Foreign aid and government fiscal behavior in Zambia

Table of Contents

List of Tables

List of Figures

List of Acronyms

Executive Summary

CHAPTER 1: INTRODUCTION
1.1 Background
1.1.1 Zambia Case Overview
1.2 Aid Effectiveness
1.2.1 Aid Volatility
1.2.2 Aid Heterogeneity
1.3 Problem Statement
1.4 Research Objective
1.5 Structure of the Study

CHAPTER 2: LITERATURE REVIE
2.1 Theoretical Review
2.1.1 Foreign Aid and Government Expenditure
2.1.2 Foreign Aid and Government Revenue
2.1.3 Foreign Aid and Domestic Borrowing
2.1.4 Disaggregated Aid and Fiscal Aggregates
2.2 Empirical Studies

CHAPTER 3: METHODOLOGY.
3.1 The Fiscal Response Model
3.2 Model Estimation
3.3 Data and Variables

CHAPTER 4: EMPIRICAL RESULTS.
4.1 Trend of Foreign Aid and Fiscal Aggregates
4.2 Unit Root and Cointegration Analysis
4.3 Vector Error Correction
4.3.1 Model of Net Foreign Aid
4.3.2 Model of Grants
4.3.3 Model of Net Foreign Loans
4.4 Results Summary and Study Limitations

CHAPTER 5: CONCLUSION AND RECOMMENDATIO

REFERENCES.

APPENDIC

List of Tables

Table 1: Summary Results of Selected Fiscal Response Studies

Table 2: Summary Statistics of the Data

Table 3: Results of Unit Root Tests

Table 4: VAR Lag Order Selection Criteria

Table 5: Johansen Test for Cointegration

Table 6: Vector Error Correction Estimates

Table 7: Vector Error Correction Estimates

Table 8: Vector Error Correction Estimates

Table 9: Impact of Foreign Aid on Fiscal Aggregates

List of Figures

Figure 1: Net ODA to Regions (1960-2011)

Figure 2: Revenue and Expenditure Flows in Zambia (1970-2014)

Figure 3: Trend of Domestic Borrowing in Zambia (1970-2014)

Figure 4: Foreign Aid Flows to Zambia (1970-2014)

Figure 5: Impulse Response Functions

Figure 6: Impulse Response Functions

Figure 7: Impulse Response Functions

List of Acronyms

Abbildung in dieser Leseprobe nicht enthalten

Executive Summary

There has been high momentum in the scaling up of foreign aid in recent years. The turn of the millennium saw calls by the development community to increase foreign aid to development countries so as to assist them in attaining the Millennium Development Goals. The mechanisms through which foreign aid flows are transmitted to recipient countries require that the aid resources are channeled through the government. Thus, for foreign aid to have any meaningful impact is highly dependent on how governments respond to inflows of aid.

The paper investigates the relationship between foreign aid and government fiscal behavior. An overview of the global trend of foreign aid flows over the last few decades is provided, as well as literature and research on fiscal response studies that have aimed to examine how these aid flows influence the fiscal decisions of aid recipient governments. The paper assesses the impact of foreign aid flows on fiscal aggregates, taking into focus the case of Zambia. In particular, the paper goes into detail examining how government investment, consumption, revenue and domestic borrowing are associated with both aggregated and disaggregated aid.

The paper adopts a mixed methods approach in its analysis by triangulating between qualitative and quantitative sources of information. A Vector Error Correction approach was used to estimate the relationship between foreign aid and fiscal aggregates data for Zambia over the period 1970­2014. The econometric estimation used annual data and analyzed both short-run and long-run effects. The following were the findings:

Foreign aid flows were found to be positively associated with government investment, consumption and domestic borrowing. While, government revenue was negatively associated with foreign aid. In the short-run, it was observed that grants were used to reduce the level of the country’s domestic debt stock. Whereas, net foreign loans were seen as a substitute to domestic revenues and were used to finance the budget deficit. The paper concluded by providing a number of recommendations which suggested improvement in government’s revenue mobilization efforts, effective management of the country’s domestic debt and the deliberate action to direct revenue resources towards investment expenditure. In order to achieve sustained growth and ensure the effective used of aid, donor partners were recognized as important actors in supporting the government’ s fiscal policy direction.

CHAPTER 1: INTRODUCTION

One of the most vital problems that most nations have faced in the past 50 years is economic growth, and this has been more crucial for less developed countries. Developing countri es face the challenge of raising the desired levels of finance needed to fuel investment from their own resources. As such, foreign aid has been an ideal means to supplement the shortfall needed to accelerate development efforts (Senbet and Senbeta, 2007). The post-colonial period saw many poor countries begin their paths to development. Former colonial powers assisted in the facilitation of the development process through the provision of foreign aid resources. This principle guided foreign aid flows from the 1960s and still remains one of the main objectives of foreign aid today (Bandyopadhyay and Vermann, 2013).

In the year 1970, aid was formally institutionalized when the United Nations (UN) made the decision to set a goal for its higher income members to provide 0.7% of their Gross National Product (GNP) for development aid. Since 1970, this target has been continually reaffirmed by the UN, country leaders and heads of international institutions (OECD/DAC, 2010). For many countries, foreign aid remains a principal source of revenue. For instance, during the mid- to late- 90s, foreign aid inflows were approximately equal in magnitude to taxation and constituted nearly half of all public expenditure in low-income countries (McGillivray and Morrissey, 2001).

Following from this, in the year 2000 the international community experienced an aid paradigm shift and committed to scaling-up aid and improving methods of aid delivery to developing countries to help these nations meet the Millennium Development Goals (MDGs). The development goals were a necessary initiative at the time because the late 1990s was a period characterized by aid fatigue as most development assistance efforts proved futile. The MDGs brought a new hope in the fight against poverty and if well implemented, the goals would have facilitated sustainable growth for these poor countries (UNCTAD, 2006).

Moreover, momentum for scaling-up aid to developing countries increased with the 2002 UN Conference on Financing for Development held in Monterrey. There was wide consensus amongst the donor countries in attendance and the various multilateral agencies on the need to provide a push in order for the MDGs to be achieved by 2015 (Mavrotas and Ouattara, 2007). The meeting was a step in the right direction for those advocating for the renovation of the aid agenda following the slow-down of the late 1990s. The consensus that was reached advocated for more aid as a 1 necessary measure to meet the MDGs, likewise opened the door to the exploration of new sources of development finance, improved domestic resource mobilization and improvements in aid effectiveness (Mavrotas and Ouattara, 2007).

The orthodox nature of official development assistance is that most of it is channeled through the public sector. This means that for there to be any real impact of foreign aid on economic growth and poverty reduction, it is highly dependent on how the public sector reacts to inflows of aid resources, and in this case the public sector being the recipient government (McGillivray and Morrissey, 2001). In turn, governments respond to scaled-up aid flows through fiscal policy, which simply put is the government’s way of using spending and taxation to influence the economy (Horton and El-Ganainy, 2009).

Fiscal policy in developing countries plays a key role in helping make the development process effective. The effective use of resource flows requires sound fiscal management. It is for this reason that aid recipient governments need to frame their spending with a medium- to- long-term perspective in mind. Additionally, they need to ensure alignment of budget priorities with those of donor financing, as well as strengthen critical fiscal institutions. This helps recipient countries absorb foreign aid in a sustainable way (Gupta et al., 2007).

The recent calls for scaling-up foreign aid to finance the development agenda has created an opportunity for researchers and policy makers to investigate the macroeconomic consequences that arise to due increased resource flows. There have been ongoing concerns about whether developing countries are able to effectively absorb additional foreign aid, also related are issues pertaining to the diminishing returns associated with foreign aid and possible adverse effects on macroeconomic aggregates such as the real exchange rate and inflation. Further, attention has also been directed towards the possible crowding-out effects that may be created by incremental aid and the soundness of policy response from recipient governments (Mavrotas and Ouattara, 2007).

Then again, the question of whether foreign aid has been successful in bringing out desired change is still a difficult one to conclude on. Different studies have yielded varying results. For example, studies by Burnside and Dollar (2000), Khan (1998) and Gomanee et al. (2003) have argued that foreign aid has a positive impact but only in countries with a good policy environment. Others like Bräutigam and Knack (2004) found evidence of a negative impact of foreign aid. While, Mosely et al (1987) and Boone (1996) argued that foreign aid had no demonstrable effect.

Regardless of standpoint, the aim of this paper is to empirically investigate the relationship between foreign aid and government fiscal behavior and in particular in Zambia. This will require assessing the impact that foreign aid has on government spending, government revenue collection and domestic borrowing. Zambia was chosen for this study because it possesses a unique history, both economically and politically. The country has been through some political shifts and has also been affected by a number economic shocks which have been significant in explaining how external financing in the form of foreign aid has impacted government’s fiscal response over the last few decades.

1.1 Background

The history of foreign aid flows to developing countries has evolved over time, particularly flows aimed towards the sub-Saharan region. What is evident from a review of the aid flows is that there has been a progressive rise in official development assistance (ODA) since 1960 and it is still on the rise till today. Foreign aid flows have surged from approximately $6 billion in 1960 to $46 billion in 2011 (Bandyopadhyay and Vermann, 2013). This has all been in support of the big push theory, where foreign aid was traditionally viewed as a tool for overcoming the savings gap experienced in most developing countries, thereby releasing them from the plague of the poverty trap (Abuzeid, 2009).

Traditional growth theories revealed that foreign aid would be the necessary key that would enable developing countries transform their economies. These theories showed that the impact of aid was dependent on its effects on savings, investment and government behavior (McGillivray and Morrissey, 2001). Conventionally, it was assumed that foreign aid was meant to support domestic saving, meaning foreign aid was to finance investment and not consumption expenditure. This notion was supported by a number of early studies on the fiscal effects of foreign aid (McGillivray, 2000). However, with time the assumption that aid was only meant to finance capital expenditure received wide criticisms and was challenged by a number of researchers who not only suggested that it was possible to allocate aid to consumption but that aid could actually reduce domestic savings thus creating a cycle of aid dependency (McGillivray, 2000; Senbet and Senbeta, 2007).

Figure 1 below gives the historical trend of ODA flows to the different regions of the world. We notice that a major portion of ODA flows after the mid-1970s have mostly been directed to sub­Saharan Africa. The rest of the regions have on average experienced constant flows, with the exceptions of the Middle East and South & Central Asia who have experienced increased ODA flows since the mid-2000s.

Abbildung in dieser Leseprobe nicht enthalten

Studies on the fiscal effects of foreign aid have been addressed by two perceptions in literature. Namely, fungibility studies and fiscal response studies. Fungibility studies are concerned with identifying whether foreign aid intended for particular expenditure items is actually allocated towards those areas. Whereas, fiscal response studies explicitly model how the impact of aid is mediated by government fiscal behavior (Franco-Rodriguez, 2000; McGillivray and Morrissey, 2001). More will be said about the two concepts in the next section, but for feasibility purposes this paper focuses primarily on government fiscal response in Zambia.

1.1.1 Zambia Case Overview

During the first few years after Zambia gained its independence in 1964, the country was one of the best performing economies in the world. It was relatively prosperous as it built its economy on trade in mineral resources, mainly copper which is Zambia’s main export good. The country was able to finance much of its development from the domestic resources it collected, and external financing was mainly used to support government initiated proj ects and to strengthen diplomatic ties (Wohlgemuth and Saasa, 2008).

In the 1970s, a combination of internal and external shocks hit the nation. Due to the oil crisis, prices of the commodity went up and this was coupled by a fall in copper prices. This affected the country’s revenue position and this was the beginning of Zambia’s debt crisis. The country borrowed to finance most of its expenditures and by the early 1980’s, Zambia’s economy was in serious trouble (Anderson et al., 2000; Wohlgemuth and Saasa, 2008). Soon after the crisis period, Zambia had no choice but to become a major aid recipient and foreign aid has been fluctuating in an increasing trend ever since. For most part of the 1980s, the country subscribed to a number of International Monetary Fund (IMF) and World Bank structural adjustment programmes (SAPs) in order to deter its slipping economy (Anderson et al., 2000).

During the 1990s, Zambia was regarded as one of the poorest nations in the World. It performed poorly in terms of human development compared the rest of sub-Saharan Africa, with its social indicators way below the region’s averages. Further, Zambia’s domestic and foreign debt levels reached unsustainable levels, which in turn affected the country’s macro-economic performance. These and many other factors led to the country to becoming highly dependent on foreign aid. From 1991, ODA inflows averaged around 30% of Gross Domestic Product (GDP) and by 1995 went as high as 56%. (Anderson et al., 2000).

The country’s economic performance improved considerably after the year 2000. Real annual GDP growth averaged 4.6 per cent in the five years from 2002 to 2007, reversing the negative trend in previous periods. The country reached a major milestone in 2005, when it reached its completion point under the Heavily Indebted Poor Countries (HIPC) initiative, triggering significant debt cancellation. Zambia’s foreign debt was reduced to US $4.0 billion, down from the $7.1 billion registered at end 2004. That same year Zambia also became eligible for debt relief under the G8 initiative, which proposed cancellation of all of the country’s debts to the IMF, the African Development Bank and the World Bank. After the G8 commitments were effected through the Multilateral Debt Relief Initiative (MDRI), Zambia’s total external debt was reduced to some $500 million (MoFAD, 2010; Wohlgemuth and Saasa, 2008).

However, much of the progress was thwarted as a result of the global financial crisis in 2008. The economic state of the country was weakened as a result of the impact on copper prices. Further, donor confidence was undermined by concerns about fiduciary risk, particularly following a major corruption scandal in the Ministry of Health in 2009 and concerns about the road sector. As a result, the Global Fund to Fight AIDS, Tuberculosis and Malaria (GFATM) and a number of Poverty Reduction Budget Support (PRBS) disbursements were suspended for some time in 2009. The situation likely weakened perceptions of the government as a responsible foreign aid recipient. However, the effects of this were off-set against Zambia’s relatively positive assessment on a number of governance indicators (Prizzon, 2013).

An assessment of recent happenings has shown that Zambia’s new middle-income status and improved access to international capital markets has changed the aid landscape in the country. It is now characterized with falling ODA volumes, and the ODA/GDP ratios are now well below the Low Income Country (LIC) average. This means Zambia is no longer an aid dependent country as it once was in earlier years (Prizzon, 2013). The move from aid dependence to self-reliance will heavily depend on the country’s ability to create inclusive institutions and have prudent fiscal and macroeconomic management. This has to be joined with the right set of incentives from donors in order to get the best results out of the current aid support the country is receives (Bräutigam, 2000)

1.2 Aid Effectiveness

The coordination of aid efforts by the donor community is paramount in ensuring the effective use of foreign aid. A number of ingenuities have emerged over the years aimed at improving the way foreign aid is managed. Some of the most important initiatives include the following; the Declaration of Aid Harmonization by aid donors in Rome in February 2003, the Paris Declaration on Aid Effectiveness in March 2005, the Accra Agenda for Action in September 2008 and the Busan Partnership for Effective Development Co-operation in November 2011 (Mavrotas and Outtara, 2007; OECD, 2011).

There are five principles that were agreed upon at these different fora which act as the pillars supporting the aid effectiveness debate. These are the encouragement of local ownership; the alignment of development programs around a country’s development strategy; the harmonization of practices with the aim of reducing transaction costs; the avoidance of fragmented efforts by donors and the creation of results-based frameworks (OECD, 2005; OECD, 2008; OECD, 2011).

Issues of absorptive capacity constraints additionally determine how much foreign aid can be spent in the short-term. Depending on the institutional capacity of the government, some foreign aid might need to be saved in order to prevent negative macroeconomic effects on the economy. Stable or smoothed spending is the preferred option as it enables recipient governments to increase expenditure in a sustainable manner over the medium term (Gupta et al., 2007).

Aid ineffectiveness is likely to be due to low productivity of aid-financed investments just as much as foreign aid which is diverted to unintended uses. Fungibility, which is defined as the transfer of aid resources towards items not accounted for¹, does play a role in ensuring effective use of foreign aid but a minimal one in most cases. According to McGillivray and Morrissey (2001), this does not mean that donors are not able to influence how foreign aid is used but that their influence is less than complete. There has been a push for more emphasis on aid effectiveness to be focused on government policy direction and general expenditure rather than categorical aid expenditure.

1.2.1 Aid Volatility

Aid volatility and uncertainty are issues of concern that affect the effective implementation of fiscal policy. Foreign aid flows are said to be 40 times more volatile than tax revenues, and more volatile than remittances (Gupta et al., 2007). These complications brought about by aid volatility are more pronounced when large parts of government spending are financed by foreign aid. Therefore, it is necessary to monitor the outcomes from scaling up aid in relation to this subject and explore different mitigation strategi es against it.

The impact of aid volatility should be judged on the influence it has on achieving the actual goals set out for foreign aid, together with any negative side effects that might be identified. Despite the limitations of studying aid volatility at an aggregated level, it is still vital to assess how foreign aid flows have evolved over time as this information can be useful for policy direction in recipient countri es (Hudson, 2012).

1.2.2 Aid Heterogeneity

Foreign aid support to developing countries is disbursed in different forms. Some of these different financing modalities include; concessional loans, grants, project aid, program aid, technical assistance, development food aid and humanitarian assistance. Likewise, these different aid modalities are meant to fulfil different developmental purposes.

Aid modalities are instruments that allow for the transfer of foreign aid from donors to recipient countri es. They give a description of the mode of aid delivery and the recipient government programs, projects, systems and institutions which the aid intends to finance (Leiderer, 2012). There is no universally accepted definition of aid modalities but the majority of development organizations accept that aid can be categorized by the different modes listed above.

Project aid is generally used for specific projects while program aid is generally distributed for non-project based activities. Donors have greater control over the use of project aid, whereas the opposite is true for program aid. Program aid goes through the recipient government budget thereby reducing the control donors have over the use of such funds (Mavrotas and Ouattara, 2007).

The relative efficiency of loans and grants likewise plays a role in determining aid effectiveness. Grants are transfers of resources, either in monetary terms or in kind, for which no payment is required. Concessional loans are monies lent at below market interest rates, extended by governments and official agencies (OECD, 2016). The preconceived view is that loans are a better measure for meeting development objectives due to their demand of efficient use because they are to be repaid. On the other hand, one can argue that recipient governments prefer grants because they do not have to be repaid and can then be used to substitute for domestic revenues (Gupta et al., 2003)

Studies by Mavrotas (2002), Gupta et al. (2003), Mavrotas and Ouattara (2003), and Clements et al. (2004) have revealed that different aid modalities yield varying effects on aid recipient economies. Therefore, it is vital to investigate the impact that these different aid modalities have on the fiscal aggregates.

1.3 Problem Statement

Much of aid effectiveness literature has focused on the effects of foreign aid on economic growth. In as much as research carried out on this topic is important, it is vital as well to understand the macroeconomic impact of foreign aid on a wide array of other factors which are just as critical in measuring the effectiveness of aid in a recipient economy. Similarly, it is equally important to assess if indeed foreign aid was used to finance public expenditure as this has an impact on the final outcome of improving the livelihoods of people in recipient countries (McGillivray and Morrissey, 2001).

As earlier mentioned, foreign aid is given to governments and as such assessing the fiscal effects of foreign aid requires an in-depth understanding of how the government is affected by external resource inflows. Accordingly, the effectiveness of aid will crucially depend on the fiscal response of recipient governments. Unfortunately, the link between foreign aid and government fiscal behavior is not straight forward because recipient governments have different prioriti es which they intend to pursue despite donor motives. How they allocate aid funds amongst consumption and investment is different from country to country, and some part of the aid is fungible so may be used for other purposes such as debt reduction.

Furthermore, it is significant to test what effect different aid modalities have on the fiscal behavior of the government. Aforementioned, different aid modalities have varying effects on government’s fiscal response. Hence, the study takes into account aid heterogeneity and creates estimation models that test the impact of disaggregated aid on the fiscal aggregates. The modalities of interest chosen for this study are concessional loans and grants.

Finally, the post-2015 development agenda which was formally adopted by the UN General Assembly presented the Sustainable Development Goals (SDGs). These development goals are the blueprint guiding development efforts from the year 2016- till- 2030 and are to be an enhancement from their predecessors the MDGs. They have been set ambitiously and in order to avoid failure, one way is to investigate previous performance of development efforts and create policies that address shortfalls identified in the different specific country contexts (CDP, 2015). Thus, it is through fiscal response studies that policy makers are informed on the best way to influence fiscal policy given increased aid resource inflows.

1.4 Research Objective

Literature on aid effectiveness and government fiscal behavior has been growing in number but still remains limited. Most studies investigating foreign aid and fiscal spending have focused on economic growth and fungibility, and these have provided little analysis on the impact of foreign aid on total government spending (Morrissey, 2015).

The objective of this paper is to add to the body of knowledge on fiscal response studies by determining the impact of foreign aid on government expenditure, revenue and domestic borrowing in Zambia.

More specifically, the paper will go into detail and:

1) Assess the trend of fiscal flows and ODA to Zambia from 1970 to 2014.
2) Measure the impact of net foreign aid inflows and disaggregated aid on government revenue.
3) Measure the impact of net foreign aid inflows and disaggregated aid on government expenditure. Government expenditure has been categorized into consumption and investment expenditure.
4) Measure the impact of net foreign aid inflows and disaggregated aid on domestic borrowing.

1.5 Structure of the Study

The paper is split into five sections. The first section introduces the topic, giving a brief background of the study. It introduces some important concepts, states the research problem and gives the research objective. The second section is the literature review, this section presents theoretical foundations and empirical case studies on the topic at hand. The third section is the methodology, presented here are the fiscal response model, the process of model estimation, as well as data issues that may be of concern. The fourth section gives the empirical results, this section discusses the findings of the study and highlights some of the study limitations. The last section is the conclusion and recommendations. From this, the main conclusions from the paper are presented and some policy implications and recommendations are drawn.

CHAPTER 2: LITERATURE REVIEW

The early works of Harrod (1939), Domar (1946), Chenery and Bruno (1962), and Chenery and Strout (1966) all provided the foundation for the necessity of foreign aid. According to early growth theories, developing countries struggled with domestic capital accumulation and external assistance was seen as the best option to assist these countries break from the trap of poverty. For economic development to take place, domestic savings and foreign exchange reserves had to be supplemented by additional resources from foreign aid otherwise there would be a constraint on long-term growth in these developing countries (Omoruyi, 2014).

Over the years many more studies, both theoretical and empirical, have been conducted on the topics of foreign aid, growth, aid effectiveness and fiscal response. The fiscal response method provides the foundation of the framework of this paper. This section will review some theoretical literature and empirical studies that have been conducted on the impact of foreign aid over the years, paying particular attention towards the fiscal effects of aid.

2.1 Theoretical Review

Despite the fact that most of the studies exploring the effect of foreign aid on growth have produced conflicting results, the early growth models supported the notion that foreign aid was beneficial to recipient countries as it supplemented the domestic savings and eased foreign exchange shortage. By means of relating to fiscal response studies, no such well-developed theories were available at the time which could explain the influence that foreign aid had on government fiscal behavior (Omoruyi, 2014).

The beginning of fiscal response studies can be traced back to the work of Heller (1975). He developed the fiscal response model to gain insight into the interactions among several categories of public expenditure and of domestic and foreign revenue. It was argued that foreign capital inflows resulted in increased public consumption rather than increased investment, taxes were also squandered on non-productive forms of public consumption and the consequence of this was that foreign capital inflows contributed less to growth. He sought to examine these questions using econometric techniques through the fiscal response model described in the next section.

Over the years, Heller’s specification of the fiscal response model was adopted and modified to consider current trends in development practice or to address country context specificities. For example, Mosley et al. (1987) adjusted the model to examine the influence that overseas capital had on growth when channeled through private and public investments. In their study, foreign aid had to be indirectly captured through prices in the private investment function, and growth was a function of both private and public capital stock thus the indirect effect of foreign aid was captured from both fronts.

Gang and Khan (1999) in one of their latter studies suggested that Heller’s model was impractical based on the argument that governments should not place the same weights for over and undershooting of their target variables. As such, their study proposed an alternative quadratic-ratio loss function which sought to minimize the deviations of target variables by attaching different weights, unlike the previous studies which took the deviations as linear.

Franco-Rodriguez et al. (1998) as well made some adjustments to earlier model specifications by treating aid as an endogenous variable in their quadratic utility function and included inequalities in their budget constraint, differing with previous studies. Further, Franco-Rodriguez etal. (1998) allowed for domestic borrowing to be a funding instrument for government consumption unlike previous studies which limited it to funding investment. Many other authors have made their own modifications depending on which hypotheses they planned to test in their studies, some of these including White (1994), Khan (1998), McGillivray (2000) and Mavrotas and Ouattara (2007). The trend of altering the fiscal response model through-out the years has been necessary for broadening the knowledge base of literature in the field of fiscal response studies.

2.1.1 Foreign Aid and Government Expenditure

Focusing on the expenditure side, theory does directly support the hypothesized intention of giving of foreign aid. As earlier explained, this has generally been to avail additional resources to recipient governments for funding their expenditure (Bandyopadhyay and Vermann, 2013). Most studies on this interaction have been centered on fungibility, assessing whether donors or governments spend foreign aid on its intended purpose. Nevertheless, numerous studies have found that even in situations where aid has been found to be fungible, the effectiveness of aid has not been lessened. To add on, studies based on fiscal response of governments have offered more analysis on the effect of foreign aid on total expenditure (Omoruyi, 2014).

McGillivray and Morrissey (2004) conducted a study to check whether aid in general was fungible, this test was done to assess if the effectiveness of aid is lessened due to the diversion of external finance from funding public investment to consumption expenditure. Nonetheless, McGillivray and Morrissey (2000) put forward an argument mentioning that the distinction between the two types of expenditure needn’t be necessary as government consumption comprises essential recurrent expenditures that are used to maintain and operate investment projects. From the studies reviewed, it was evident that both government investment and consumption spending were vital in ensuring that foreign aid was effective in accomplishment its target. Therefore, it is then necessary that any upsurge in government spending does not coincide with a fall in domestic revenue, whether foreign aid is fungible or not, such a counteraction does not have positive effects on the fulfilment of development assistance goals.

2.1.2 Foreign Aid and Government Revenue

As for government revenue, the few studies that have been done concerning aid and tax collection have not provided solid evidence that aid has a behavioral effect on revenue collection. Foreign aid may have a positive fiscal effect and spring up government revenue collection efforts. Though such a condition is more likely in situations where donors tie the aid to specific projects and the government has to mobilize domestic revenues to fund part of the project, or in situations where certain revenue collection conditionalities accompany the development assistance (Omoruyi, 2014). However, most of the studies in fiscal response have hypothesized that foreign aid inflows may actually lessen government’s revenue mobilization efforts (Heller, 1975; Mosley et. al., 1987). This is particularly likely to be more prone in countries with weak institutional setups. This is so because such recipient governments view the foreign assistance as a direct resource substitute to domestic revenues that can be used to fund their expenditures (Gupta et al., 2003).

In addition, some policy reforms associated with aid conditionality are difficult to address when they tend to reduce government revenue. Morrissey (2015) gives an example of economic liberalization as one such policy reform that is characterized by aid increases but is associated with revenue reductions. According to the World Bank’s (1998) argument, in a situation where foreign aid inflows reduce domestic revenue, there are misguided policies, encouraged acts of corruption, incompetence within the recipient government and ultimately the hindrance of growth. Therefore, it is important that foreign aid inflows to a recipient country are accompanied by corresponding increases in government revenue.

2.1.3 Foreign Aid and Domestic Borrowing

As posited above, the effect of foreign aid on fiscal spending is generally positive but in most instances is rarely fully additional. In some cases, this may be due to the fact that aid is fungible. But more importantly, it is because foreign aid is used by recipient governments to support reductions in domestic borrowing. For most multilateral agencies, reducing levels domestic borrowing is one of the requirements that a recipient government has to meet in order to obtain external assistance (Morrissey, 2015). As well, most developing countries have limited capacity to affect domestic revenues in the short term. Thus, domestic borrowing, which is much easier to manipulate, acts as a major determinant for government spending depending on the amount of foreign aid that it receives and the public fiscal policies it has in place (Morrissey, 2015).

2.1.4 Disaggregated Aid and Fiscal Aggregates

Foreign aid in its disaggregated form provides the opportunity to look at the dynamics of aid from a different perspective. This paper has focused on two modalities in particular, aid grants and net foreign loans. Each of these aid modalities has its own characteristics in terms of disbursement, composition and concessionality. As such, the choice of modality for aid delivery serves to fulfil different purposes both for donors and recipient governments (Mavrotas and Ouattara, 2007). The main aim behind the investigation of the impact of grants and net foreign loans was to understand if there exists a significant difference in the way each of these aid modalities affects the different fiscal aggregates.

In terms of revenue collection, there are concerns that have been highlighted in different literature about aid discouraging tax effort if it is given purely in grant form. The reason for this is that grants have no repayment obligations. On the other hand, loans have to be repaid and are then viewed to encourage tax effort in order to make it possible for recipient governments to meet loan repayments (Gupta et al., 2003). Furthermore, due to the same differences of non-repayment and repayment between grants and loans respectively, it is expected that grants are more likely to be directed towards consumption while loans are to be directed towards investment (Gupta et al., 2003). As for domestic borrowing, both grants and loans are expected to reduce the amount of domestic borrowing but given the characteristics of both modalities, it is expected that recipient governments would be more in favor of grants to serve this purpose.

In general, a number of different research studies have revealed that the impact of foreign aid as a whole and in its different distinctions, on government expenditure, revenue and in some cases domestic borrowing, differ from study to study. What has been evident is that the government’s spending has depended on three basic sources of revenue which are; foreign aid, government revenue, and domestic borrowing. Therefore, it should not be assumed that if all the foreign aid that goes through the government budget is spent then all these resources are spent on public expenditure. Spending might increase by more or less the amount of aid received depending on the other dynamics at play amongst the fiscal variables (Morrissey, 2015). Therefore, it is important to examine the outcomes from some empirical cases that have tested these hypotheses so as to assess their validity.

2.2 Empirical Studies

To begin with, we look at some of the early empirical studies on fiscal response with the first being that of Heller (1975). Heller’s paper considered the impact of different kinds of aid on public expenditure, domestic borrowing and government revenue. The study used cross-section time series data from eleven African countries, distinguishing between English-speaking and French­speaking countries. Results revealed that public consumption and investment expenditure were strongly interdependent on each other and were positively influenced by aid. Increases in tax revenue also positively influenced public expenditure and corresponded with a reduction in net borrowing. Di saggregating aid between loans and grants, the author noticed that both loans and grants reduced domestic borrowing and taxes. Heller (1975) also observed that loans were more pro-investment while grants were pro-consumption. In addition, the reduction of taxes contributed indirectly towards private consumption. The results from Heller’s study confirmed that public decision makers differed in their preferences amongst the two types of expenditure and the mode in which these expenditures were to be financed.

Gang and Khan (1991) later adopted Heller’s model and empirically investigated the fiscal behavior of the Indian government to foreign financial inflows. They employed time series data for the period 1961-1984. The study proposed a two-step procedure that firstly tested the effect of foreign aid on public investment, taxation and government consumption. Secondly, they estimated the impact of public investment and consumption on growth and income distribution. After using a non-linear three-stage least square (3SLS) method of estimation, Gang and Kang (1991) found that there results varied with earlier empirical work. Grants and loans were found not to have a 15 significant effect on government consumption. Though, results concerning tax revenues were similar with Heller’s findings in that tax revenues were used to fund consumption. Their work received some criticisms from other authors, including White (1994) who critiqued their work placing concern over theoretical and methodological aspects.

Now turning the focus towards some recent studies on fiscal response in select sub-Saharan countries, we begin with McGillivray and Ouattara (2003) who applied a fiscal response model to look at the interactions between foreign aid and government fiscal behavior in Cote d’Iviore during the period 1975-1999. Their paper differed from previous works as it recognized the significant impact that debt service expenditure had on other fiscal aggregates and foreign aid. There findings showed that a significant amount of foreign aid was used to service public debt compared to other types of government expenditure. Foreign aid induced a reduction in taxation effort, though the authors noted that this could have actually benefited the private sector by indirectly increasing private sector consumption. In addition, inflows of foreign aid did not correspond with reductions in the level of public debt entailing that aid and public debt were not substitutes for financing government expenditure contrary to conventional thinking (McGillivray and Ouattara, 2003).

According to Osei et al. (2003), they found similar results in terms of taxation and domestic borrowing. Their study analyzed the effect of aid on fiscal behavior in Ghana using annual data over the period 1966-1998 within a cointegrating Vector Autoregressive (VAR) framework. Results from their study revealed that the impact of aid on tax revenues and government spending was relatively insignificant with the two estimates almost having identical magnitudes. Furthermore, results suggested that domestic borrowing had been the main long-term financing instrument for the government and that aid was used as a short- to medium-term measure to alleviate budget constraints (Osei et al., 2003).

Moving on, Fagernäs and Schurich (2004) conducted a study that discussed movements in aid and fiscal aggregates in Malawi. The paper used several Vector Error Correction (VEC) models to estimate the fiscal effects of net ODA, aid grants and concessional loans on fiscal aggregates over a thirty-year time period. Estimation results suggested that increases in all the three types of external financing had a positive impact on government investment and a negative effect on domestic borrowing. Increases in foreign aid did not appear to discourage tax effort but in terms of government consumption, increases in grants and net ODA had a negative association with consumption while loans were positively associated with consumption.

Another case study by M’Amanja et al. (2005) employed time series econometric techniques to investigate the relationship between fiscal aggregates, foreign aid and economic growth in Kenya. Tax revenues were found not to have a significant direct influence on growth but had an indirect effect on government expenditure. The effect was dependent on the consideration of either loans or grants, of which grants appeared to have a positive effect on growth in the long-run. Conversely, loans appeared to have a negative effect on growth as they substituted taxes and accordingly were used to finance fiscal deficits. Comparing these results to the study of Gupta et al. (2003), they found that their results contradicted one another as in their study grants were associated with reduced tax effort while loans were associated with an increase in tax effort.

The final case to be reviewed was the study by Bwire et al. (2013). They assessed the dynamic relationship between foreign aid and domestic fiscal variables in Uganda over the period 1972­2008. They used a Cointegrated VAR model for their empirical analysis. The key results from the study were that foreign aid and the fiscal variables formed a long-run stationary relation. A number of hypotheses of this long-run effect of aid on government fiscal behavior were tested, and the assumption of variable endogeneity held for all the variables. The findings showed that foreign aid encouraged tax effort, reduced domestic borrowing and increased public spending. Although the increases in public spending were less than proportional to incremental aid, the existence of a budget constraint suggests that foreign aid to the government was likely to be fully additional. Improved public finance management and reduced domestic borrowing are common policy conditions attached to aid, and this was the policy direction portrayed in the study as results revealed that foreign aid was either associated with or caused beneficial policy responses in Uganda. Alternatively, it could be the case that in fiscal terms, foreign aid was utilized reasonably. From the cases above, we notice that researchers have found it difficult to generalize the effect that foreign aid inflows have on recipient governments’ fiscal behavior. Findings have varied from study to study, with each case producing context specific results. To conclude the review, Table 1 below gives a summary of other empirical studies that have investigated the association between foreign aid and government fiscal response.

[...]

---

¹ As defined by White and Dijkstra (2003: 468)