The influence of China and emerging donors aid allocation: A recipient perspective

    The influence of China and Emerging donors aid allocation: A recipient perspective Eric Gabin Kilama PII: DOI: Reference:

S1043-951X(15)00150-9 doi: 10.1016/j.chieco.2015.11.010 CHIECO 897

To appear in:

China Economic Review

Received date: Revised date: Accepted date:

22 June 2015 18 November 2015 18 November 2015

Please cite this article as: Kilama, E.G., The influence of China and Emerging donors aid allocation: A recipient perspective, China Economic Review (2015), doi: 10.1016/j.chieco.2015.11.010

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT The influence of China and Emerging donors aid allocation:

T

A recipient perspective

SC R

IP

KILAMA Eric Gabin CERDI, Universit d‟ uvergne, France

Abstract :

MA

NU

From the perspective of recipients, the increasing influence of China and emerging donors in the aid landscape represents an opportunity to attract additional resources to finance development and improve their control over their development agenda. This paper investigates how African countries and other LMICs deal with this complex and changing aid landscape and explores how government fiscal behaviors and private agents anticipations regarding aid flows are affected.

CE P

TE

D

We used several measures of fiscal behaviors and we were not able to confirm empirically the fear of traditional donors about a macroeconomic disaster that would follow emerging donors aid allocation. Moreover, the results indicate that economies receiving additional aid flows from China enhance their fiscal response to aid through an increased domestic economy aid absorption rate.

AC

Keywords: China, Africa, Emerging donors, Aid allocation, Ownership, Fiscal policy, Absorption rate of aid

1

ACCEPTED MANUSCRIPT Introduction: The growing influence of emerging economies is leading to more choices and more finances,

T

and developing country governments‟ welcome this situation. Woods (2008) argued that

IP

competitive pressures between donors give governments a certain amount of leverage.

SC R

Analysing the aid landscape in Nicaragua, Roussel (2013) reported the following citation from an interview with a representative from the German Embassy, illustrating the empowerment of

NU

recipient government:

„we were basically told “this is your money…we can talk about it and if we agree, then let's do

MA

it, and if not, just take your money home”. (…) This is a new attitude or a new strength that the government didn't show before.‟

TE

D

The same empowerment is also observed in Ethiopia, Cambodia, Fiji or Zambia by Sato et al. (2011), Greenhill et al. (2013) and Schmaljohann and Prizzon (2014). These recent country

CE P

case studies describe the changing aid landscape and show that emerging donors provide a policy comfort to the recipient governments and diminish the leverage of traditional donors,

AC

who are disinclined to impose policy terms1. Thus, donor diversity would enhance recipient ownership of the development agenda. According to Paulo and Reisen (2010), more than 30 donor countries operate outside the Development Assistance Committee (DAC).

We choose the term "non-DAC donors" to

describe these actors, even if it defines the group by what they are not, rather than by what they are. The most important non-DAC donors included China, India, Brazil, Venezuela, Mexico, Russia, United Arab Emirates, Saudi Arabia, South Africa, Kuwait and Thailand. The category of non-DAC donors represents a broad spectrum of actors with diversity in various aspects of development cooperation, from strategic priorities to regional and sectorial focus to

2

ACCEPTED MANUSCRIPT institutional arrangements 2 . However, they have in common an at least partial rejection of DAC-related principles and practices and a rhetoric promoting a different kind of engagement

T

with the countries to which they provide assistance. The 2005 Paris Declaration on Aid

IP

Effectiveness emphasised that aid should support ownership, harmonisation, alignment, results

SC R

and mutual accountability. Non-DAC donors like China have sets similar principles to govern their aid, but a key difference is in the application of conditionality. The members of the DAC are practised in banding together to impose economic and political conditions (policy reforms,

NU

structural economic changes or good governance) to aid allocations (Brautigam, 2011).

MA

Moreover, as South-South Cooperation is based on mutual interest and local ownership, another difference is the practice among the DAC donors to develop country assistance strategies which sometimes reflect the donors‟ goals more than those of the country they are

TE

D

assisting.

If concerns about aid fragmentation, increasing transactions costs, or debt sustainability are

CE P

preoccupying issues for DAC donors, recent country case studies (Greenhill et al. 2013; Sato et al. 2011; Roussel, 2013) about recipients of non-DAC aid showed that these concerns did not

AC

appear as priorities for recipient governments. The overwhelming priority is to raise additional resources to finance development despites the associated transactions costs. Some recipient countries are taking a strategic approach to manage funds from traditional and emerging donors. They intend to use the presence of new donors to increase their negotiating capital in relation to traditional donors, and secure better outcomes in relation to their priorities (Greenhill et al., 2013). The purpose of this analysis is to better understand the implications of the changing aid landscape for developing countries. Our analysis will explore both recipient governments and private agents perspectives regarding the “welcoming emerging donors” strategy.

3

ACCEPTED MANUSCRIPT First, we would observe what happened to fiscal variables during the last decades in countries that have received significant amount of aid flows from China and other non-DAC donors. Our

T

paper investigates the behavior of a comprehensive set of fiscal variables (government

IP

spending, the level of tax revenues, borrowing behavior) during the period 1980-2010 on a

SC R

sample of LICs and other developing countries receiving aid flows from DAC donors and emerging donors. The objective is to find if the presence of non-DAC donors in the aid landscape as new development finance providers is leading to specific fiscal behaviors from

NU

recipients or similar to those demonstrated by Ouattara (2006) about DAC aid flows.

MA

Evaluating the absorption of aid flows, Berg et al. (2007) and Foster & Killick (2006) found that most aid appears to finance capital flight, rather than an increase in net imports. In

D

countries where political institutions are weak or corrupt, aid transfers may lead to capital flight

TE

and undermine growth. Unlike the western donors providing budget support which often finances capital flight from developing countries to high income countries, emerging donors

CE P

like China primarily deliver aid through discrete projects. This could have positive implication for the effective absorption of aid. Furthermore as the debate about emerging donors is

AC

analysed by the literature of aid effectiveness largely focused on traditional donors behaviors, the value and the implication of the contribution of the new southern donors to development cooperation seems to be under-appreciated by economists. To fill this theoretical gap, Lin and Wang (2015) propose to look the contribution of China and other emerging donors to development cooperation through the New Structural Economics theory. In fact, China, Brazil or India like other rising donors are utilising their own tacit knowledge and intimate experiences of structural transformation in the past 40 years in their approach of development cooperation. The originality precisely lies in the close linkage of development assistance to the provision of resources, foreign direct investment, trade flows and technical knowledge transfer.

4

ACCEPTED MANUSCRIPT They finance infrastructure projects targeting and addressing African countries and other LICs bottlenecks to growth (water; electricity; road and rail; air transport; and telecom). Overall,

T

infrastructure resources committed to Africa by these countries jumped from US$1 billion per

IP

year in the early 2000s to over US$10 billion in 2010. Chen (2013) found that China alone

SC R

accounts for 34 per cent of all aid to infrastructure in SSA, higher than other multilateral and bilateral donors. However, even if South-South development cooperation is rather perceived as an opportunity for doing business, trade and investments for new donors, developing countries

NU

believe that working together with emerging donors could enhance their capacity in self-

MA

development.

The second part of our analysis in this paper seeks to find out how the private sector behavior

D

regarding aid flows is also influenced by the specific characteristics of non-DAC donors

TE

activities.

CE P

Using reported OECD-CRS data on non-DAC donors aid allocation and taking advantage of the comprehensive dataset AidData on China aid to Africa, we succeed to build a representative view of emerging donors‟ influence on development finance. To our best knowledge, there is

AC

no empirical work that formally addresses the impact of emerging donors aid allocation on recipient economy3.

Our results indicate that there is no evidence supporting the macroeconomic disaster scenario surrounding the aid allocation from non-DAC to LICs. Moreover our empirical evidences suggest that emerging donors offer more opportunities to finance development through additional support for public investment or provision of Turn-key infrastructures projects and enhance development capacity through specific knowledge transfer. However, the analysis of the specific case of African countries shows that their level of borrowing is increasing and tax exemptions still a problem. Addressing the question about how the empowerment of recipients

5

ACCEPTED MANUSCRIPT with the competition between donors is being put to use, we performed an analysis of the absorption rate of aid flows by private agents. We investigate whether African countries, with

T

the Chinese influence expanding, are doing worse or better with increasing aid flows over the

IP

period 2000-2011. We find a robust positive relationship between the number of China projects

SC R

realized in the economy and the absorption rate of aid flows by African economies. This evolution of private agents behavior regarding aid flows - aid appears to finance more imports

NU

than capital flight – is confirmed outside Africa.

The structure of this paper is as follows. Section 1 presents some stylized facts about fiscal

MA

behavior of aid recipient governments and the context of the emergence of non-DAC donors. Sections 2 and 3 discuss respectively the fiscal authorities‟ responses to the increasing

D

influence of emerging donors in development assistance and the consequence for the domestic

TE

economy aid absorption through a rigorous empirical analysis. In section 4 we present our

AC

CE P

conclusions.

6

ACCEPTED MANUSCRIPT Section 1. Context and Recent stylized facts The development agenda took a major shift in priorities from production and economic sectors

IP

T

to social sectors, accentuated by the Millemnium Development Goals. Describing the sectorial aid evolution, Frot and Santiso (2010) explained that while economic infrastructures spending

SC R

have declined, the social sectors represent more than 60% of the total number of projects, up from 30% in the 1970‟s. Since 2000, infrastructure aid per year has dropped from an average of

NU

29.5% of total development assistance over the period 1973–1990 to an average of just 10% of total bilateral aid of members of the Development Assistance Committee (DAC). In response,

MA

some low-income countries (LICs) and lower-middle income countries (LMICs) have adopted a deliberate and systematic strategy to welcome new donors to secure additional resources to

D

finance development, and emerging donors are seen as particularly valuable in helping them to

TE

meet this objective (Sato et al., 2011; Roussel, 2013). Although information on development

CE P

assistance of emerging donors is rather patchy, and it appears relatively small (less than 0.10% of total official development assistance in the last five years), at the country level their importance as development partner is increasing rapidly, with finance on non-concessional

AC

terms used mainly to fund infrastructure projects such as roads and communications. It seems that the neglect of economics infrastructure by traditional donors, paired with the departure of some traditional donors due to dissatisfaction with governance and human rights standards4, and the recent falling of traditional official development assistance (ODA) have opened up the way for emerging donors in these sectors and helped them to extend their influence in development cooperation with low income countries (LICs) and other developing countries.

7

ACCEPTED MANUSCRIPT An observation of the highly fragmented aid market may lead to the conclusion that the arrival of emerging donors exacerbates the situation and would undermine macroeconomic

T

management in recipients, because of the heavy transactions costs associated with donor-

IP

recipient coordination. Furthermore, the specificities of emerging donors‟ aid allocation

SC R

regarding the DAC principles suggest that they will cause some changes on recipients‟ fiscal behaviors. But there is no consensus (evidences) in the way this will influence the existing aid

NU

system.

At the height of contention is the non-interference policy of emerging donors; they are accused

MA

to support rogue states (Naim, 2007). Another aspect of their non-interference policy, emerging donors traditionally do not apply conditionality on their aid, while traditional donors are known

D

for requiring specific changes to governance and macroeconomic policies in recipient

TE

countries5. Some traditional donors fear that countries with weak rule-of-law, particularly those with abundant natural resources, have gained greater freedom to circumvent the demanded

CE P

policy and institutional reforms. Some of them argued that emerging donors are encouraging poor policies, lowering standards and increasing the debt burden of recipient countries.

AC

Manning (2006) discusses the possible risk that loans from emerging donors to LICs may prejudice their debt situation and may waste resources on unproductive investments. This paper exposes some of the present fiscal management challenges faced by countries where non-DAC donors are becoming more powerful. Previous studies have found that foreign aid might influence government spending, the level of tax revenue and borrowing behavior (Morrissey, 2012) 6 . Thus our analysis will focus on the influence of emerging donors‟ aid allocations on the recipient government‟s fiscal behavior, in terms of the decisions between various sources of revenue (e.g. taxation and borrowing), and areas of expenditure (e.g. public investment and recurrent government expenditure).

8

ACCEPTED MANUSCRIPT Even if the divergence between traditional donors and emerging donors is over-stated (Kragelund, 2014) as they are all concerned by their economic benefits, by delivering aid that is

T

“non-conditional” to political and environmental standards or the sustainability of recipients‟

IP

debt, emerging donors are including a moral hazard on recipients‟ relationship with traditional

AC

CE P

TE

D

MA

NU

SC R

donors that could undermine established donors norms and efforts to improve aid effectiveness.

Figure1: Correlation Graphs: Emerging donors influence in ODA and Evolution of fiscal behaviors in recipients Source: Author's calculation

9

ACCEPTED MANUSCRIPT As showed by Figure 1, the increasing aid from emerging donors has coincided with some deterioration in fiscal variables during the last decade, justifying some traditional donors

T

concerns about the increasing influence of non-DAC donors.

IP

It seems that there is a positive correlation between short-term debt and non-DAC donors aid

SC R

surge; as the recipients borrowing opportunities increase significantly, recipients have less incitation to enhance domestic resource mobilization. Furthermore, emerging donors aid is

NU

correlated to additional government expenditures and current account deficit. These observations suggest that the presence of non-DAC donors may induce different

MA

recipients' fiscal behaviors from those described by Ouattara (2006) about the implications of traditional ODA. In fact using panel data over the period 1980–2000, he found that there is a

D

substitution between borrowing and foreign aid. Moreover, aid flows exert a positive impact on

TE

government developmental expenditure and a negative significant impact on non-

CE P

developmental expenditure.

The other side of the emerging donors activities relies on private investment and commercial

AC

transactions, influencing directly the private sector which is exposed to competition, new advances in technology and modern labour skills. China for example is largely concentrate on state-to-state deals, extractive industries and infrastructure development, while India presence in Africa is largely commercially driven, private, and facilitated by the Export–Import Bank of India and the Confederation of Indian Industries (Taylor, 2012). But in this paper we would only investigate the indirect effect on macroeconomic management of aid flows and explore how the changing aid landscape is influencing private sector behavior regarding aid flows.

10

ACCEPTED MANUSCRIPT Section 2. Empirical evidences on government fiscal behaviors

Empirical and theoretical background

T

2.1.

IP

Fraser and Whitfield (2008)
 described the relationship between recipients and donors in

SC R

development cooperation as a negotiation process in which both sides have a set of (potentially divergent) interests and priorities they need to negotiate. The “competitive pressure” that new donors‟ growing presence in the aid landscape exerts on the traditional donors development

NU

policy regime affects the policy space of recipient governments.

The biggest challenge

presented by China and other emerging donors is the fact that foreign aid and business

MA

engagement from the Western donors has not been very successful in fostering growth in recipient countries. Even if (all) donors often compete for influence over the recipient country‟s

TE

D

policies (Acharya et al., 2006), emerging donors will give countries a more profitable option. Their approach to development cooperation clearly offers opportunities. These benefits include

CE P

greater ownership, and more equal partnerships, lower transaction costs, a new emphasis on infrastructure and productive activities, and policy space (Brautigam, 2011). Greenhill et al.

AC

(2013) highlighted the need for additional resources
 to finance development is a key priority for recipient governments, and emerging donors are seen as particularly valuable in helping countries meet the country‟s substantial infrastructure needs. But the presence of emerging donors also entails some risks; China‟s large-scale finance arrived just as African countries were finally successful in getting multilateral debt relief through the Highly Indebted Poor Countries (HIPC) program. Donors have worried about a new debt burden. Manning (2006) explained that greater access to aid might once again condemn recipient countries to unsustainable debt. In the paper it is clearly recognized that emerging donors and traditional donors have in some aspects similar aid allocation motives and processes. Aid from emerging donors and the DAC 11

ACCEPTED MANUSCRIPT countries are often programmed in similar ways, including project support, technical assistance, food aid, debt relief, humanitarian assistance and so on. The members of the DAC and the

T

Chinese also both have sets of principles that are supposed to govern their aid. The key

IP

difference is in practice and approach proposed by emerging donors.

SC R

This paper is based on observations of on-the-ground operations of traditional donors in several recipients (Cambodia, Ethiopia, Nicaragua, Zambia, Fiji) describing their perceptions and the

NU

incidence of the presence of emerging donors in their relationship with recipients. These previous studies discussed the mechanism of how the approach of emerging donors is giving

MA

policy space to fiscal authorities in recipient countries. Moreover, Manning (2006) and Naim (2007) already established a causal link between the terms of non-DAC development assistance

D

and deterioration of fiscal variables in developing countries when they are talking about debt

TE

unsustainability in countries welcoming emerging donors aid. From that literature we propose

2.2.

CE P

to move from interviews-based explanations to an explicit empirical analysis.

Baseline Specification and Data

AC

a) The model

Following Ouattara (2006), we perform a panel data analysis using the specification above: Yit  ai    NDACit  Z it   t  eit (1)

This paper estimates several dynamic panel equations linking a given fiscal outcome with the increasing influence of emerging donors in development assistance while controlling for standard determinants of the given fiscal variable. First, we analyze the government spending behavior through the public investment and government consumption expenditures. Two other variables we are describing in our framework are the level of domestic borrowing, and the current account balance. Finally, we explore tax revenues. We distinguish between different types of taxes since it would make our analysis richer in terms of understanding if the

12

ACCEPTED MANUSCRIPT governments changed the composition of tax revenue efforts by collecting less a particular tax (trade tax) to attract emerging donors (notably China). We estimate our equation for each

T

policy variable.

IP

Equation [2] is a dynamic specification and would be considered when administrative inertia is

SC R

characterizing a fiscal variable. Government administrations are constrained by budgets, and the current budget largely determines the next period‟s appropriations.

lthough such inertia

has been argued to provide some stability and predetermines fiscal spending (Schuknecht,

NU

2000), the presence of lagged dependent variables and the country-specific effects renders the

MA

OLS estimator biased since the lagged dependent variable is correlated with the error term: Yit  ai  Yit 1    NDACit  Z it   t  eit

(2)

D

In order to deal with this issue, we could remove the fixed effect by differencing

TE

Yit  Yit 1    NDACit  Z it  t  eit (3)

CE P

The problem here is that the differenced residual, eit , is correlated with the lagged dependent variable, Yit 1 , leading to non-consistent OLS estimates. Angrist and Pischke (2008)

AC

suggested that one solution is to use Yit  2 as an instrument for Yit 1 in [3]. However as our first-differencing strategy is only efficient for large panel, we would prefer fixed effects estimator for small samples and to control for unobserved time-invariant countries characteristics influencing countries fiscal behaviors. Furthermore, we also need to consider two other issues: the non-stationarity of macroeconomic variables (Martins, 2011) and the potential cross-section dependence (Eberhardt & Teal, 2011) 7 . Thus, this paper applies the Driscoll and Kraay (1998) standard errors correction for cross-sectional dependence.

13

ACCEPTED MANUSCRIPT The paper also presents results of the System-GMM estimator with Windmeijer‟s (2005) correction of standard errors for finite sample bias 8 because of his popularity in empirical

SC R

IP

suggesting that “system" GMM estimators can be seriously biased.

T

analysis despites recent research (see Bun and Windmeijer (2010) and Hayakawa (2007))

The econometric models used control for several variables that ensure that the emerging donors influence‟ effects are well identified so that any shift in fiscal variables associated with a period

NU

of increase of NDAC aid must be interpreted as a discretionary fiscal policy by the recipient

MA

government.

Our key independent variable is the measure of emerging donors influence in development

D

assistance proxied alternatively by the volume of ODA from emerging donors and the number

TE

of Chinese projects implemented (only for the African sample). Because the recipients are engaged with both groups of donors, we control for the existing relationship between the

CE P

recipient and the traditional donors as separate units; that said we include a proxy of DAC aid fragmentation. By including this covariate instead of the overall aid flows received from DAC

AC

donors, we would be able to control at the same time for the total aid received from DAC and the transactions costs faced by the recipients9. Following Kimura et al. (2012) we built the Herfindhal-Hirschman index (HHI) of DAC aid shares. The HHI is calculated by taking the sum of squared aid shares of all traditional donors: N

HHI   si2 , i 1

where donor i's aid share in total aid received is defined as si  aid i . The DAC donors total .aid aid fragmentation variable is obtained by subtracting the HHI from 1.

14

ACCEPTED MANUSCRIPT Because aid flows are tied with commercial transactions, an indicator of average trade intensity with emerging donors is included. To estimate the influence of the bilateral trade, we use the

T

International Monetary Fund's Direction of Trade Statistics (DOTS). Following the empirical

IP

literature on the determinants of government consumption, tax revenues and budget deficits in

SC R

developing countries (Rodrik, 1998; Keen and Lockwood, 2010; Combes and Saadi-Sedik, 2006), we control for the level of economic development (GDP per capita provides a proxy for the overall level of development within a country and also represents a measure of the total tax

NU

base), the level of imports (imports in relation to GDP is included since internationally traded

MA

goods are still taxed directly in many developing countries), or the agriculture value added (the share of agriculture value added in GDP is included because the industrial sector of the economy is easier to tax than agriculture (Gupta, 2007)). The other covariates included are the

TE

D

inflation rate, the financial openness, the savings ratio, the government size, the debt service

AC

b) Data

CE P

and population.

The data on control variables used in this article were mainly collected from the IMF‟s World Economic Outlook (WEO) and the World Bank‟s World Development Indicators (WDI). Our data cover 82 LICs and LMICs over the period 1980–2010. Data on African countries tax revenues are provided by the African Economic Outlook Fiscal database. The size of the samples varies due to data availability with respect to our dependent variables.

The first part of data on foreign aid of non-DAC donors come from the OECD online statistics database, which provides also data on DAC donors aid flows. The overall “non-D C aid” data is estimated by the OECD based on various sources but incomplete as major non‐DAC donors

15

ACCEPTED MANUSCRIPT (China, India, Brazil, the Russian Federation, etc…) do not publish detailed data on their aid activities. In 2010, the PLAID (Project-level Aid) and Development Gateway compiled idData is designed to address some of the limitations of the OECD-CRS dataset.

T

“ idData”.

IP

A major advantage of AidData is that it includes more data from non-DAC donors. Data were

SC R

collected from various sources, such as annual reports, media reports, public websites or the statistical agencies of the donors. The last AidData project "China-Africa aid database" (Strange et al., 2013) compiled all Chinese development finance to Africa from 2001 to 2011

NU

using a media-based data collection methodology. Given that much of the discussion about new

MA

donors has centered on China‟s role in development cooperation, and the differences between its approach to development cooperation and the DAC principles, we performed robustness analyses using the data on China Aid to Africa.

TE

D

Strange et al. (2013) created a comprehensive database of Chinese development finance flows to Africa from 2000 to 2011. They estimated that China‟s financial commitments to

frica

CE P

accounted for approximately US$75billion through 1673 projects from 2000 to 2011, of which approximately US$ 15 billion is comparable to Official Development Assistance according to

AC

the OECD definition. Even if Chinese development finance flows do not easily align with the well-defined OECD-DAC definitions of Official Development Assistance (ODA) and Other Official Flows (OOF), they provided a Classification of projects as ODA-like or OOF-like. Based on Kitano and Harada‟s (2015) approach, which provided the most recent estimations of Chinese OD , China‟s net foreign aid between 2001 and 2013 is US$ 34,5 billion.

s the

frica share in China aid is around 45.7% (China‟s White Paper, 2011), it corresponds to approximately US$ 15,7 billion which is comparable to the amount of Chinese “OD -like” activities estimated by Strange et al. (2013) in AidData. If the methodology (media reports) is quite unusual we think that their estimations are not spurious. However, given the conceptual issue related to the definition of Chinese foreign aid, we use the “Number of projects

16

ACCEPTED MANUSCRIPT implemented” by China (and classified “OD -like”) as a proxy of Chinese influence in the aid landscape.

T

We remind that in this paper our goal is to capture the effects of the existence of alternative and

IP

active sources of development finance outside the DAC on the behaviors of less developed

SC R

countries governments. Thus taking the number of Chinese projects implemented could be an interesting alternative to measure the presence of China in Africa.

NU

Even if OECD-CRS data and AidData on China aid to Africa do not give the full picture of the

MA

development of finance activities of emerging donors, we believe that in using both datasets, we can build a representative view of recipient country perceptions of the new balance in the

TE

D

aid landscape.

CE P

2.3. Baseline Estimates

Table 1.A and 1.B present the results for the various fiscal outcomes10. Table 1.D reports the

AC

results for the sample of African countries with a comparison between China aid and DAC donors‟ aid. Our comments are based on First Differences-IV estimates with Driscoll-Kraay correction. But given that FD-IV strategy, using Yit  2 as an instrument, requires at least three periods to obtain data (t; t-1; and t-2), we alternatively used fixed effect estimator with Driscoll-Kraay correction as explained above. The first equation reports results for public investment behavior. The coefficient on NDAC is significant which shows that government investment as a share of GDP increases with additional resources from emerging donors. This result is statistically significant and robust across specifications. Recipient governments use aid from DAC donors (Table 1.D) and the

17

ACCEPTED MANUSCRIPT additional resources from emerging donors to increase public investment and finance development (Table 1.A). This finding is related to the LICs‟ objective to maintain their

T

support to public investment and prevent a collapse in ODA through a diversification of official

IP

donors. The positive and statistically significant sign of the coefficient on Aid Fragmentation

SC R

Index confirms that strategy. However, as Chinese do not allocate much of their aid through budget support, we found no significant effect of Chinese aid on public investment of African countries (Column 1 Table 1.D). The results of Public Investment equation also indicate that

NU

the availability of savings and the imports (through increases on tax revenues) have a positive

MA

and significant impact on investment.

Turning to the Government Consumption Expenditures variable, the results show that there is

D

no evidence that emerging donors aid flows increase consumption expenditures (Column 2

TE

Table 1.A). Moreover, as explained by Brautigam (2011), Chinese development assistance is simpler and does not overstretch the weak capacity of many African countries (provision of

CE P

turnkey infrastructure projects) and Chinese experts do not cost much. The estimated coefficient on China aid variable shows that in comparison to DAC donors who commonly

AC

require skilled staff from African ministries to work in their own country offices, China seems to reduce administrative costs related to aid management (Table 1.D). The positive and statistically significant sign on Aid Fragmentation index captures the adverse effect of donor plurality and confirms our finding.

We now focus on the question of borrowing behavior. In the full sample, our regressions‟ results show that the coefficient on NDAC variable has no statistically significant sign, suggesting that non-DAC aid allocation is not on average associated with a change in government borrowing behavior (Table 1.B). However, the analysis of African governments borrowing behavior showed that the increasing partnership with China is associated with an 18

ACCEPTED MANUSCRIPT increase of the level of borrowing (Table 1.D column 3). The Chinese government and the China Eximbank offer several different kinds of loan finance at competitive commercial rates

T

similar to those charged by export credit agencies around the world. If this finding does not

IP

confirm the fear of a rise of debt unsustainability in countries welcoming emerging donors‟ aid,

SC R

expressed by Manning (2006), Naim (2007) and others, this issue should not be ignored by African governments. Our results also confirm the importance of open financial market for the

NU

lending–Borrowing ratio.

Findings about current account balance show two different effects of the influence of emerging

MA

donors. The negative and significant coefficient on NDAC variable suggests that an increase of non-DAC influence in aid allocation leads to an increase of current account deficit (Table 1.B).

D

We know from Buffie et al. (2010) that the current account deficit has to increase by the same

TE

amount as aid to realise a complete transfer of resources. It seems that countries receiving additional resources from emerging countries also increase the transfer of aid resources in the

CE P

economy, either directly via imports or indirectly via increased public expenditure.

The

African case is different as current account deficit is a persistent feature. As such, countries

AC

with high debt service cannot sustain a high current account deficit and they are urged (by the IMF) to implement reforms and generate revenues to service the debt. The results of Table 1.D also show that China with important infrastructure projects implemented helps African countries to reduce the consequence of low savings on current account balance.

Finally we describe the effort put by the government in collecting taxes when they are receiving additional resources from non-DAC donors. In fact, to evaluate fiscal effort, we disaggregate taxation between fiscal potential and real fiscal effort as explained in Brun et al.

19

ACCEPTED MANUSCRIPT (2009) 11 . The coefficients on NDAC show an improvement of recipient government fiscal effort. Aid would affect the level of tax revenue, either because it influences tax effort (because

T

policy reforms associated with conditional lending) or the tax base. Our findings here could be

IP

explain by the fact that in the case of emerging donors, their aid allocation is tied to foreign

SC R

direct investment and commercial transactions, increasing the tax base in recipient countries. Our analysis of disaggregated tax revenues (income tax revenues and tax revenues on

NU

international trade) shows declining trade-related tax revenues in African countries as their number partners is increasing and as a result of trade liberalization (Table 1.D column 7).

MA

African countries tax administrations should end tax exemptions on aid-funded goods, services and personnel, making aid more conducive to effective domestic resource mobilisation by

D

generating new fiscal revenues and also by sending a signal that tax systems are not open to

TE

negotiation (African Economic Outlook, 2010). If the new partners of African countries like the old ones are sensible to reduction of trade taxes, the partnership with NDAC donors like

CE P

China associated with trade and outward FDI will allow African countries to increase their tax base, and through a learning process proper to South-South cooperation enhance their fiscal

AC

effort by moving capacities towards productive sectors contributing most to domestic revenue (Table 1.D column 8). Table 1.D also reveals two other interesting results. When the primary deficit is increasing (sign of an expansive fiscal policy), the fiscal effort also increases. To assume the debt service governments improve their tax revenues mobilization. And with high inflation (an earlier sign of expansive monetary policy), the real value of tax revenues is reduced and thus lower tax revenues mobilisation.

20

ACCEPTED MANUSCRIPT Section 3.

Indirect effect on domestic economy aid absorption

As discussed in introduction, the existing literature points that donor plurality encourages local

T

ownership of the development agenda. Although there seems to be an agreement that this

IP

competition and plurality empowers the local government, the important question of how this

SC R

empowerment is being put to use has not been effectively addressed. One important contribution of this paper is to provide a comprehensive study that could inform this issue

NU

through an analysis of the absorption rate of aid flows, considering that with more partners and a better control over the development agenda a government can improve the absorption of aid

MA

by the domestic economy. Below, we provide a theoretical framework describing how the fiscal authority can use the presence of emerging donors to influence the anticipations of

D

private agents about their wealth allocations, thus the degree to which aid finances an inflow of

TE

external resources (imports) rather than an accumulation of foreign reserves.

CE P

3.1. A theoretical framework

AC

Since Buffie et al. (2010) we are aware of private sector concerns about the durability of aid and the government's capacity for expeditious fiscal retrenchment, and the implications for aid absorption. Diversification of the type of donors would help private sector to believe that the aid surge is not temporary, and larger fiscal deficits and rapid money growth will not loom on the horizon. Given that our purpose is to capture an indirect effect of the presence of non-DAC donors on private sector behavior regarding aid absorption, we do not provide a detailed analysis of the nature of private sector in developing countries.

21

ACCEPTED MANUSCRIPT Our model derives from the conceptual framework in Buffie et al. (2010) and Berg et al. (2007):

T

Let suppose that all economic decisions in the private sector are assumed to be controlled by a



U  E0   t



Ct1 1

is a constant elasticity of substitution (CES)

NU



where Ct   (CtT )   (1   )(CtN )

t 0

1   

SC R

composite bundle of tradable and non-tradable goods, thus:

IP

representative agent who maximizes his expected lifetime utility and has preferences over a

MA

aggregator function, and ω is the weight households place on tradable consumption. The elasticity of substitution of consumption between tradables and non-tradables is 1/(1+µ).

D

The private agent receives labor income, rents capital to firms, and makes investment decisions.

TE

In addition, the private sector receives lump-sum transfers from the government. Thus, the

CE P

private agent chooses asset holdings and expenditure that maximize his utility with the following wealth and budget constraints (WC and BC respectively): W  m   p   b  F  e p   b  m BC : W  pC  g  r     e

AC

WC :

where χ=ė/e is the rate of currency depreciation, m≡M/e, M is domestic currency, r is the real interest rate, g is real lump-sum transfers received from government, foreign currency is F, and government bonds is B. Bonds are indexed to the price level P, so B=Pb, where b≡B/P.

- Aid, public sector, and reserve accumulation When aid flows increase from X0 to X1 at t=0, the government and the private sector make expectations about the end of the aid surge with probabilities pg and pp. These probabilities determine the proportion of the increased aid spend by the government, or used as buffer stocks 22

ACCEPTED MANUSCRIPT in central bank reserves, but also the success of the fiscal management policy of government due to the credibility level accorded by the private sector. Thus we have, ψ<= 1

T

Public transfer: g1  g 0    X 1  X 0  

Z  1    X 1  X 0 

IP

Reserves:

SC R

ψ determines the fiscal management scenario chosen by the government, and according to Buffie et al. (2010) even in the best-case scenario the success of aid surge management

NU

depends on private agent expectations about government capability, and fears about a future period of large fiscal deficits and high inflation, while the government is struggling to curtail

MA

expenditure after the end of the aid surge.

D

Following the "absorb and spend" scenario of Hussain et al. (2009) the government spends the

TE

extra aid inflow, and the central bank sells the foreign exchange in the currency market -

CE P

corresponding to Z  0 and ψ=1:

(1) In the full credibility case, the public sector budget constraint is thus

AC

p  p b  Z  g1  rb  X 1  m with b  Z  0 e e p m  g1  rb  X 1  m e

m 

(2) low credibility case: As shown by the wealth constraint, the private sector divides its wealth between domestic currency M, foreign currency F, and government bonds B. Therefore, in a low credibility period, private agents believe that the aid surge is temporary, and have concerns about the government's capacity for expeditious fiscal retrenchment (Buffie et al. 2010). They move their wealth allocation towards F and M, generating capital flight and high inflation. Thus, the public sector budget constraint becomes:

23

ACCEPTED MANUSCRIPT m  g1 

p p rb  X 1  b  m e e

A part of the fiscal deficit is now financed by issuing debt. The intuition here is that the

IP

T

increase in current account deficit (CAD) will be higher in the full credibility scenario: CADFC > CADLC,

CAD = C+g – Investment – (Transfert+Income) - (rt-1-1)bt-1/π ;

SC R

where

An increase in aid flows can serve some combination of three purposes: an increase in reserve

NU

accumulation, an increase in capital outflows, and an increase in the non-aid current account deficit. If the private sector fears that after the aid surge there might be a period of large fiscal

MA

deficits and high inflation, their expectations could lead to capital outflows.

D

Given that the success of macroeconomic management of aid depends also on private sector

TE

expectations, the presence of emerging donors could be perceived by private sector as a signal that there would not be a collapse in aid flows and that the government by increasing his

CE P

control over the development agenda would also perform better fiscal management. The expression of that change in expectations can be measure by the absorption rate of aid flows.

AC

Domestic economy actors‟ behavior regarding aid flows would also depend on the capacity of the government to affect donors to specific sectors and form an implicit division of labour between donors as a starting point; traditional DAC donors into the social sectors and non-DAC donors into infrastructure for example. Recipients could also use technical cooperation from emerging donors to enhance their government's ability to manage their development priorities on projects for which a grant has already been received from traditional donors, thereby liberating funds otherwise allocated to the technical component. The freed-up resources can then be invested elsewhere.

24

ACCEPTED MANUSCRIPT 3.2. Empirical analysis Using the econometric model provided in our baseline framework, we estimate the Eq(3) with a

IP

T

new dependent variable, the “rate of absorption of an increase in aid” defined as the change (∆) in the current account (excluding aid) deficit as a share of the change in aid inflows (Hussain et

SC R

al., 2009):

NU

Absorption of aid ∆Non-aid current account deficit∆Aid.

For the measure of absorption rate of aid flows we used the Net Aid Transfers (NAT) data

MA

collected from the Center for Global Development (CGD). Arguing that OECD-CRS data are not appropriate to estimate the „true‟ amount of aid, Martins (2011) collected from Central

D

Banks for his analysis on aid absorption and spending in African countries. The NAT dataset

TE

also tries to overcome this aid measurement issue with a broader coverage.

CE P

Previous studies have found that the response to an aid surge is bad in countries with weak records of macroeconomic stability (Hussain et al, 2009), so we control for the existing debt

AC

burden and the current level of inflation. However, the absorption rate would also depend of additional factors including the financial openness, the government expenditures and the economic structure (agriculture versus industry). Table 2.A presents the results for the System-GMM, First difference, and Fixed Effects estimators of aid absorption rate equation. The coefficient on NDAC variable is positive and statistically significant, suggesting that countries welcoming emerging donors‟ aid increase their aid absorption rate. These findings sustain the hypothesis that the presence of emerging donors in the development cooperation would strengthen the ability of recipients to engage with donors on their own terms and finally use their overall funds more effectively. Our results are robust to the Driscoll-Kraay standard errors correction.

25

ACCEPTED MANUSCRIPT We faced some data limitations with the evaluation of emerging donors activities since some important non-DAC donors clearly resent the traditional dominance of the DAC and do not

T

report their activities to the DAC. China and India, in particular, frame their financial,

IP

economic and technical support to other emerging and developing countries as South-South

SC R

cooperation. They do not want to be perceived as aid donors but rather as partners. Moreover, they are reluctant to closely coordinate their aid activities with other donors if doing so compromises their policy autonomy. However, the most important in our analysis was to

NU

capture the perception of the changing aid landscape by recipients here estimated by the

MA

amount of foreign aid available outside the DAC group of donors. Table 2.B presents specific findings related to the increasing influence of China in Africa. The

D

coefficient on China Aid variable is statistically significant. The presence of China as a strong

TE

development partner is well perceived by private agents and changes their expectations regarding the fiscal management of aid flows by the government. Furthermore by exposing

CE P

African companies to competition, new advances in technology and modern labour skills, they would become more efficient (Taylor, 2012). African governments could potentially use the

AC

opportunity of an increased presence of corporates from India or China in Africa as sources of appropriate technology, skills and advice for economic development. The presence of new donors outside of the DAC, with an alternative in development cooperation policy, increases the bargaining power of African countries over the development agenda. Greenhill et al. (2013), Sato et al. (2011) and Roussel (2013) found that fiscal authorities of some recipients do not want new donors to cooperate with traditional donors; they prefer to deal with them separately to control the concurrence among donors and increase their empowerment over the development agenda. Our results indicate that during the last decade African countries succeed to translate the presence of emerging donors into a strategic policy to improve their macroeconomic management of aid flows. 26

ACCEPTED MANUSCRIPT Our analysis also indicates other interesting findings about aid absorption. The first one concerns public investment level. The coefficient on the Public Investment variable is negative

T

and statistically significant, describing a negative effect of public investment on the absorption

IP

rate. In fact, a substantial percentage of aid inflows went to finance public investment

SC R

expenditures leading to a crowding out effect of the private sector that reduces the absorption rate of aid. Furthermore, as suggested by the negative and statistically significant coefficient of Agriculture Value Added, less the economy structure is oriented towards manufacturing and

NU

industry less the domestic economy would be able to absorb aid flows. Finally the coefficients

MA

on Inflation rate and Debt Service show that macroeconomic instability and debt burden are important issues to deal with when exploring the question of the absorption of aid flows by

AC

CE P

TE

D

recipient countries12.

27

ACCEPTED MANUSCRIPT Section 4. Conclusion:

The aid landscape is changing with emerging donors like China increasing their influence in

IP

T

development cooperation. By spurring competitive pressures in aid architecture, they are

SC R

introducing flexibility into what was formerly a traditional donor-driven space, increasing the negotiating space of recipients. Moreover, following a market based approach by combining trade

and investment to foreign aid, emerging donors are addressing bottlenecks in lowThey are utilising their recent

NU

income countries with inexpensive and tangible results.

experiences of structural transformation as a comparative advantage when defining their If recent country case studies showed that some

MA

approach of development cooperation.

countries adopted active strategies to develop partnership with emerging donors and to boost

D

infrastructure construction, as well as the development of agriculture, manufacturing, and

TE

small- and medium-sized enterprises, they do not investigated the macroeconomic impact of

CE P

such policy given the specificities of emerging donors‟ aid allocation. This article informs our understanding of how China and the new donors are influencing development cooperation relationships and fiscal policies taking place in developing countries. We estimated the direct

AC

effect of emerging donors aid on government fiscal behaviors and indirect effect on private sector behavior regarding aid absorption. Like traditional donors, some non-DAC rely on budget support to channel their aid and finance public investment while others like China provide discrete infrastructures projects reducing the investment gap and targeting bottlenecks to growth in low-income countries. Results presented in this paper suggest that emerging donors‟ development cooperation approach merits greater attention. Our empirical analysis also showed that in comparison to DAC donors, non-DAC donors like China reduce the administrative costs related to aid management and help their partners learning from aid projects to enhance their capacity to mobilise domestic resources and

28

ACCEPTED MANUSCRIPT growth. However the increasing demand of African countries and other LICs for new types of financial instruments to finance their development like Resources for Infrastructures packages,

T

equity investment, concessional and non-concessional loans and export credit provided by

IP

EximBank of China and other southern commercial banks, is also leading to significant

SC R

increase of the borrowing level of governments. Thus there is a need to be cautious and define a clear strategy to deal with the problem of debt sustainability.

NU

In the second part of our analysis, we were interested in the implications of the increased bargaining power and local ownership of the development agenda given the competitive

MA

pressure in aid architecture. So we shift the focus to how foreign aid is absorbed by the domestic economy in countries receiving also aid from emerging donors.

D

Our empirical results indicate that during the last two decades, countries receiving an

TE

increasing amount of emerging donors‟ aid had a better aid absorption rate. Moreover, we

CE P

found that African countries benefit from the competition between traditional donors and China, at least in terms of better absorption rate of aid. The presence of emerging donors changes the expectations of private agents about the fiscal management of aid flows by the

AC

government, redefining their wealth allocations towards foreign reserves accumulation and thus increasing aid absorption by the domestic economy. Developing countries could use the increasing influence of emerging donors in development cooperation as strategic policy to improve their macroeconomic management of aid flows and control over the development agenda. Findings from this study suggest that for developing countries any future partnership with traditional donors will exist alongside other cooperation framework, especially with emerging donors as they are providing them more negotiating space. The presence of emerging donors should not be perceived as a threat, they are enhancing the capacity of developing countries to become stronger partners. Furthermore, the increasing influence of emerging donors in 29

ACCEPTED MANUSCRIPT development cooperation also reveals the kind of partnership developing countries are looking for in the future development agenda; a partnership that is no longer primarily about aid. The

T

future of DAC development policy would depend of the capacity of traditional donors to

IP

increase their knowledge and understanding of developing countries priorities, thus improving

SC R

their policies responses, and also create a framework to work with emerging donors towards

NU

more integrated and effective development cooperation.

MA

Acknowledgments :

TE

D

The authors are grateful to P. Guillaumont and FERDI (Fondation Pour les Etudes et Recherches sur le Développement International) for technical and financial support. We thank two anonymous referees for helpful comments and valuable suggestions on an earlier version of this paper. A special thanks to Micours Milca and Eden Eliot for their assistance.

AC

CE P

The data and code for this paper are available on request.

30

ACCEPTED MANUSCRIPT Endnotes: 1. Due to limited capacity in line ministries, the governments of Fiji and Vanuatu expressed a stronger preference for projects that target Technical assistance and capacity-building activities.

IP

T

The Ethiopian government clearly prioritises projects that fit with national strategy with higher

SC R

concessionality in financing, lower conditionality and maximum commitment.

2. For example, much of Brazil‟s development cooperation initiatives revolve around its flag-

NU

ship programme, termed „technical cooperation‟, which emerged in the late 1960s and eventually became a core theme of the country‟s overseas aid activities. The primary goal of its

MA

technical cooperation initiative is to contribute to the development of the partner countries. Following another policy, India through ITEC (Indian Technical and Economic Cooperation

D

Program) offers training facilities to both civilians and military officials, supports project

TE

management activities, provides disaster relief and supports other development initiatives. The

CE P

country offers lines of credit (LOCS) via the Exim Bank of India to foreign governments, regional multilateral banks and other financial institutions (Quadir; 2013).

paper.

AC

3. Preliminary results about aid absorption rate are available online in an early version of the

4. Under the administration of former revolutionary leader Daniel Ortega (since 2007), traditional donor dissatisfaction with governance and human rights standards in Nicaragua compromises their bilateral cooperation relationships. After the 2006 coup, Australia, New Zealand, the European Union (EU) and other bilateral DAC donors imposed a series of diplomatic and financial sanctions on the Fijian government after it failed to hold elections.

31

ACCEPTED MANUSCRIPT 5. However, most of these new donors are in a quest for energy security, enlarged trading opportunities and new economic partnerships. Therefore, although there may not be policy

T

conditionalities, the majority of aid from emerging countries, especially China, India, and

SC R

IP

Venezuela, is combined with special trade arrangements and commercial investments.

6. Morrissey (2012) provides an excellent survey of fiscal effects studies.

NU

7. Unit root, panel co-integration and cross-section dependence tests are presented in Appendix.

MA

8. The System-GMM estimator controls for unobserved country-specific effects as well as potential endogeneity of the explanatory variables.

TE

D

9. Indeed, there is evidence that donor fragmentation has negative consequences, both for the effectiveness of aid, because of higher transactions costs (Djankov et al, 2009), and for the

CE P

domestic institutions in recipient countries (Knack and Rahman, 2007). With the increasing number of donors, administrative requirements tend to overburdening local authorities

AC

(Easterly, 2007).

Our estimations with DAC aid (% GDP) instead of Aid Fragmentation variable as covariate provide similar results. 10. 3-year average estimates are available upon request.

11. We build an indicator of the “revealed” policy by computing the difference between the observed flows and the “structural” flows that result from the non-political or structural determinants of these flows. These “structural” flows are the fitted values derived from a regression of observed flows on economic determinants. The residuals of this regression, the flows that remain unexplained by the regression, represent the impact of the policy and can

32

ACCEPTED MANUSCRIPT then be used to build an indicator of this policy. This methodology was also used by Combes and Saadi-Sedik (2006), and Combes Motel et al. (2009) to build policies indicators on

IP

T

commercial and financial openness and domestic policies on deforestation.

SC R

12. About the macroeconomic instability (proxied by the inflation rate) two challenging hypotheses could be consider: The stabilizing hypothesis expressing the fact that aid is more effective (thus more absorbed) in period of instability (Table 2.A) and a more classic

AC

CE P

TE

D

MA

economy to absorb aid flows (Table 2.B).

NU

hypothesis explaining that macroeconomic instability would reduce the capacity of domestic

33

ACCEPTED MANUSCRIPT References

Acharya A, Fuzzo de Lima AT, and M. Moore (2006) Proliferation and fragmentation:

T

transactions costs and the value of aid. Journal of Development Studies 42(1), pp. 1-21.

IP

Adam, C., S. O'Connell, E. Buffie, and C. Pattillo (2009) Monetary Policy Rules for Managing

SC R

Aid Surges in Africa. Review of Development Economics 13(3), pp. 464–490. Aiyar, S. and U. Ruthbah (2008) Where did all the aid go? An empirical analysis of absorption

NU

and spending. Working Paper 08/34, International Monetary Fund, Washington, DC. Angrist, J. D., and J. S. Pischke (2008). Mostly harmless econometrics: An empiricist's

MA

companion. Princeton university press.

Arellano, M., and S. Bond (1991) Some Tests of Specification for Panel Data: Monte Carlo

D

Evidence and an Application to Employment Equations. Review of Economic Studies 58, 277–

TE

297.

CE P

Arellano, M. and O. Bover (1995) Another look at the instrumental variable estimation of error-components models. Journal of Econometrics 68, 29–51.

AC

Arellano, C., Bulir, A., Lane, T. and L. Lipschitz (2009) The dynamic implications of foreign aid and its variability. Journal of Development Economics 88, 87–102. Berg, A., Aiyar, S., Hussain, M., Roache, S., Mirzoev, T. and A. Mahone (2007) The macroeconomics of scaling up aid: lessons from recent experience. Occasional Paper 253, International Monetary Fund, Washington, DC. Berg, A., T. Mirzoev, Portillo, R. and L.F. Zanna (2010) The Short-Run Macroeconomics of Aid Inflows: Understanding the Interaction of Fiscal and Reserve Policy. Working Paper 10/65, International Monetary Fund, Washington, DC.

34

ACCEPTED MANUSCRIPT Bräutigam D. (2011)

id „with Chinese characteristics‟: Chinese foreign aid and development

finance meet the OECD–DAC aid regime. Journal of International Development 23(5), pp.752-

T

764.

IP

Brech, V. and N. Potrafke (2014) Donor ideology and types of foreign aid, Journal of

Brun J.-F., Chambas G., and S. Gu rineau (2009)

SC R

Comparative Economics 42(1), pp. 61-75.

ide et mobilisation fiscale dans les pays en

NU

d veloppement, Rapport Th matique Jumbo 21, fD D partement de la Recherche, Paris. Buffie, E., O'Connell, S. and C. Adam (2009) Fiscal inertia, donor credibility, and the monetary

MA

management of aid surges. Journal of Development Economics 93, pp. 287-298.

D

Bun, M. and F. Windmeijer (2010) The weak instrument problem of the system GMM

TE

estimator in dynamic panel data models, Econometrics Journal 13(1), pp. 95-126.

CE P

Chen, Chuan, (2013) South-South Cooperation in Infrastructure in Sub-Saharan Africa, Working paper for ECOSOC, United Nations. Combes, J.-L. and T. Saadi-Sedik (2006) How does trade openness influence budget deficits in

AC

developing countries? Journal of Development Studies 42, pp. 1401-1416. Combes Motel P., Pirard R. and J-L. Combes (2009) A methodology to estimate impacts of domestic policies on deforestation: Compensated successful efforts for “avoided deforestation” (REDD), Ecological Economics 68, pp. 680-691. Djankov, S., Montalvo, J. G. and M. Reynal-Querol (2009) Aid with Multiple Personalities. Journal of Comparative Economics 37, pp. 217–229. Dreher, A., Nunnenkamp, P. and R. Thiele (2011) re „New‟ Donors Different? Comparing the Allocation of Bilateral Aid Between non-DAC and DAC Donor Countries. World Development 39 (11), pp. 1950-1968.

35

ACCEPTED MANUSCRIPT Driscoll, J. C. and A. C. Kraay (1998) Consistent Covariance Matrix Estimation with Spatially Dependent Panel Data. Review of Economics and Statistics 80, 549-560.

IP

T

Easterly, William (2007) Are Aid Agencies Improving? Economic Policy 22:633–678. Eberhardt, M. and F. Teal (2011) Econometrics for Grumblers: A New Look at the Literature

SC R

on Cross-Country Growth Empirics. Journal of Economic Surveys 25(1), pp.109-155. Foster, M. and T. Killick (2006) What Would Doubling Aid Do for Macroeconomic

NU

Management in Africa? Mimeo (London; Overseas Development Institute).

MA

Greenhill, R., Prizzon, A. and A. Rogerson (2013) How are developing countries managing the new aid landscape? Overseas Development Institute Report

D

Harrison, G (2004) The World Bank and Africa. The Construction of Governance States.

TE

Routledge: London and New York.

CE P

Hayakawa, K (2007) Small sample bias properties of the system GMM estimator in dynamic panel data models, Economics Letters 95(1), pp. 32-38.

AC

Heller, P (2005) Pity the finance minister: issues in managing a substantial scaling up of aid flows. Working Paper 05/180, International Monetary Fund, Washington, DC. Hudson, J. and P. Mosley (2008) Aid Volatility, Policy and Development. World Development 36, pp. 2082–2102. Hussain, M., Berg, A. and S. Aiyar (2009) The macroeconomic management of increased aid: policy lessons from recent experience. Review of Development Economics, 13(3), pp. 491–509. Keen, M. and B. Lockwood (2010) The value added tax: Its causes and consequences. Journal of Development Economics 92, pp. 138–151.

36

ACCEPTED MANUSCRIPT Kimura, H., Mori, Y. and Y. Sawada (2012) Aid Proliferation and Economic Growth: A CrossCountry Analysis. World Development 40 (1), pp. 1-10.

(forthcoming issue).

IP

Development


T

Kitano, N. and Y. Harada (2015) Estimating China‟s foreign aid. Journal of International

SC R

Knack, S. and A. Rahman (2007) Donor fragmentation and bureaucratic quality in aid recipients. Journal of Development Economics 83, pp. 176-197

regime: Closing

frica‟s policy space?

Cambridge Review of International Affairs,

MA

Forthcoming

NU

Kragelund, P (2014) Towards convergence and cooperation in the global development finance

D

Kragelund, P (2008) The Return of Non-DAC Donors to Africa: New Prospects for African

TE

Development? Development Policy Review 26 (5), pp. 555-584

CE P

Kye, W (2012) Do emerging donors allocate aid as DAC members do? The case of Korea in the millennium era. Journal of International Development 24 (8), pp. 977-988. Lin, J. Y. (2010) New Structural Economics: A Framework for Rethinking Development.

AC

Policy Research Working Paper 5197, World Bank, Washington, DC. Lin, J. Y., and Y. Wang (2015) China‟s Contribution to Development Cooperation: Ideas, Opportunities and Finances. Development Policies Working Paper 119, FERDI, France. Manning, R. (2006) Will „Emerging Donors‟ Change the Face of International Co-Operation? Development Policy Review, 24(4), pp. 371-385. Martins, P.G. (2011) Aid Absorption and Spending in Africa: A Panel Cointegration Approach. Journal of Development Studies 47(12), pp. 1925-1953.

37

ACCEPTED MANUSCRIPT Morrissey, O. (2012) Aid and Government Fiscal Behaviour: What Does the Evidence Say? UNU-wider Working paper No. 2012/01.

IP

T

Naim, M. (2007) Rogue aid. Foreign Policy 159, pp. 95-96.

International Development 17, pp. 1037-1053.

SC R

Osei, R., Morrissey, O. and T. Lloyd (2005) The Fiscal Effects of Aid in Ghana. Journal of

Otim, S. (1996) Foreign Aid and Government Fiscal Behaviour in Low-income South Asian

NU

Countries. Applied Economics 28, pp. 927-933.

MA

Ouattara, B. (2006) Foreign aid and government fiscal behaviour in developing countries: Panel data evidence. Economic Modelling 23, pp. 506– 514.

D

Pesaran, H., Shin, Y. and R. Smith (1999) Pooled mean group estimation of dynamic

TE

heterogeneous panels. Journal of the American Statistical Association 94(446), pp. 621–634.

CE P

Paulo, S. and H. Reisen (2010) Eastern Donors and Western Soft Law: Towards a DAC Donor Peer Review of China and India? Development Policy Review 28(5), pp. 535–552.

AC

Quadir, F. (2013) Rising Donors and the New Narrative of „South–South‟ Cooperation: what prospects for changing the landscape of development assistance programmes? Third World Quarterly 34(2), pp. 321-338. Rodrik, D. (1998) Why Do More Open Economies Have Bigger Governments? Journal of Political Economy 106, pp. 997–1032. Roussel, L.W. (2013) The changing donor landscape in Nicaragua: rising competition Enhances ownership and fosters Cooperation. Journal of International Development 25, pp. 802-818.

38

ACCEPTED MANUSCRIPT Sato, J., H. Shiga, Kobayashi, T. and H. Kondoh. (2011) Emerging Donors from a Recipient Perspective: An Institutional Analysis of Foreign Aid in Cambodia. World Development 39

T

(12), pp. 2091-2104.

IP

Schuknecht, L. (2000) Fiscal Policy Cycles and Public Expenditure in Developing Countries.

SC R

Public Choice 102, 113–128.

Taylor, I. (2012) India‟s rise in frica. International Affairs 88 (4), pp. 779-798

NU

Walz, J. and V. Ramachandran (2011) Brave New World: A Literature Review of Emerging

MA

Donors and the Changing Nature of Foreign Assistance. CGDEV Working Paper Whitfield, L. (Eds). (2008) The Politics of Aid. African Strategies for Dealing with Donors.

D

Oxford University Press: Oxford.

TE

Windmeijer, F. (2005) A finite sample correction for the variance of linear efficient two-step

CE P

GMM estimators. Journal of Econometrics 126, 25–51. Woods, N. (2008) Whose aid? Whose influence? China, Emerging Donors and the Silent

AC

Revolution in Development Assistance. International Affairs, 84(6), pp. 1205-1221. Zimmermann, F. and K. Smith (2011) More Actors, More Money, More Ideas for International Development Co-operation. Journal of International Development, 23(5), pp. 722–738.

39

ACCEPTED MANUSCRIPT

T

Tables of Results

Imports (%GDP) Savings ratio Debt service (%GDP) Aid Fragmentation

0.85*** [0.123] -0.004 [0.003] 0.096** [0.038] 0.05** [0.022] 0.009 [0.011] 0.602 [0.628]

0.74* [0.39] -0.002 [0.003] 0.061** [0.027]

0.21*** [0.07] -0.0025 [0.002] 0.055*** [0.019]

0.72*** [0.10] -0.0004 [0.001] 0.035 [0.024]

0.024** [0.009] 3.01*** [0.89] -0.000 [0.002]

0.009 [0.007] 1.56** [0.65] -0.0007 [0.001]

0.021 [0.018] 0.53 [1.25] 0.007 [0.005]

Inflation rate (%)

TE 1208 0.11

CE P

Observations R2 AR(2) Hansen OID Instruments N° of countries

56

0.76*** [0.08] -0.002** [0.001] 0.127** [0.048] 0.084*** [0.018] 0.015 [0.015] 2.39** [1.18]

-5.29** [1.95]

D

Intercept

0.18*** [0.04] -0.001 [0.002] 0.104*** [0.024] 0.06*** [0.024] 0.01 [0.006] 0.65* [0.36]

SC R

GDP per capita

Govt Expenditures IV-FD IV-FD-DK GMM 0.0001 0.0001 -0.0001 [0.0002] [0.0001] [0.001]

NU

Lag. dependent

Public Investment IV-FD IV-FD-DK GMM 0.0012*** 0.0006** 0.0018* [0.0002] [0.0003] [0.002]

MA

NDAC influence (ODA in %GDP)

IP

Table 1.A: Estimates of the effects of emerging donors presence on fiscal variables

1208 0.15

1289 0.21 0.39 29 56

56

1.60 [2.24]

1236 0.10

64

1236 0.10

64

1323 0.25 0.02 29 64

AC

Note: Standard errors in brackets. Lagged Dependent variable and Time dummies included but not reported. Columns (3) and (6) present estimates using the two-step SystemGMM with Windmeijer (2005) correction of standard errors. Note on estimation Method:

IV-FD-DK = Instrumented First Difference as specified in Equation [4] with time dummies included but not reported, with Driscoll Kraay (1998) correction for potential cross-sectional dependence. * p < 0.10, ** p < 0.05, *** p < 0.01.

Please refer to the Data description table in the appendix for details on data sources and definitions of variables

40

ACCEPTED MANUSCRIPT

-0.05 [0.046] -0.009** [0.004] -0.30*** [0.053] 0.477*** [0.135] -0.026 [0.019] -1.79* [1.03] 0.01* [0.006] 0.006 [0.299]

GDP per capita Imports (%GDP) Savings ratio (%GDP) Debt service (%GDP)

MA

Aid Fragmentation Inflation rate (%) Financial Openness

D

Ln (population)

CE P

TE

Intercept

Observations R2 AR(2) Hansen OID Instruments N° of countries

-0.16*** [0.052] -0.001 [0.006] -0.40*** [0.05] 0.253** [0.12] -0.05** [0.018] -1.60 [1.00] 0.013 [0.017] 0.023 [0.32]

1089 0.22

58

Lending/Borrowing IV-FD-DK GMM 0.0000 -0.015* [0.001] [0.008]

IP

IV-FD 0.0014 [0.002]

SC R

Lagged dependent

0.48*** [0.12] 0.003 [0.002] -0.394*** [0.096] 0.364** [0.173] -0.067** [0.032] -3.85* [2.19] -4.86 [0.002] -0.469 [0.906]

NU

NDAC influence (ODA %GDP)

Current Account Balance IV-FD IV-FD-DK GMM -0.001*** 0.0001 -0.008* [0.0004] [0.001] [0.004]

T

Table 1.B: Estimates of the effects of emerging donors presence on fiscal variables

-0.42*** [0.090] 0.006*** [0.001] -0.004 [0.024] 0.068** [0.028]

-0.54*** [0.10] 0.0058* [0.003] 0.002 [0.02] 0.074* [0.043]

0.39*** [0.11] 0.002*** [0.000] -0.031 [0.028] 0.123*** [0.034]

-1.40 [1.27] -0.023** [0.011] -0.18 [0.49] -2.22 [1.53]

-0.078 [0.91] -0.013 [0.012] -0.80* [0.45] -2.66 [1.30]

-0.68 [2.20] -1.68** [0.02] -0.942* [0.577] 0.736 [1.33]

3.19 [2.29]

1089 0.35

1188 0.17 0.18 29 58

58

4.92 [4.17]

742 0.21

61

742 0.32

61

821 0.26 0.52 26 62

AC

Note: Standard errors in brackets. Lagged Dependent variable and Time dummies included but not reported. Columns (3) and (6) present estimates using the two-step SystemGMM with Windmeijer (2005) correction of standard errors. Note on estimation Method:

IV-FD-DK = Instrumented First Difference as specified in Equation [4] with time dummies included but not reported, with Driscoll Kraay (1998) correction for potential cross-sectional dependence. * p < 0.10, ** p < 0.05, *** p < 0.01. Please refer to the Data description table in the appendix for details on data sources and definitions of variables

41

ACCEPTED MANUSCRIPT

Imports (%GDP) Savings ratio (%GDP) Aid Fragmentation Inflation rate (%) Ln(population) Agriculture value added (%GDP)

-0.19*** [0.04] 0.001 [0.002] 0.096*** [0.021] 0.011 [0.03] 0.40 [0.25] -0.0005 [0.011] -7.3** [0.28] -0.005 [0.06]

D

Debt Service (%GDP) Fiscal Deficit

418 0.13

43

Revealed Fiscal Effort GMM 0.001** [0.000]

0.061 [0.16] 0.0000 [0.000] 0.0003*** [0.000] 0.0002 [0.0003] -0.001 [0.002] -0.000 [0.000] 0.0003*** [0.000] -0.032 [0.03]

0.57*** [0.124]

IP

Tax on Trade IV-FD -0.00006** [0.00002]

-0.31 [0.21] 0.002 [0.001] 0.0002** [0.000] -0.0001 [0.000] 0.001 [0.003] 0.0001 [0.0001] -0.028* [0.016] 0.0001 [0.0002]

0.015 [0.020] 0.047 [0.058] 53.58 [47.07]

-4.59 [2.06]

429 0.15 0.71 0.13 29 41

-0.023 [0.024] 0.916 [2.56] -16.38* [0.84]

413 0.28

43

415 0.11

42

452 0.88 0.62 20 47

AC

Observations R2 AR(2) Hansen OID Instruments N° of countries

CE P

TE

Intercept

0.69*** [0.13] 0.000 [0.001] 0.104*** [0.025] -0.018 [0.062] -1.28 [0.95] -0.16 [0.013] 0.25 [0.56] -0.080 [0.112]

SC R

GDP per capita

-0.034 [0.22] 0.001 [0.002] 0.096*** [0.021] 0.010 [0.04] 0.43 [0.39] -0.000 [0.018] -7.56* [0.42] -0.002 [0.05]

Tax on Income IV-FD -0.00003 [0.0001]

NU

Lag. dependent

MA

NDAC influence (ODA%GDP)

Tax Revenues IV-FD-DK GMM -0.0015 0.005* [0.002] [0.002]

IV-FD -0.002 [0.002]

T

Table 1.C: Estimates of the effects of emerging donors presence on fiscal variables

Note: Standard errors in brackets. Lagged Dependent variable and Time dummies included but not reported. Columns (3) and (6) present estimates using the two-step SystemGMM with Windmeijer (2005) correction of standard errors. Note on estimation Method: IV-FD-DK = Instrumented First Difference as specified in Equation [4] with time dummies included but not reported, with Driscoll Kraay (1998) correction for potential cross-sectional dependence. * p < 0.10, ** p < 0.05, *** p < 0.01. Please refer to the Data description table in the appendix for details on data sources and definitions of variables

42

ACCEPTED MANUSCRIPT

0.001** [0.0004]

0.00003 [0.0005]

0.0159*** [0.0044]

0.065** [0.015]

0.104 [0.075]

0.254*** [0.053] 0.697*** [0.092] 0.102** [0.03]

0.0346*** [0.005]

-0.0013 [0.0023] -.0036*** [0.00046]

-0.00096 [0.00148]

-4.73* [2.29] -0.0049 [0.016]

-0.0034 [0.0057] 0.0051*** [0.0008] -10.37*** [1.63] -0.0081 [0.0123]

-.0666** [0.027] 0.0011 [0.0016]

-0.011 [0.043] -0.0021*** [0.0003]

0.095*** [0.021] -0.0184 [0.017]

0.011 [0.03] -0.061*** [0.01]

0.43 [0.39] -0.000 [0.018]

3.28*** [0.99] -0.011 [0.016]

-7.56* [0.42] -0.002 [0.05]

-7.3** [0.28] -0.005 [0.06]

TE

CE P -1.02 [2.94]

Observations R2 N° of countries

146 0.28 32

AC

Intercept

0.805 [2.68] 0.0007 [0.008]

D

Savings ratio(%GDP)

Fiscal Deficit

IP

0.0006 [0.004]

SC R

0.0014*** [0.0003] -0.121*** [0.025]

0.0601*** [0.013]

Agriculture va. (%GDP) Financial openness

7 Tax on Trade -0.0176* [0.008] -0.0015* [0.007]

-0.0615 [0.059] -0.0034 [0.006]

Imports (%GDP)

Ln (Population)

6 RFE direct 0.0039 [0.0026] 0.0004 [0.002]

0.522** [0.181] -0.026 [0.045]

0.0092*** [0.0021] 0.0493*** [0.013]

Inflation rate (%)

5 Direct Tax

-0.922* [0.491] 0.083*** [0.015]

0.0089 [0.0056]

Aid Fragmentation

4 CAB

-0.340** [0.145] 0.0078 [0.016]

GDP per capita

Debt Service (%GDP)

3 L/B ratio

NU

Aid DAC (%GDP)a

2 Gov. Exp

MA

China_projects

1 Public Inv. -0.002 [0.002] 0.045*** [0.010]

T

Table 1. D: Estimates of the effects of China presence on fiscal variables: Africa (2000-2011)

156 0.57 31

-27.04** [8.03]

8 RFE Trade 0.0038*** [0.0008] -0.0007 [0.0008]

0.00056 [0.00058] 0.0012*** [0.00013] -0.039 [0.0262] -.00056 [0.00054]

0.569** [0.182] -0.0072* [0.0034]

-6.89* [3.46] 4.34*** [1.27]

-4.79* [1.77]

1.70*** [0.26]

0.0033*** [0.0007] 2.20*** [0.045]

181 0.24 34

218 0.49 33

274 0.31 34

175 0.96 32

0.047 [0.058]

0.001* [0.0004] 0.687*** [0.015]

268 0.58 33

175 0.92 32

Note: Standard errors in brackets. Dependent variables: Public Investment (1), Government cons. Expenditures (2), Current account balance (3), Lending/Borrowing ratio (4), Direct Taxes (on income and profit) (5), Revealed Fiscal Effort on direct taxation (6), Tax on Trade (7), Revealed Fiscal Effort on Trade (8). a

To compare the effect of DAC aid to the presence of China we replaced the Aid Fragmentation index and reported only the coefficient of DAC aid. b

Lagged Dependent variable and Time dummies included but not reported. c Time dummies included but not reported. d Missing intercept term: The intercept term excluded (omitted) by the FE-DK estimator. * p < 0.10, ** p < 0.05, *** p < 0.01. Method: Fixed Effects regression with Driscoll-Kraay standard errors correction Please refer to the Data description table in the appendix for details on data sources and definitions of variables

43

IP

T

ACCEPTED MANUSCRIPT

[1980-2010]

(1)

Trade intensity with NDAC

Aid Fragmentation Debt Service (%GDP)

D

Public Investment (%GDP)

TE

Agriculture Value Added (%GDP)

CE P

AC

Observations R2 AR(2) Hansen OID Instruments N° of countries

(3)

0.0011 [0.001] -0.002 [0.058] 12.45 [27.7] -0.29 [0.37] -3.64* [2.10] -6.48 [6.83] 2.15 [1.40]

0.000 [0.0001] 0.005 [0.009] 23.4 [21.1] 0.05 [0.24] -0.86 [0.63]

-0.0005** [0.0002] 0.002 [0.003] 49.3 [40.79] -0.69** [0.21] -3.02*** [0.82]

-13.2 [12.11]

-41.9 [22.2]

350

350 0.15

MA

Inflation rate (%)

Financial Openness

(2) Annual S-GMM IV-FD-DK 0.032** 0.002** [0.014] [0.0007]

NU

NDAC Influence (ODA%GDP)

Intercept

SC R

Table 2.A: Estimates of the effects of emerging donors presence on aid absorption rate

-0.0004 [0.0003] 0.024** [0.01] -3.92 [4.36] -0.16*** [0.01] -1.34*** [0.18] 0.81 [0.54] -1.49*** [0.109]

14.6 [18.2] 821 0.34 0.74 48 55

572 0.80

56

(4) 3-Year average S-GMM FE-DK 0.02** 0.23* [0.008] [0.12]

0.70 0.54 18 58

58

Note: Standard errors in brackets. Dependent variable is the absorption rate of aid. Lagged Dependent variable and Time dummies included but not reported. Columns (1) and (3) present estimates using the two-step System-GMM with Windmeijer (2005) correction of standard errors. Column (4) = Fixed effect regression with Driscoll-Kraay standard errors correction. Note on estimation Method: IV-FD-DK = Instrumented First Difference as specified in Equation [4] with time dummies included but not reported, with Driscoll Kraay (1998) correction for potential cross-sectional dependence. * p < 0.10, ** p < 0.05, *** p < 0.01. IV-FD-DK strategy using

Yit  2

as an instrument requires at least three periods to obtain data (t; t-1; and t-2), thus we apply instead FE-DK when using 3-year average data.

Please refer to the Data description table in the appendix for details on data sources and definitions of variables

44

ACCEPTED MANUSCRIPT

(2)

Debt Service (%GDP) Public Investment (%GDP) Agriculture Value Added (%GDP) Aid Fragmentation Financial Openness

CE P

N° of countries

TE

Observations R2

0.036*** [0.006]

0.012 [0.009] -0.048 [0.051] -2.44*** [0.38] -0.906*** [0.162] -4.38 [4.18]

0.011 [0.01] -0.042 [0.049] -2.54*** [0.34] -0.72*** [0.11] -2.14 [3.35] -0.91 [1.21]

0.013 [0.01] -0.033 [0.055] -2.45*** [0.38] -0.88*** [0.15] -4.50 [4.19] -0.74 [1.05]

0.012 [0.009] -0.056 [0.05] -2.45*** [0.38] -0.84*** [0.146] -4.90 [4.43] -0.53 [1.23]

4.44*** [0.42]

4.75*** [0.62]

3.94*** [0.53]

_

_

150 0.58

150 0.58

129 0.60

129 0.72

129 0.73

32

32

31

31

31

D

Intercept

(5)

0.009 [0.008] -0.056 [0.045] -2.45*** [0.38] -0.91*** [0.16]

NU

Inflation rate (%)

(4) 0.89** [0.40]

MA

NDAC Influence (ODA %GDP)

(3)

SC R

(1) China_projects

IP

T

Table 2.B: Estimates of the effect of emerging donors‟ presence on aid absorption rate in Africa (2000-2011)

AC

Note: Standard errors in brackets. Dependent variable: Aid absorption rate a Time dummies included but not reported. b Missing intercept term: The intercept term excluded (omitted) by the FE-DK estimator. * p < 0.10, ** p < 0.05, *** p < 0.01. Note on estimation Method: Fixed Effects regression with Driscoll-Kraay standard errors correction

Please refer to the Data description table in the appendix for details on data sources and definitions of variables

45

ACCEPTED MANUSCRIPT

IP

T

Appendix

SC R

Summary Statistics of Aid and Fiscal Variables: Variable

SD

Min

Max

-6.18 -3.00 18.81 7.57 0.073 0.68 13.22 15.20 4.47

10.80 7.07 6.51 5.72 0.56 0.18 11.74 7.52 3.90

-124.56 -46.23 1.3 0.08 0 0 -2.56 1.56 1

34.84 125.44 59.9 50.62 0.1046 1 146.6 61.4 35

TE

D

MA

NU

Current Account Balance to GDP Lend/Borrowing ratio Total tax revenue ratio Public investment ratio NDAC share in total ODA Aid Fragmentation index Official Development Assistance ratio Government consumption ratio China projects

Mean

Panel Unit Root Tests

AC

Breitung t-stat Lag 1 2

Absorb -17.10*** 92.03

Inv.P -6.43*** -1.76**

-1.91*** 0.15

-1.21 -0.32

-10.01*** -1.62*

-3.09*** -2.34***

CE P

Levin, Lin & Chu t-stat Lag 1 2

Im, Pesaran and Shin W-stat Lag 1 2

ODA -16.08*** 0.47

Inflation -9.93*** 6.32

Debt S. -8.80*** -2.46***

Gov. Exp -2.45*** 1.82

-3.01*** -1.41*

-7.85*** -0.77

-3.88*** -2.79***

-2.66*** -0.52

-7.97*** -6.01***

-9.82*** -3.28***

-6.63*** -3.94***

-1.90** -0.59

** Probabilities for Fisher tests are computed using an asymptotic Chi-square distribution. All other tests assume asymptotic normality. Notes: Test results generated by Eviews. The asterisks represent significance at the 10 per cent (*), 5 per cent (**), and 1 per cent (***) confidence levels.

46

ACCEPTED MANUSCRIPT

Panel Cointegration Test Debt S. Gov. Exp Trade_NDAC

T

Absorb Inv.P ODA Inflation

Pedroni Residual Cointegration test Prob. 0.997 0.853 0.001 0.842

D

MA

NU

-2.835 1.050 -91.11 1.003

SC R

Statistic Panel v-Statistic Panel rho-Statistic Panel PP-Statistic Panel ADF-Statistic

IP

Series:

TE

List of Countries:

AC

CE P

Afghanistan, Angola, Burundi, Benin, Burkina Faso, Bangladesh, Belize, Bolivia, Bhutan, Central African Republic, Côte d'Ivoire, Cameroon , Republic of Congo, Republic Democratic of Congo, Comoros, Cape Verde, Djibouti, Egypt, Eritrea, Ethiopia, Fiji, Ghana, Guinea, Gambia, GuineaBissau, Equatorial Guinea, Guatemala, Guyana, Honduras, Haiti, Indonesia, India, Iraq, Kenya, Cambodia, Kiribati, Lao People's Democratic Rep, Liberia, Sri Lanka, Lesotho, Morocco, Moldova, Madagascar, Marshall Islands, Mali, Myanmar, Mongolia, Mozambique, Mauritania, Malawi, Niger, Nigeria, Nicaragua, Nepal, Pakistan, Philippines, Papua New Guinea, Korea Democratic Rep, Paraguay, Rwanda, Sudan, Senegal, Solomon Islands, Sierra Leone, El Salvador, Somalia, São Tomé and Principe, Swaziland, Syrian Arab Republic, Chad, Togo, Timor-Leste, Tonga, Tuvalu, Tanzania, Uganda, Vietnam, Vanuatu, Samoa, Yemen, Zambia, Zimbabwe.

47

ACCEPTED MANUSCRIPT Data description: Definition

Source

Non-Dac (NDAC)Influence

ODA received from emerging donors (%GDP)

OECD-CRS

China Projects

Number of projects implemented “Proxy of China influence”

AidData

Net ODA NAT

The Net Aid Transfers (NAT)

CAB

Current account balance (% GDP)

Fiscal Deficit

Government primary deficit (as % GDP)

Variation of reserves

Change in international reserves (% GDP)

Borrowing (LB)

General government net lending/borrowing ratio (% GDP)

WEO

Financial openness

Chinn-Ito Financial Openness Index

Chinn and Ito (2006)

Aid Fragmentation

Fragmentation of the aid landscape Herfindhal-Hirschman index (HHI) of DAC aid shares

Authors

Human capital

Human asset index (HAI)

CERDI

Imports

Imports of goods and services (% of GDP)

WDI

Trade intensity with NDAC

Average Bilateral trade with BRICS

IMF-DOTS

IP SC R

NU

MA

D

TE

WEO Authors WDI WEO

Foreign direct investment (% of GDP)

WDI

Inflation rate (CPI, percentage change)

WDI

AC

Inflation rate

CGD data http://davidroodman.com/data/

CE P

FDI

T

Variable

Government size

General government final consumption expenditure (% of GDP)

WDI

GDP per capita

Gross domestic product per capita (USD)

WDI

Public Investment

Public Investment (% GDP)

IMF-IFS

Tax revenues

Tax revenue (% GDP) Direct taxes and Trade-related taxes

WDI-CERDI AEO database

Savings ratio

Gross savings (% GDP)

WDI

Debt service

Debt service (% GDP)

WDI

M2

Money and quasi money ( % of GDP)

WDI

Agri value added

Agriculture, value added (% of GDP)

WDI

48

ACCEPTED MANUSCRIPT Highlights Some non-DAC donors rely on budget support to channel their aid and increase directly public investment.

SC R

IP

T

China provides discrete infrastructures projects reducing the investment gap and targeting bottlenecks to growth in low-income countries;

In comparison to DAC donors, China seems to reduce administrative costs related to aid management.

NU

However, the level of borrowing of African countries is increasing and tax exemptions on aid-funded goods and services reduce tax revenues.

AC

CE P

TE

D

MA

Finally, Countries receiving an increasing amount of emerging donors‟ aid have a better aid absorption rate.

49