridge, MA.
Eubank, N. 2012: Taxation, political accountability and foreign
aid: Lessons from Somaliland. Journal of Development Studies 48(4):
465-480.
Ghura, D, Hadjimichael, MT, Mahleisen, M, Nord, R and Ucer, EM.
1995: Sub-Saharan Africa: Growth, savings and investment, 1986-93. IMF
Occasional Paper No. 118, Washington DC.
Gyimah-Brempong, K and Racine, JS. 2014: Aid and economic growth: A
robust approach. Journal of African Development 16(1): 465-480.
Hansen, H and Tarp, F. 2001: Aid and growth regressions. Journal of
Development Economics 64(2): 547-570.
Feeny, S. 2003: The impact of foreign aid on poverty and human
well-being in Papua New Guinea. Asia-Pacific Development Journal 10(2):
73-93.
Feredeaned, E and Dahlby, B. 2012: The impact of tax cuts on
economic growth: Evidence from the Canadian provinces. National Tax
Journal 65(3): 563-594.
Fielding, D, McGillivray, M and Torres, S. 2006: A wider approach
to aid effectiveness: Correlated impacts on health, wealth, fertility
and education. The World Institute for Development Economics Research
(UNU-WIDER): Helsinki.
Filmer, D, Hammer, J and Pritchett, L. 2000: Weak links in the
chain: A diagnosis of health policy in poor countries. The World Bank
Research Observer 15(2): 199-224.
Filmer, D and Pritchett, L. 1997: What educational production
functions really show: A positive theory of education spending. World
Bank Policy Research Paper No. 1795.
Fosu, AK. 2015a: Growth and institutions in African development.
First edited by Augustin KFRoutledge Studies in Development Economics:
New York.
Fosu, AK. 2015b: Growth and institutions in African development.
First edited by Augustin KF. Growth and Institutions in African
Development. 2015, Chapter 1, Routledge Studies in Development
Economics: New York, pp. 1-17.
Gomane, K, Morrissey, 0, Mosley, P and Verschoor, A. 2003: Aid,
pro-poor government spending and welfare. CREDIT Working paper 03/01.
Guillaumont, P and Chauvet, L. 2001: Aid and performance: A
reassessment. Journal of Development Studies 37(6): 66-92.
Herrick, C. 2008: The Southern African famine and genetically
modified food aid: The ramifications for the United States and European
Union's trade war. Review of Radical Political Economics 40(1):
50-66.
Hines, J. 1995: Forbidden payments: Foreign bribery and American
business. NBER Working paper 5266.
Ishfaq, M. 2004: Aid effectiveness, debt capacity and debt
management in the economy of Pakistan. A PhD dissertation at Quad-e-Azam
University, Islamabad.
Kargbo, PM and Sen, K. 2014: Aid categories that foster pro-poor
growth: The case of Sierra Leone. African Development Review 26(2):
416-429.
Koechlin, T. 2007: Fighting global poverty, three ways. Review of
Radical Political Economics 39(3): 377-384.
Kuada, J. 2015: Private enterprise-led economic development in
Sub-Saharan Africa the human side of growth, 1st Edition by Kuada, J.
Palgrave Macmillan: New York.
La Porta, R, Lopez-de-Silanes, F, Shleifer, A and Vishny, RW. 1998:
Law and finance. Journal of Political Economy 106(6): 1113-1155.
La Porta, R, Lopez-de-Silanes, F, Shleifer, A and Vishny, RW. 1999:
The quality of government. Journal of Law, Economics and Organization
15(1): 222-279.
La Porta, R, Lopez-de-Silanes, F and Shleifer, A. 2008: The
economic consequences of legal origin. Journal of Economic Literature
46(2): 285-332.
Levy, D and Dumenil, G. 2006: Imperialism in the neoliberal era:
Argentina's reprieve and crisis. Review of Radical Political
Economics 38(3): 388-396.
Manly, TS, Thomas, DW and Schulman, CT. 2006: Tax incentives for
economic growth: Capital investment or research. In: Luttman, S. (ed).
Advances in Taxation, vol. 17. Emerald Group Publishing Limited:
Bingley, pp. 95-120.
Martinez-Vazquez, J, Arze, FJ and Boex, J. 2006: Corruption, fiscal
policy and fiscal management: Fiscal reform in support of trade
liberalization? United States Agency for International Development:
Washington DC.
Masud, N and Yontcheva, B. 2005: Does foreign aid reduce poverty?
Empirical evidence from nongovernmental and bilateral aid. IMF Working
paper, 05/100, Washington DC.
Mauro, P. 1998: Corruption and the composition of government
expenditure. Journal of Public Economics 69(2): 263-279.
Minou, C and Reddy, SG. 2010: Development aid and economic growth:
A positive long-run relation. The Quarterly Review of Economics and
Finance 50(1): 27-39.
Morrissey, O. 2012: Aid and government fiscal behavior. What does
the evidence say? UNU-WIDER Working paper no. 2012/01, Helsinki.
Mosley, P, Hudson, J and Horrell, S. 1992: Aid, the public sector
and the market in less developed countries: A return to the scene of
crime. Journal of International Development 4(2): 139-150.
Mosley, P, Hudson, J and Verschoor, A. 2004: Aid, poverty reduction
and the 'new' conditionality. Economic Journal 114(496):
F217-F243.
Okada, K and Samreth, S. 2012: The effect of foreign aid on
corruption: A quantile regression approach. Economic Letters 11(2):
240-243.
Oya, C. 2006: The political economy of development aid as main
source of foreign finance for poor African countries: Loss of policy
space and possible alternatives from East Asia. School of Oriental and
African Studies, University of London: London.
Payne, JE and Kumazawa, R. 2005: Capital mobility, foreign aid and
openness: Further panel data evidence from Sub-Saharan Africa. Journal
of Economics and Finance 29(1): 122-126.
Pedersen, KP. 1996: Aid, investment and incentives. Scandinavian
Journal of Economics 98(3): 423-438.
Pritchett, L and Woolcock, M. 2004: Solutions when the solution is
the problem: Arraying the disarray in development. World Development
32(2): 191-212.
Reichel, R. 1995: Development, aid, savings and growth in the
1980s: A cross-sectional analysis. Savings and Development 19(3):
279-296.
Remmer, KL. 2004: Does foreign aid promote the expansion of
government. American Journal of Political Science 48(1): 77-92.
Resnick, D. 2012: Foreign aid in Africa, tracing channels of
influence on democratic transitions and consolidation. UNU-WIDER Working
paper no. 2012/15, Helsinki.
Rostow, WW. 1960: The stages of economic growth: A non-communist
manifesto. Cambridge University Press: Cambridge.
Shleifer, A and Vishny, R. 1993: Corruption. Quarterly Journal of
Economics 108(3): 599-617.
Snyder, DW. 1996: Foreign aid and private investment in developing
countries. Journal of International Development 8(6): 735-745.
Taliercio, R. 2003: Designing performance: The semi-autonomous
revenue authority model in Africa and Latin America. World Bank Policy
Research Working paper no. 3423, Washington DC.
Tchamyou, VS. 2015: The role of knowledge economy in African
business. Journal of the Knowledge Economy. Revised and Resubmitted.
Teko, J and Nkote, J. 2014: Fiscal decentralization, donor aid and
capacity building: Empirical experiences from Uganda. Journal of
Research in International Business and Management 4(3): 64-73.
Weeks, J. 2012: Macroeconomic impact of capital flows in
Sub-Saharan African countries, 1980-2008. Association of Concerned
Africa Scholar Bulletin (87): 1-7.
World Bank. 2015: 'World Development Indicators', World
Bank Publications http://www.gopa.de/
fr/news/world-bank-release-world-development-indicators-2015, accessed
25 April 2015.
APPENDIX
Table Al: Categorization of countries
Category Panels Countries Num
Income Upper Middle Algeria, Botswana, Equatorial 10
levels Income Guinea, Gabon, Libya, Mauritius,
Namibia, Sao Tome & Principe,
Seychelles, South Africa.
Lower Middle Angola, Cameroon, Cape Verde, Cote 12
Income d'Ivoire, Egypt, Lesotho, Morocco,
Nigeria, Senegal, Sudan, Swaziland,
Tunisia.
Middle Income Algeria, Angola, Botswana, Cameroon, 22
Cape Verde, Cote d'Ivoire, Egypt,
Equatorial Guinea, Gabon, Lesotho,
Libya, Mauritius, Morocco, Namibia,
Nigeria, Sao Tome & Principe,
Senegal, Seychelles, South Africa,
Sudan, Swaziland, Tunisia.
Low Income Benin, Burkina Faso, Burundi, 31
Central African Republic, Chad,
Comoros, Congo Democratic Republic,
Congo Republic, Djibouti, Eritrea,
Ethiopia, The Gambia, Ghana, Guinea,
Guinea-Bissau, Kenya, Liberia,
Madagascar, Malawi, Mali,
Mauritania, Mozambique, Niger,
Rwanda, Sierra Leone, Somalia,
Tanzania, Togo, Uganda, Zambia,
Zimbabwe.
Legal English Botswana, The Gambia, Ghana, Kenya, 20
oriqins Common-law Lesotho, Liberia, Malawi, Mauritius,
Namibia, Nigeria, Seychelles, Sierra
Leone, Somalia, South Africa, Sudan,
Swaziland, Tanzania, Uganda, Zambia,
Zimbabwe.
French Algeria, Angola, Benin, Burkina 33
Civil-law Faso, Burundi, Cameroon, Cape Verde,
Central African Republic, Chad,
Comoros, Congo Democratic Republic,
Congo Republic, Cote d'Ivoire,
Djibouti, Egypt, Equatorial Guinea,
Eritrea, Ethiopia, Gabon, Guinea,
Guinea-Bissau, Libya, Madagascar,
Mali, Mauritania, Morocco,
Mozambique, Niger, Rwanda, Sao Tome
& Principe, Senegal, Togo, Tunisia.
Religious Christianity Angola, Benin, Botswana, Burundi, 33
domination Cameroon, Cape Verde, Central
African Republic, Congo Democratic
Republic, Congo Republic, Cote
d'Ivoire, Equatorial Guinea,
Eritrea, Ethiopia, Gabon, Ghana,
Kenya, Lesotho, Liberia, Madagascar,
Malawi, Mauritius, Mozambique,
Namibia, Rwanda, Sao Tome &
Principe, Seychelles, South Africa,
Swaziland, Tanzania, Togo, Uganda,
Zambia, Zimbabwe.
Islam Algeria, Burkina Faso, Chad, 20
Comoros, Djibouti, Egypt, The
Gambia, Guinea, Guinea-Bissau,
Libya, Mali, Mauritania, Morocco,
Niger, Nigeria, Senegal, Sierra
Leone, Somalia, Sudan, Tunisia.
Regions Sub-Saharan Angola, Benin, Botswana, Burkina 47
Africa Faso, Burundi, Cameroon, Cape Verde,
Chad, Central African Republic,
Comoros, Congo Democratic Republic,
Congo Republic, Cote d'Ivoire,
Djibouti, Equatorial Guinea,
Eritrea, Ethiopia, Gabon, The
Gambia, Ghana, Guinea, Guinea-
Bissau, Kenya, Lesotho, Liberia,
Madagascar, Malawi, Mali, Mauritius,
Mozambique, Namibia, Niger, Nigeria,
Senegal, Sierra Leone, Somalia,
Sudan, Rwanda, Sao Tome & Principe,
Seychelles, South Africa, Swaziland,
Tanzania, Togo, Uganda, Zambia,
Zimbabwe.
North Africa Algeria, Egypt, Libya, Mauritania, 6
Morocco, Tunisia.
Resources Petroleum Algeria, Angola, Cameroon, Chad, 10
Exporting Congo Republic, Equatorial Guinea,
Gabon, Libya, Nigeria, Sudan.
Non-Petroleum Benin, Botswana, Burkina Faso, 43
Exporting Burundi, Cape Verde, Central African
Republic, Comoros, Congo Democratic
Republic, Cote d'Ivoire, Djibouti,
Eritrea, Ethiopia, Egypt, The
Gambia, Ghana, Guinea, Guinea-
Bissau, Kenya, Lesotho, Liberia,
Madagascar, Malawi, Mali,
Mauritania, Mauritius, Morocco,
Mozambique, Namibia, Niger, Senegal,
Sierra Leone, Somalia, Rwanda, Sao
Tome & Principe, Seychelles, South
Africa, Swaziland, Tanzania, Togo,
Tunisia, Uganda, Zambia, Zimbabwe.
Stability Conflict Angola, Burundi, Chad, Central 12
African Republic, Congo Democratic
Republic, Cote d'Ivoire, Liberia,
Nigeria, Sierra Leone, Somalia,
Sudan, Zimbabwe.
Non-Conflict Algeria, Benin, Botswana, Burkina 41
Faso, Cameroon, Cape Verde, Comoros,
Congo Republic, Djibouti, Egypt,
Equatorial Guinea, Eritrea,
Ethiopia, Gabon, The Gambia, Ghana,
Guinea, Guinea-Bissau, Kenya,
Lesotho, Libya, Madagascar, Malawi,
Mali, Mauritania, Mauritius,
Morocco, Mozambique, Namibia, Niger,
Senegal, Rwanda, Sao Tome &
Principe, Seychelles, South Africa,
Swaziland, Tanzania, Togo, Tunisia,
Uganda, Zambia.
Openness Landlocked Botswana, Burkina Faso, Burundi, 15
to sea Chad, Central African Republic,
Ethiopia, Lesotho, Malawi, Mali,
Niger, Rwanda, Swaziland, Uganda,
Zambia, Zimbabwe
Not Algeria, Angola, Benin, Cameroon, 38
landlocked Cape Verde, Comoros, Congo
Democratic Republic, Congo Republic,
Cote d'Ivoire, Djibouti, Egypt,
Equatorial Guinea, Eritrea, Gabon,
The Gambia, Ghana, Guinea, Guinea-
Bissau, Kenya, Liberia, Libya,
Madagascar, Mauritania, Mauritius,
Morocco, Mozambique, Namibia,
Nigeria, Senegal, Sierra Leone,
Somalia, Sudan, Sao Tome & Principe,
Seychelles, South Africa, Tanzania,
Togo, Tunisia.
Num: Number of cross sections (countries).
(1) For instance Snyder (1996) has concluded from a panel of 36
developing countries that nations that receive large aid allocations are
associated with lower levels of private investment.
(2) The debate has recently been reframed by Koechlin (2007) who
has examined three ambitious book (Sachs's The End of Poverty,
Bhagwati's In Defense of Globalization, and Easterly's The
Elusive Quest for Growth) and concluded that, the insights and drawbacks
of underlying books remind us that the 'status quo' is not
working. The author has concluded that a rich understanding of
globalization and development requires a serious reconsideration of
alternative visions of each phenomenon. For instance, new ways of
theorizing development in light of the globalized system of food
production has involved the European Union heavily criticizing the
USA-led 'genetically modified food aid' program to the
Southern African region (Herrick, 2008).
(3) As a case in point, China's policy in Africa of
non-interference in development assistance and foreign direct investment
is perceived as a better alternative (Asongu and Ssozi, 2016). Hence,
the results of this study could either confirm or reject the narrative.
(4) The conditionality-oriented debate has recently intensified
when some Western governments (British and US for instance) have
threatened to cut-off aid from some African countries because of the
prosecution of gays, lesbians and transsexuals by governments of
recipient countries. In response, activists, analysts and African
government officials have viewed the threat as an insult to African
values in particular and moral well-being in general.
(5) Accordingly, structural adjustment policies by the IMF have
also been criticized. There is a wealth of literature documenting that
the IMF's neoliberal policies have not been: (i) sound for South
Korean development after the 1997 crisis (Crotty and Lee, 2002, 2006,
2009); (ii) the principal cause of the Argentinean crisis in the late
1990s and early 2000s (Levy and Dumenil, 2006) and (iii) responsible for
the failed privatization projects across Africa (Bartels et al., 2009).
(6) Whereas Okada and Samreth (2012) have concluded that aid
mitigates corruption in developing countries, Asongu (2012a) in response
has established that the Okada and Samreth (2012) findings may not be
relevant for Africa because aid fuels (mitigates) corruption (the
control of corruption) on the continent. In response to some informal
discussions that the Okada and Samreth and Asongu (2012a) findings are
not directly comparable, Asongu (2013a) has maintained his position in
the African context without partially negating the empirical
underpinnings of Okada and Samreth on the one hand and extending the
horizon of inquiry from corruption to eight government quality
variables.
(7) Addison et al. (2005) have established that development
assistance encourages pro-poor public spending and has a positive effect
on economic prosperity (growth) since it broadly aligns with poverty
reduction. Their position that poverty would be higher in the absence of
aid had earlier been raised by Ishfaq (2004). Among proponents of a
positive aid-development nexus in the first strand of Table 1, Fielding
et al. (2006) have been the most optimistic in their conclusion that aid
has a straight forward positive effect on development objectives.
(8) The time span is consistent with those employed by Okada and
Samreth (2012), Asongu (2012a) and Asongu (2013a) in the highlighted
debate. The first authors have use data from 120 developing countries
for the period 1995-2009, the second has used data from 52 African
countries for the period 1996-2010 whereas the third has used data for
the period 1996-2010 from 53 African countries.
(9) Please see last paragraph of the 'Methodology'
section for further insights.
(10) Insignificant estimate or variable not included in model.
(11) [omicron] negative coefficient of determination, significant
Sargan OIR test (invalid instruments) or insignificant Fisher
statistics.
(12) It is therefore not surprising that the worst post-apartheid
corruption scandal that has embroiled the current president (Jacob Zuma)
has been linked to the purchase of military equipment. In the same line
of thinking (from a high technology standpoint), the
'Albatross' jet affair that has rocked the Cameroonian
institutional landscape has seen the arrest of many high profile
politicians over the spectacular disappearances of $ 25 million destined
for the purchase of a presidential plane.
(13) To further illustrate this point, a recent budget scandal in
South Africa has resulted from the government's spending of R4
billion on entertainments, travel allowance and catering in 2011 while
under-spending in health initiatives, which has left about 47% of
metropolitan South Africans dissatisfied.
SIMPLICE ASONGU (1,2) & MOHAMED JELLAL (3)
(1) African Governance and Development Institute, P.O. Box 8413,
Yaounde, Cameroon. E-mail: asongus@afridev.org
(2) Department of Economics, Business School, Oxford Brookes
University, Wheatley Campus, Wheatley, Oxford, 0X33 1HX, UK.
(3) Al Makrizi Institut d'economie, Rabat, Morocco.
Table 1: Summary of controversial views in the literature
Researchers Main findings
First-strand: Aid improves growth (development)
Ghura et al. (1995) Aid positively impacts savings for
good adjusters.
Burnside and Dollar (2000) Aid can be good when economic
management and policies are appealing.
Guillaumont and Chauvet (2001) Aid effectiveness is conditional on
environmental factors (hazards and
shocks).
Collier and Dehn (2001) Aid effectiveness is contingent on
negative supply shocks. Targeting aid
conditional on negative supply shocks
is better than a targeting based on
good policies.
Collier and Dollar (2001) The positive impact of aid on poverty
depends on its effect on per- capita
income growth and the effect of
per-capita income growth on poverty
mitigation.
Feeny(2003) The sectoral allocation of foreign aid
to Papua New Guinea has been broadly
in line with a strategy to effectively
mitigate poverty and increase human
well-being.
Gomanee etal. (2003) Aid has both a direct impact on
welfare and indirect effect via public
spending on social services.
Clement et al. (2004) Aid has a short-run appealing impact
on growth.
Ishfaq (2004) Though in a limited way, aid has
helped in reducing the extent of
poverty in Pakistan.
Mosley et al. (2004) Aid has an indirect impact on
well-being and poverty in recipient
countries.
Addison et al. (2005) Aid augments pro-poor public
expenditure and has a positive impact
on economic prosperity. Aid broadly
works to reduce poverty, and poverty
would be higher in the absence of aid.
Fielding et at. (2006) There is a straight forward positive
impact of aid on development
objectives.
Minou and Reddy (2010) Aid positively impacts economic
prosperity in the long-run.
Okada and Samreth (2012) Aid mitigates corruption.
Resnick (2012) Aid has promoted democratic
transitions in the 1990s in African
countries.
Second-strand: Aid does not lead to growth (development)
Mosley et al. (1992) Aid promotes unproductive public
consumption and fails to promote
growth.
Reichel (1995) Aid does not encourage savings because
of the substitution effect.
Ghura et al (1995) Aid has a negative incidence on
savings.
Boone (1996) Aid is insignificant in promoting
economic development on two main
counts: poverty is not the effect of
capital shortage and it is not optimal
for politicians to adjust
distortionary policies when they
receive aid flows.
Pedersen (1996) Aid distorts development and
eventually leads to aid dependency.
Asongu(2012a) Aid fuels (mitigates) corruption (the
control of corruption).
Asongu and Nwachukwu (2016) Aid has a negative nexus with
government quality dynamics.
Asongu(2013a) Aid is unappealing to institutional
quality irrespective of initial levels
of institutional development.
Asongu (2014a) Aid leads to less pro-poor
development.
Source: Authors
Table 2: Variable definitions and summary statistics
Variables Variable definitions
Corruption Control Index 'Control of corruption (estimate):
captures perceptions of the extent
to which public power is exercised
for private gain, including both
petty and grand forms of
corruption, as well as "capture" of
the state by elites and private
interests'.
Voice & Accountability 'Voice and accountability
(estimate): measures the extent
to which a country's citizens are
able to participate in selecting
their government and to enjoy
freedom of expression, freedom of
association and a free media'.
Government Expenditure Government Final Consumption
Expenditure (% of GDP)
Tax Revenue Tax Revenue (% of GDP)
Fixed Capital Formation Gross Fixed Capital Formation
(% of GDP)
Private Investment Gross Private Investment
(% of GDP)
Foreign Aid (1) Total Net Official Development
Assistance (% of GDP)
Foreign Aid (2) NODA from DAC Countries
(% of GDP)
Foreign Aid (3) NODA from Multilateral Donors
(% of GDP)
Grants Grants excluding technical
cooperation (% of GDP)
Variables Mean S.D Min. Max. Obs.
Corruption Control Index -0.607 0.623 -2.495 1.086 622
Voice & Accountability -0.674 0.734 -2.174 1.047 636
Government Expenditure 4.392 12.908 -57.815 90.544 468
Tax Revenue 17.693 10.096 0.116 61.583 262
Fixed Capital Formation 19.708 10.715 -23.76 113.58 706
Private Investment 12.979 9.400 -2.437 112.35 658
Foreign Aid (1) 10.811 12.774 -0.251 148.30 704
Foreign Aid (2) 6.244 8.072 -0.679 97.236 704
Foreign Aid (3) 6.244 8.072 -0.679 97.236 704
Grants 0.069 0.115 0.000 1.477 773
Source of variables: World Development Indicators. S.D: Standard
Deviation. Min: Minimum. Max: Maximum. Obs: Observations
Table 3: Summary of results
Income levels Legal origins
UMI LMI MI LI English French
Panel A: Specifications in Panel A of Table 3
(Restricted Private Investment Modeling)
Gov. Exp. - na na na + na
Tax Rev. + + (a) + + + + (a)
Panel B: Specifications in Panel B of Table 3
(Unrestricted Private Investment Modeling)
Gov. Exp. na na na na na na
Tax Rev. na na na na + na
Panel C: Specifications in Panel A of Table 4
(Restricted Fixed Capital Formation Modeling)
Gov. Exp. - na na na + - (a)
Tax Rev. + + (a) + + + + (a)
Panel D: Specifications in Panel B of Table 4
(Unrestricted Fixed Capital Formation Modeling)
Gov. Exp. na na na na na na
Tax Rev. na + na na + na
Religious dom. Regions Resources
Christ. Islam SSA NA Oil Non-oil
Panel A: Specifications in Panel A of Table 3
(Restricted Private Investment Modeling)
Gov. Exp. + na + - (a) + (a) +
Tax Rev. + + (a) + + (a) na +
Panel B: Specifications in Panel B of Table 3
(Unrestricted Private Investment Modeling)
Gov. Exp. na na na na + na
Tax Rev. na na na + + na
Panel C: Specifications in Panel A of Table 4
(Restricted Fixed Capital Formation Modeling)
Gov. Exp. na - + - (a) + +
Tax Rev. + + + + (a) + +
Panel D: Specifications in Panel B of Table 4
(Unrestricted Fixed Capital Formation Modeling)
Gov. Exp. na na na na + na
Tax Rev. na + na + + na
Stability Landlocked(LL) Africa
Conflict Non-co. LL Not LL
Panel A: Specifications in Panel A of Table 3
(Restricted Private Investment Modeling)
Gov. Exp. - (a) + + na na
Tax Rev. na + + + +
Panel B: Specifications in Panel B of Table 3
(Unrestricted Private Investment Modeling)
Gov. Exp. na na na na na
Tax Rev. + na + na na
Panel C: Specifications in Panel A of Table 4
(Restricted Fixed Capital Formation Modeling)
Gov. Exp. - + + na na
Tax Rev. na + + + (a) +
Panel D: Specifications in Panel B of Table 4
(Unrestricted Fixed Capital Formation Modeling)
Gov. Exp. na na na na na
Tax Rev. + na na na na
(a) Negative coefficient of determination, significant Sargan OIR
test (invalid instruments) or insignificant Fisher statistics.
+(-): positive (negative) effect. Gov. Exp: Government Expenditure.
Tax Rev: Tax Revenue. UMI: Upper Middle Income. LMI: Lower Middle
Income. MI: Middle Income. LI: Low Income. English: English
Common-law. French: French Civil-law. Christ.: Christianity
dominated countries. Islam: Islam dominated countries. SSA:
Sub-Saharan Africa. NA: North Africa. Oil: Petroleum exporting
countries. Non-oil: Countries with no significant exports in
petroleum. Conflict: Countries with significant political
instability. Non-co: Countries without significant political
instability. Dorn: Domination, na: insignificant estimate or
variable not included in model.
Table 4: Baseline assessment with private investment (HAC standard
errors)
Income levels
UMI LMI MI LI
Panel A: Restricted Modeling
Constant -- -- -- --
Gov. Exp. -0.60# ** 1.064 1.951 0.301
(0.016)# (0.325) (0.270) (0.446)
Tax Rev. 0.66# *** 0.568# ** 0.512# * 0.822# ***
(0.000)# (0.017)# (0.071)# (0.000)#
C. Control 4.383 6.888 -- --
(0.752) (0.492)
Voice & A. -- -- 17.69# ** --
(0.029)#
Hausman 67.6# *** 18.37# *** 51.87# *** 19.88# ***
(0.000)# (0.000)# (0.000)# (0.000)#
Sargan OIR 0.425# 14.29 *** 1.210# 1.538#
(0.808)# (0.000) (0.545)# (0.673)#
Adjusted 0.215 0.203 0.110 0.032
[R.sup.2]
[chi square] -- -- -- 147# ***
Fisher 152# *** 45.07# *** 23.73# *** --
Observations 34 51 87 77
Panel B: Unrestricted Modeling
Constant 63.22 21.83# *** 16.06# ** 12.925
(0.595) (0.000)# (0.039)# (0.207)
Gov. Exp. -0.108 0.135 0.039 -0.028
(0.917) (0.840) (0.975) (0.890)
Tax Rev. -1.614 0.056 0.219 0.092
(0.718) (0.459) (0.361) (0.873)
C. Control -7.935 20.55# *** -- --
(0.578) (0.000)#
Voice & A. -- -- 15.51# ** 2.062
(0.014)# (0.526)
Hausman 33.5# *** 6.758# * 27.76# *** 0.988
(0.000)# (0.080)# (0.000)# (0.804)
Sargan OIR 0.013# 1.033# 1.110# 1.134#
(0.907)# (0.309)# (0.292)# (0.286)#
Adjusted -0.065 0.494 0.109 -0.024
[R.sup.2]
[chi square] -- -- -- --
Fisher 14.0# *** 42.97# *** 2.409# * 0.525
Observations 34 51 87 59
Instruments Constant, Total NODA, NODADAC, NODAMD, Grants
Legal origins Religious dom.
English French Christ. Islam
Panel A: Restricted Modeling
Constant -- -- -- --
Gov. Exp. 0.475# ** 0.004 1.062# * -0.43# **
(0.046)# (0.983) (0.097)# (0.011)#
Tax Rev. 0.52# *** 1.16# *** 0.567# *** 0.84# ***
(0.000)# (0.000) (0.000)# (0.000)#
C. Control -0.164 9.567 -- --
(0.972) (0.103)
Voice & A. -- -- 2.710 -1.469
(0.828) (0.733)
Hausman 45.5# *** 17.9# *** 56.78# *** 35.0# ***
(0.000)# (0.000)# (0.000)# (0.000)#
Sargan OIR 1.391# 6.74 ** 1.035# 1.354#
(0.498)# (0.034) (0.595)# (0.508)#
Adjusted 0.083 0.251 0.073 -0.061
[R.sup.2]
[chi square] -- -- --
Fisher 107# *** 40.7# *** 20.13# *** 34.2# ***
Observations 72 72 111 35
Panel B: Unrestricted Modeling
Constant 5.582# ** 15.7# *** 20.405 11.6# ***
(0.031)# (0.003)# (0.184) (0.000)#
Gov. Exp. 0.179 -0.038 0.325 0.061
(0.305) (0.874) (0.570) (0.783)
Tax Rev. 0.34# *** 0.371 -0.212 0.140
(0.000)# (0.200) (0.714) (0.565)
C. Control -1.549 13.8# *** -- -2.309
(0.496) (0.005)# (0.729)
Voice & A. -- -- 6.749 --
(0.532)
Hausman 4.854 2.948 9.767# ** 2.007
(0.182) (0.399) (0.020)# (0.570)
Sargan OIR 1.641# 1.132# 1.365# 1.240#
(0.200)# (0.287)# (0.242)# (0.265)#
Adjusted 0.052 0.273 0.029 -0.006
[R.sup.2]
[chi square] -- -- -- --
Fisher 7.92# *** 4.25# *** 0.450 0.768
Observations 72 72 111 35
Instruments Constant, Total NODA, NODADAC, NODAMD, Grants
Regions Resources
SSA NA Oil Non-oil
Panel A: Restricted Modeling
Constant -- -- -- --
Gov. Exp. 1.003# * -1.32# ** 0.385# * 0.894# *
(0.050)# (0.049)# (0.053)# (0.070)#
Tax Rev. 0.55# *** 2.10# *** 0.154 0.58# ***
(0.000)# (0.000)# (0.212) (0.000)#
C. Control -- -- -- --
Voice & A. -- 22.60# ** -- --
(0.013)#
Hausman 38.0# *** 57.5# *** 0.035 34.7# ***
(0.000)# (0.000)# (0.982) (0.000)#
Sargan OIR 3.768# 0.294# 7.594 * 4.484#
(0.287)# (0.862)# (0.055) (0.213)#
Adjusted 0.054 -0.074 0.878 0.073
[R.sup.2]
[chi square] 91.7# *** -- 14,391# *** 120# ***
Fisher -- 46.6# *** -- --
Observations 155 26 8 176
Panel B: Unrestricted Modeling
Constant 12.843 43.4# *** 7.414# *** 13.340
(0.433) (0.003)# (0.000)# (0.335)
Gov. Exp. -0.105 -- 0.180# *** -0.078
(0.912) (0.000)# (0.917)
Tax Rev. -0.014 0.61# *** 0.092# *** 0.027
(0.982) (0.000)# (0.000)# (0.959)
C. Control -6.986 -- -- -4.247
(0.459) 44.6# *** (0.637)
Voice & A. -- (0.009)# -- --
Hausman 7.002# * 15.8# *** 4.719# * 4.359
(0.071)# (0.000)# (0.094)# (0.225)
Sargan OIR 2.725 * 1.076# 1.514# 3.027 *
(0.098) (0.583)# (0.468)# (0.081)
Adjusted 0.181 0.150 0.818 0.150
[R.sup.2]
[chi square] -- -- -- --
Fisher 0.421 31.9 *** 155# *** 0.176
Observations 118 26 8 138
Instruments Constant, Total NODA, NODADAC, NODAMD, Grants
Stability Landlocked (LL)
Conflict Non-co. LL Not LL
Panel A: Restricted Modeling
Constant -- -- -- --
Gov. Exp. -0.128# ** 0.977# * 0.673# *** 0.506
(0.031)# (0.070)# (0.000)# (0.354)
Tax Rev. 0.066 0.643# *** 0.504# *** 0.919# ***
(0.751) (0.000)# (0.000)# (0.000)#
C. Control -5.07# *** -- -- --
(0.000)#
Voice & A. -- 2.966 -- 14.728# **
(0.793) (0.041)#
Hausman 110# *** 46.60# *** 10.58# *** 32.40# ***
(0.000)# (0.000)# (0.005)# (0.000)#
Sargan OIR 0.381# 3.007# 2.598# 3.138#
(0.826)# (0.222)# (0.457)# (0.208)#
Adjusted -0.101 0.088 0.123 0.147
[R.sup.2]
[chi square] -- -- 113# *** --
Fisher 2e^4# *** 26.06# *** -- 22.40# ***
Observations 13 140 57 103
Panel B: Unrestricted Modeling
Constant -25.1# *** 13.340 -9.278 17.29# ***
(0.000)# (0.335) (0.408) (0.004)#
Gov. Exp. 0.104 -0.078 -0.020 0.103
(0.186) (0.917) (0.933) (0.771)
Tax Rev. 0.75# *** 0.027 0.66# *** 0.131
(0.000)# (0.959) (0.000)# (0.680)
C. Control -18.1# *** -4.247 -17.347 --
(0.000)# (0.637) (0.166) 13.49# ***
Voice & A. -- -- -- (0.000)#
Hausman 90.40# *** 4.359 9.713# ** 12.18# ***
(0.000)# (0.225) (0.021)# (0.000)#
Sargan OIR 0.004# 3.027 * 0.773# 0.160#
(0.945)# (0.081) (0.379)# (0.688)#
Adjusted 0.395 0.150 0.009 0.127
[R.sup.2]
[chi square] -- -- -- --
Fisher 72.74# *** 0.176 20.12# *** 4.565# ***
Observations 13 138 42 103
Instruments Constant, Total NODA, NODADAC, NODAMD, Grants
Africa
Panel A: Restricted Modeling
Constant --
Gov. Exp. -0.310
(0.862)
Tax Rev. 0.475# **
(0.014)#
C. Control -18.134
(0.369)
Voice & A. --
Hausman 84.39# ***
(0.000)#
Sargan OIR 2.074#
(0.354)#
Adjusted 0.115
[R.sup.2]
[chi square] --
Fisher 16.75# ***
Observations
Panel B: Unrestricted Modeling
Constant 14.294
(0.372)
Gov. Exp. -0.501
(0.766)
Tax Rev. 0.005
(0.991)
C. Control -7.932
(0.643)
Voice & A. --
Hausman 4.254
(0.235)
Sargan OIR 1.326#
(0.249)#
Adjusted 0.138
[R.sup.2]
[chi square] --
Fisher 0.090
Observations 144
Instruments
***, **, *: significance levels of 1%, 5% and 10%, respectively.
P-values in parentheses. OIR: Over-identifying Restrictions test.
UMI: Upper Middle Income. LMI: Lower Middle Income. MI: Middle
Income. LI: Low Income. English: English Common-law. French: French
Civil-law. Christ.: Christianity dominated countries. Islam: Islam
dominated countries. SSA: Sub-Saharan Africa. NA: North Africa.
Oil: Petroleum exporting countries. Non-oil: Countries with no
significant exports in petroleum. Conflict: Countries with
significant political instability. Non-co: Countries without
significant political instability. Gov. Exp: Government
Expenditure. Voice & A: Voice & Accountability. Tax Rev: Tax
Revenues. HAC: Heteroscedasticity and Autocorrelation Consistent.
NODA: Net Official Development Assistance. DAC: Development
Assistance Committee. MD: Multilateral Donors. NODADAC: NODA from
DAC countries. NODAMD: NODA from Multilateral Donors. The relevance
of bold values that depict the information criteria is threefold.
(1) Rejection of the null hypothesis of the Hausman test for the
presence of endogeneity. (2) The significance of estimated
coefficients and the Fisher statistics. (3) The failure to reject
the null hypothesis of theSargan OIR test for instrument validity.
Note: The relevance of values that depict the information criteria
is threefold is indicated with #.
Table 5: Robust assessment with fixed capital formation (HAC standard
errors)
Income levels
UMI LMI MI LI
Panel A: Restricted Modeling
Constant -- -- -- --
Gov. Exp. -0.59# * 1.366 2.328 0.957
(0.099)# (0.229) (0.293) (0.265)
Tax Rev. 0.87# *** 0.819# *** 0.775# ** 1.10# ***
(0.000)# (0.005)# (0.026)# (0.000)#
C. Control 4.956 8.550 -- --
(0.000) (0.555)
Voice & A. -- -- 18.691# ** --
(0.045)#
Hausman 58.4# *** 31.97# *** 59.21# *** 48.28# ***
(0.000)# (0.000)# (0.000)# (0.000)#
Sargan OIR 0.627# 15.98 *** 1.504# 1.123#
(0.730)# (0.000) (0.471)# (0.771)#
Adjusted 0.075 0.272 0.068 0.105
[R.sup.2]
[chi square] -- -- -- 104# ***
Fisher 485# *** 29.92# *** 25.83# *** -
Observations 34 57 93 80
Panel B: Unrestricted Modeling
Constant 62.37 26.3# *** 21.89# ** 3.831
(0.405) (0.000)# (0.017)# (0.915)
Gov. Exp. -0.106 -0.099 -0.358 0.555
(0.870) (0.870) (0.785) (0.191)
Tax Rev. -1.375 0.235# * 0.381 0.747
(0.631) (0.074)# (0.166) (0.723)
C. Control -7.194 27.1# *** -- --
(0.676) (0.000)#
Voice & A. -- -- 16.05# ** -6.247
(0.015)# (0.624)
Hausman 27.0# *** 18.34# *** 36.09# *** 5.545
(0.000)# (0.000)# (0.000)# (0.135)
Sargan OIR 0.265# 0.170# 1.209# 1.451#
(0.606)# (0.680)# (0.271)# (0.228)#
Adjusted -0.099 0.573 0.074 0.002
[R.sup.2]
[chi square] -- -- -- --
Fisher 1.096 28.41# *** 2.535# * 0.843
Observations 34 57 93 62
Instruments Constant, Total NODA, NODADAC, NODAMD, Grants
Legal origins Religious dom.
English French Christ. Islam
Panel A: Restricted Modeling
Constant -- -- -- --
Gov. Exp. 0.541# ** -0.267# * 1.301 -0.4# ***
(0.022)# (0.086)# (0.239) (0.000)#
Tax Rev. 0.75# *** 1.29# *** 0.69# *** 1.42# ***
(0.000)# (0.000)# (0.000)# (0.000)#
C. Control -2.740 -2.196 -- --
(0.601) (0.885)
Voice & A. -- -- -8.026 3.475
(0.719) (0.641)
Hausman 71.9# *** 25.8# *** 84.99# *** 70.5# ***
(0.000)# (0.000)# (0.000)# (0.000)#
Sargan OIR 0.506# 25.0 *** 3.389# 3.542#
(0.776)# (0.000) (0.183)# (0.170)#
Adjusted 0.165 0.100 0.014 0.305
[R.sup.2]
[chi square] -- -- -- --
Fisher 133# *** 56.5# *** 36.95# *** 273# ***
Observations 72 81 111 44
Panel B: Unrestricted Modeling
Constant 6.996# ** 28.2# *** 42.601 13.7# ***
(0.026)# (0.000)# (0.453) (0.000)#
Gov. Exp. 0.170 -0.332 -0.237 -0.111
(0.396) (0.279) (0.909) (0.175)
Tax Rev. 0.52# *** 0.088 -0.936 0.48# ***
(0.000)# (0.794) (0.667) (0.000)#
C. Control -4.476 15.187# * -- -0.764
(0.107) (0.067)# (0.828)
Voice & A. -- -- 0.406 --
(0.987)
Hausman 8.400# ** 12.6# *** 24.59# *** 4.334
(0.038)# (0.000)# (0.000)# (0.227)
Sargan OIR 0.025# 2.604# 1.679# 0.054#
(0.874)# (0.106)# (0.194)# (0.815)#
Adjusted 0.132 0.289 0.146 0.319
[R.sup.2]
[chi square] -- -- -- --
Fisher 98.6# *** 2.082 0.495 6.80# ***
Observations 72 81 111 44
Instruments Constant, Total NODA, NODADAC, NODAMD, Grants
Regions Resources
SSA NA Oil Non-oil
Panel A: Restricted Modeling
Constant -- -- -- --
Gov. Exp. 1.833# ** -1.1# *** 0.330# * 1.589# *
(0.031)# (0.009)# (0.090)# (0.036)#
Tax Rev. 0.72# *** 1.75# *** 0.47# *** 0.808# ***
(0.000)# (0.000)# (0.000)# (0.000)#
C. Control -- -- -- --
Voice & A. -- 7.467 -- --
(0.103)
Hausman 83.8# *** 127# *** 1.153 89.88# ***
(0.000)# (0.000)# (0.561) (0.000)#
Sargan OIR 3.740# 0.162# 3.321# 4.755#
(0.290)# (0.921)# (0.344)# (0.190)#
Adjusted 0.050 -0.052 0.872 0.081
[R.sup.2]
[chi square] -- -- 6e^4# *** 150 ***
Fisher 91.3# *** 276# *** -- --
Observations 158 32 8 186
Panel B: Unrestricted Modeling
Constant 23.449 38.4# *** 7.11# *** 21.828
(0.185) (0.002)# (0.000)# (0.145)
Gov. Exp. 0.318 -- 0.13# *** 0.146
(0.795) (0.000)# (0.889)
Tax Rev. -0.235 0.54# *** 0.41# *** -0.092
(0.744) (0.000)# (0.000)# (0.878)
C. Control -2.976 -- -- -3.778
(0.744) (0.687)
Voice & A. -- 27.9# ** -- --
(0.036)#
Hausman 24.9# *** 23.1# *** 0.043 17.1# ***
(0.000)# (0.000)# (0.978) (0.000)#
Sargan OIR 2.692# 0.328# 3.532# 3.005 *
(0.100)# (0.848)# (0.171)# (0.082)
Adjusted 0.045 0.370 0.937 0.127
[R.sup.2]
[chi square] -- -- -- --
Fisher 0.792 57.6# *** 463# *** 0.373
Observations 121 32 8 147
Instruments Constant, Total NODA, NODADAC, NODAMD, Grants
Stability Landlocked(LL)
Conflict Non-co. LL Not LL
Panel A: Restricted Modeling
Constant -- -- -- --
Gov. Exp. -0.112# ** 1.574 ** 1.388# *** 0.460
(0.016)# (0.025) (0.000)# (0.504)
Tax Rev. 0.255 0.784 * 0.69# *** 1.291# ***
(0.166) (0.050)# (0.000)# (0.000)#
C. Control -5.03# *** -- -- --
(0.000)#
Voice & A. -- -1.407 -- 16.26# **
(0.941) (0.047)#
Hausman 115 *** 79.14# *** 32.77# *** 40.55# ***
(0.000) (0.000)# (0.000)# (0.000)#
Sargan OIR 0.243 4.413# 3.909# 5.038 *
(0.885) (0.110)# (0.271)# (0.080)
Adjusted -0.135 0.064 0.070 0.182
[R.sup.2]
[chi square] -- -- 162# *** 28.77# ***
Fisher 5e^4# *** 51.07# *** -- 40.55# ***
Observations 13 149 60 109
Panel B: Unrestricted Modeling
Constant -19.4# *** 21.828 11.966 24.76# ***
(0.000)# (0.145) (0.244) (0.000)#
Gov. Exp. 0.067 0.146 0.033 -0.141
(0.314) (0.889) (0.818) (0.717)
Tax Rev. 0.79# *** -0.092 0.363 0.153
(0.000)# (0.878) (0.196) (0.628)
C. Control -15.1# *** -3.778 -5.617 --
(0.000)# (0.687) (0.536)
Voice & A. -- -- -- 13.857# **
(0.017)#
Hausman 64.8# *** 17.16# *** 2.364 22.79# ***
(0.000)# (0.000)# (0.500) (0.000)#
Sargan OIR 0.017# 3.005# 0.012# 0.183#
(0.896)# (0.082)# (0.909)# (0.668)#
Adjusted 0.310 0.127 0.132 0.108
[R.sup.2]
[chi square] -- -- -- --
Fisher 112# *** 0.373 1.600 2.754# **
Observations 13 147 45 109
Instruments Constant, Total NODA, NODADAC, NODAMD, Grants
Africa
Panel A: Restricted Modeling
Constant --
Gov. Exp. 0.034
(0.987)
Tax Rev. 0.662# ***
(0.000)#
C. Control -23.717
(0.351)
Voice & A. --
Hausman 186# ***
(0.000)#
Sargan OIR 2.831#
(0.242)#
Adjusted 0.087
[R.sup.2]
[chi square] --
Fisher 25.05# ***
Observations 153
Panel B: Unrestricted Modeling
Constant 23.127
(0.157)
Gov. Exp. -0.217
(0.902)
Tax Rev. -0.125
(0.820)
C. Control -6.447
(0.645)
Voice & A. --
Hausman 18.03# ***
(0.000)#
Sargan OIR 1.796#
(0.180)#
Adjusted 0.248
[R.sup.2]
[chi square] --
Fisher 0.266
Observations 153
Instruments
***, **, *: significance levels of 1%, 5% and 10%, respectively. P-
values in parentheses. OIR: Over-identifying Restrictions test. UMI:
Upper Middle Income. LMI: Lower Middle Income. MI: Middle Income. LI:
Low Income. English: English Common-law. French: French Civil-law.
Christ.: Christianity dominated countries. Islam: Islam dominated
countries. SSA: Sub-Saharan Africa. NA: North Africa. Oil: Petroleum
exporting countries. Non-oil: Countries with no significant exports
in petroleum. Conflict: Countries with significant political
instability. Non-co: Countries without significant political
instability. Gov. Exp: Government Expenditure. Voice & A: Voice &
Accountability. Tax Rev: Tax Revenues. HAC: Heteroscedasticity and
Autocorrelation Consistent. NODA: Net Official Development
Assistance. DAC: Development Assistance Committee. MD: Multilateral
Donors. NODADAC: NODA from DAC countries. NODAMD: NODA from
Multilateral Donors. The relevance of bold values that depict the
information criteria is threefold.(1) Rejection of the null
hypothesis of the Hausman test for the presence of endogeneity.(2)
The significance of estimated coefficients and the Fisher
statistics.(3) The failure to reject the null hypothesis of the
Sargan OIR test for instrument validity.
Note: The relevance of values that depict the information criteria is
threefold is indicated with #.
