De Facto Fiscal Space in Donor-countries and Their Aid Supply: To What Extent is Trade-related Aid Supply Affected?

Abstract

This article investigates the impact of fiscal space in donor-countries on their official development aid (ODA) supply. It relies on the indicator of ‘De Facto Fiscal Space’ proposed by Aizenman and Jinjarak (The Fiscal Stimulus of 2009–10: Trade Openness, Fiscal Space and Exchange Rate Adjustment, NBER Working Paper 17427, 2011) and on a panel of 22 donor-countries over the period 1964–2015. The analysis considers four measures of ODA, including the total net aid transfers (NAT), ODA allocated to all sectors in the recipient-countries (ODAALLSECT), ODA allocated to the trade sector and ODA provided for the non-trade sector. The empirical results show that greater fiscal space in donor-countries influences positively donors’ NAT, their ODA allocated to all sectors as well as their ODA allocated to the non-trade sector in recipient-countries. At the same time, greater fiscal space in donor-countries does not influence ODA relating to the trade sector. Furthermore, the impact of fiscal space on ODA supply to the trade and non-trade sectors depends on donor-countries’ level of economic wealth.

Jel Classification: E62, F35

Keywords

De Facto Fiscal Space development aid development related to trade sector development aid related to non-trade sector

Introduction

The mobilization of financial resources is more critical than ever to achieve the sustainable development goals (SDGs) that were adopted by the Members of the United Nations in September 2015 (see UN document A/RES/70/1). International public financial resources, including the official development aid (henceforth referred to as ODA) are particularly important for developing countries, including the poorest among them. In principle, the provision of ODA by donors to developing countries, in fulfilment of their ODA commitments¹ vis-à-vis these countries, could depend on securing a higher fiscal space. The literature on the determinants of ODA has emphasized the importance of fiscal policy in donor-countries for their ODA supply. However, the empirical strand of this literature concerning whether (and if so in which direction) fiscal policy (including government revenue, government spending and public debt) in donor-countries influences the latter’s aid supply remains inconclusive. To the best of our knowledge, this literature has overlooked the implication of fiscal space for the development aid supplied by donors. In addition, while the bulk of the empirical work on the relationship between fiscal policy and donors’ aid supply has been carried out on the aggregate (i.e., total) ODA, to the best of our knowledge, none of the study has examined the impact of fiscal policy, notably fiscal space on ODA components, including for example, sectoral ODA flows.

The current analysis aims to contribute to filling this gap in the literature by investigating how donor-countries’ fiscal space influences their aid supply, including total ODA as well as sectoral ODA, in particular ODA allocated to the trade sector in recipient-countries and the ODA allocated to the non-trade sector in recipient-countries. The choice for distinguishing in the analysis among these two types of sectoral ODA rests on the fact that the members of the World Trade Organization (WTO) have recognized the importance of mobilizing greater financial resources in order to help developing countries and least developed countries (LDCs) expand their trade and better integrate into the multilateral trading system. This particularly involves the increase in donors’ ODA allocated to the trade sector in recipient-countries (the so-called ‘Aid for Trade’, i.e., ‘AfT’). Hence, AfT, which aims to reconcile the trade and development communities, remains an important part of the global ODA. Incidentally, the critical role of the trade sector as a means for the implementation of SDGs has been well highlighted in the Agenda 2030.²

To perform the analysis, we use the indicator of ‘De Facto Fiscal Space’ proposed by Aizenman and Jinjarak (2011). This indicator is defined as the ratio of the outstanding public debt to the de facto tax base. The empirical analysis is conducted using a panel dataset comprising 22 donor-countries over the period 1964–2015. Results show that greater De Facto Fiscal Space in donor-countries influences positively donors’ Net Aid Transfers (NAT), their ODA allocated to all sectors and particularly their ODA allocated to the non-trade sector. At the same time, greater fiscal space in donor-countries does not influence their supply of ODA to the trade sector. Furthermore, the impacts of fiscal space on donors’ supply of ODA relating to the trade and non-trade sectors appear to be dependent on the economic wealth of donor-countries.

The remaining part of the analysis is organized as follows. The next section provides a brief literature review on the definitions of fiscal space, while the third section reviews briefly the measures of fiscal space. The fourth section presents the ‘De Facto Fiscal Space’ indicator used to conduct the analysis. The fifth section discusses how De Facto Fiscal Space could influence donors’ aid supply. The sixth section lays out the empirical model underpinning the assessment of the impact of fiscal space on foreign aid. The seventh section presents the empirical methodology and interprets the results of the analysis and the last section concludes the article.

Definitions of the Concept of ‘Fiscal Space’

The concept of fiscal space has been defined in various ways in the literature.³ Heller (2005) defined fiscal space as the room in a government’s budget that allows it to provide resources for a desired purpose without jeopardizing the sustainability of its financial position or the stability of the economy. According to this definition, fiscal space would provide governments with the capacity to finance, in the short and longer term, their desired expenditure programmes as well as to service their debt without compromising macroeconomic stability and fiscal sustainability. Hence, the concept of fiscal space, as defined by Heller (2005) is closely related to the concept of ‘debt sustainability’ (Ostry, Ghosh, Kim, & Qureshi, 2010).

With respect to developing countries, fiscal space has been considered as the ‘growth-enhancing investment in physical and human capital that a government can finance with borrowed funds without prejudicing the long-run sustainability of its fiscal position’ (Schick, 2009). This definition of fiscal space justifies allowing cash-short governments to borrow for productive expenditures that have a strong prospect of being repaid through the additional revenues produced by an expanding economy (Schick, 2009).

Roy, Heuty, and Letouzé (2007) defined fiscal space as the financing available to government as a result of concrete policy actions for enhancing resource mobilization, and the reforms necessary to secure the enabling governance, institutional and economic environment for these policy actions to be effective for a specific set of development objectives.

Roy et al. (2006) also referred to the concept ‘fiscal space conjecture’ to explain the existence of a tension between fiduciary and developmental outcomes. In their view, the fiscal space conjecture does not deny the possibility that a harmonious solution exists in which fiscal paybacks and development paybacks are simultaneously secured.

The definition of the Development Committee (2006) links fiscal space to concerns relating to the short-term effects of an increase in public expenditure and its impact on the macroeconomic stability.

Literature Review on Measures of Fiscal Space

Fiscal space can be measured in various ways, including in terms of losing market access, achieving long-term sustainability, government revenue-based measures or synthetic indicators like the interest rate to growth differential and the ‘De Facto Fiscal Space’.

Measures of fiscal space in terms of losing market access or achieving long-term sustainability are, in practice, interrelated because long-term sustainability considerations often affect market access through risk premia (Botev et al., 2016). However, in light of the difficulties to rely on a single method to capture in a comprehensive way all factors that affect fiscal space, studies usually either focus on market access (Fournier & Fall, 2015; Ghosh, Kim, Mendoza, Ostry, & Qureshi, 2013) or on long-term sustainability (Blanchard, Chouraqui, Hagemann, & Sartor, 1990). For instance, a country with an expected marked rise in ageing and health public spending can have some fiscal space according to the market access approach, but none according to the fiscal sustainability approach.

The most common approach of fiscal sustainability to calculate the fiscal space of a country relies on fiscal reaction functions to obtain the primary balance to changes of the debt level (Bohn, 1998, 2008). Subsequent studies have used the fiscal gap, that is, the difference between the current fiscal balance and the medium-term debt-stabilizing balance to estimate the fiscal space. This methodology is frequently employed by the International Monetary Fund (IMF) for operational purposes (see Escolano, 2010) and in publications like Fiscal Monitors, or by the European Commission (European Commission [EC], 2007). Nerlich and Reuter (2013) noted that in spite of its advantage of being forward-looking and hence, incorporating fiscal plans announced by governments, this approach has several limitations. These include the fact that its macroeconomic forecasts tend to rely on ad hoc assumptions rather than on formal, testable models. Additionally, by being forward looking, it implicitly ignores a country’s track record of willingness to adjust while markets pay close attention to this track record.

Fiscal space in terms of distance to loss in market access can be thought of as the difference between the current debt level and the debt limit at which the government would lose market access. This approach has been developed by Ostry et al. (2010) and Ghosh et al. (2013) who have defined fiscal space as the difference between the debt limit and the current debt level, where the debt limit is the point beyond which either extraordinary (i.e., more than historical) efforts are necessary or where a country defaults. The main limitation of this approach is the difficulty to define, in practice, what constitutes a prudent debt level (see Adedeji, Ahokpossi, Battiati, & Farid, 2016; Ghosh et al., 2013). According to Saxegaard (2013), this difficulty lies in the fact that the prudent debt level depends on many country-specific circumstances as well as factors outside its control. To address these limitations, other approaches of determining a prudent debt level have been proposed, including a probabilistic approach (see Adedeji et al., 2016; Saxegaard, 2014) for low-income developing countries, and a theoretical model that embeds both the fiscal reaction of government to rising debt and the market reaction (see Fournier & Fall, 2015).

Other alternative approaches to calculate fiscal space include, for example, the tax gaps and sustainable tax rates, as well as potential, maximum or structural tax revenues. The tax gap indicator has been proposed by Blanchard et al. (1990) who assessed fiscal space by examining the tax gap, which is the difference between the sustainable and the actual tax-to-GDP ratio (‘tax rate’). This gap can be computed at different horizons and therefore depends on the horizon retained. The sustainable tax rate approach is the rate which, if constant, would achieve an unchanged debt-to-GDP ratio at the end of the relevant horizon, for a given forecast of spending and transfers. Brun et al. (2006) relied on structural indicators of government revenue to define fiscal space as the ratio of the current level of revenues to potential tax revenues. Park (2012) defined fiscal space as the distance between current tax revenues and the peak of the Laffer curve (i.e., the maximum tax revenues possible). Mario (2013) used a fiscal space indicator, which captures the required fiscal adjustment to reach a debt level of 60 per cent of GDP by 2030.

Synthetic indicators of fiscal space include the interest rate to growth differential and the ‘De Facto Fiscal Space’. The interest rate-growth differential indicator rests on the assumption that the debt dynamics will be favourable and lead to an increase in fiscal space when for a given primary balance, the nominal growth rate is sufficiently high to offset the impact of the nominal interest rates on the debt ratio. In addition, the market interest rate includes a risk premium, which should encompass the information that the market can use to assess default probability (see Botev et al., 2016, for more details). On the other hand, the ‘De Facto Fiscal Space’ indicator has been proposed by Aizenman and Jinjarak (2010, 2011) who consider it as the number of years of tax revenues that are necessary for a country to repay its debt, that is, the public debt divided by the ‘De Facto Tax Base’ of the concerned country.

As noted by Botev et al. (2016), these two indicators are simple to compute, send clear signals and are easy to communicate. However, in spite of these advantages, they fail to capture important factors determining fiscal space. As highlighted by Botev et al. (2016), all the above-described methods to compute fiscal space face common limitations. They all provide a stylized representation of a very complex problem, and many of them consider a closed economy. In addition, they also do not account for a number of country-specific factors, for instance, the maturity structure of the public debt.

Our Measure of Fiscal Space: The ‘De Facto Fiscal Space’

The current study relies on Heller (2005)’s definition of fiscal space and derives the measure of fiscal space from the previous literature review. Indeed, it uses a simple synthetic indicator of fiscal space to perform the analysis. Therefore, it uses the ‘De Facto Fiscal Space’ indicator, which compared to the interest rate-growth differential, appears as the simplest synthetic indicator that draws on fiscal indicators to provide an indication of a country’s fiscal space. In addition, it is more easily computable for all countries as compared to the interest rate-growth differential. To recall, the ‘De Facto Fiscal Space’ has been defined by Aizenman and Jinjarak (2011) as the ratio of the outstanding public debt to the ‘De Facto Tax Base’ or as the inverse of the average tax-years it would take to repay the public debt. This indicator provides information on the relative fiscal tightness of fiscal policy. In other words, it indicates the availability of the tax revenue to support fiscal policy. It is important to note that the ‘De Facto Tax Base’ refers to the realized tax collection and reflects both the ability and willingness of a country to fund fiscal expenditure and transfers. Formally, the ‘De Facto Fiscal Space’ is computed as the ratio of the outstanding public debt to the realized tax collection, the latter being averaged across several years to smooth for business cycle fluctuations. In the current study, we compute this indicator using the ‘De Facto Tax Base’ averaged over 5-year period. Hence, for a given country in a given year, ‘De Facto Fiscal Space’ is the ratio of the total public debt (% GDP) in year t to the average of total government revenue (% GDP) over years t-4 to t. Low public debt relative to tax base implies greater fiscal capacity to fund stimuli using the existing tax capacity. In other words, a decline in this ratio indicates a greater fiscal space. Conversely, a rise in the ‘De Facto Fiscal Space’ in a country implies that this country is experiencing difficulties to meet its debt obligations, as it would take more years for it to repay its outstanding public debt.

Discussion on the Impact of De Facto Fiscal Space on Donor-countries’ Aid Supply

To discuss the theoretical impact of the ‘De Facto Fiscal Space’ on donors’ aid supply, we draw on the literature on the impact of fiscal policy on donors’ aid effort. The empirical literature on the impact of fiscal policy on aid supply has reached mixed results. It has been argued that a weaker fiscal position, characterized, for example, by larger budget deficits and/or high levels of public debt will, ceteris paribus, lead to a reduction in the level of discretionary spending (including aid flows) because of strong pressures to reduce deficits and public debt and to preserve scarce foreign currency (e.g., Bertoli, Giovanni, & Manaresi, 2008; Faini, 2006; Fuchs et al., 2014; Gnangnon, 2013; Mosley, 1985; Round & Odedokun, 2004). Put differently, an improvement in the fiscal position in a donor-country will be associated, ceteris paribus, with higher overall spending, including spending on official development assistance (ODA). In line with this, Boschini and Olofsgård (2007) demonstrated empirically the existence of a positive relationship between aid flows and military expenditures in donor-countries. They explain this result by the fact that aid was used as a strategic instrument by donor-countries during the Cold War period. Bertoli et al. (2008) also argued that ‘given the small volume of aid relative to GDP, it is the overall level of public expenditures rather than its allocation among different expenditures chapters that influences the volume of aid’ (see also Faini, 2006). Likewise, it is also possible to argue following Bertoli et al. (2008) that weak budgetary positions or significant debt overhang may not have a detrimental impact on foreign aid provided that governments adopt an accommodating attitude towards the fiscal disequilibria over the medium term.

Against this backdrop, we can argue that a greater fiscal space in a donor-country could reflect greater capacity of this donor to meet its debt obligations, and hence to release financial resources for other priorities, including higher aid expenditures to recipient-countries. Nevertheless, it is still possible that higher De Facto Fiscal Space in donor-countries would be associated with lower aid supply or would exert no significant effect on aid supply if further to the improvement in this fiscal space, donors allocate more financial resources to other spending priorities than aid in their budgets. Nevertheless, if a donor considers development aid as an important tool to influence its economic and political relationships with recipient-countries, it would increase its aid expenditures, including if it experiences lower De Facto Fiscal Space.Overall, while we expect that greater De Facto Fiscal Space would exert a positive impact on donor-countries’ aid supply, we don’t rule out the possibility of a negative impact of De Facto Fiscal Space on donors’ aid supply.

Empirical Assessment of the Impact of De Facto Fiscal Space on Donors’ Aid Supply

To examine empirically the impact of fiscal space in donor-countries on their aid supply, we draw from the existing empirical literature on the determinants of donors’ aid supply (see, for example, the detailed literature survey provided by Fuchs et al., 2014) and postulate the following model:

$\begin{array}{l} O D A_{i t} = α_{0} + α_{1} O D A_{i t - 1} + α_{2} F A C T O S P_{i t} + α_{3} L o g {(G D P C)}_{i t} \\ + α_{4} U R_{i t} + α_{5} E X P E N D_{i t} + α_{6} O P E N_{i t} + α_{7} L o g {(P O P)}_{i t} \\ + α_{8} C O L D W A R_{i t} + μ_{i} + ε_{i t} \end{array}$ (1)

where i is the subscript associated with a country; t denotes the time period. The panel dataset used in the analysis contains 22 countries⁴ over the annual period 1964–2015. The period has been chosen on the basis of data availability.

The variable ‘ODA’ denotes the aid variable and is measured here in different ways. It could be the NAT, which, according to Dang, Knack, and Rogers (2013), approximates more closely the current budgetary outlays associated with aid. The ‘NAT’ is calculated by subtracting from gross official development assistance the repayments of principal, interest payments and the cancellation of non-ODA loans (i.e., debt relief). The exclusion of debt relief from the definition of aid is justified by the fact that debt cancellation does not give rise to an actual disbursement of funds and may even imply a double counting of aid if the debt that is cancelled was granted on a concessional basis. We use this variable in order to examine the impact of De Facto Fiscal Space in donor-countries on their total NAT to recipient-countries. The variable ‘ODA’ could also be the total sectoral ODA (denoted ‘ODAALLSECT’) provided to recipient-countries, the trade component of this total sectoral ODA (denoted ‘ODATRADE’), or the non-trade component of this total sectoral ODA (denoted ‘ODANONTRADE’).

Following the empirical literature on the determinants of donors’ aid supply, we include the 1-year lag of the ‘ODA’ variable as a right-hand side regressor. Indeed, Wildavsky (1964) pointed out that the current year’s spending in any public agency is predominantly influenced by the budget of the previous year. Mosley (1985) also stressed that this argument particularly holds for aid agencies since aid projects often run over several years, with financial flows being committed in year 1.

‘FACTOSP’ is the measure of ‘De Facto Fiscal Space’ described earlier.

‘GDPC’ stands for the real per capita income in a given donor-country, while ‘UR’ is the unemployment rate in a donor-country. ‘EXPEND’ represents the difference between overall government expenditure and the measure of the ‘ODA’ variable considered in model (1). The latter has been subtracted from the donor’s overall government expenditure because ‘ODA’ is part of overall government expenditure. This helps avoid the endogeneity of our ‘EXPEND’ variable in model (1).

The variables ‘OPEN’ and ‘POP’ are respectively the degree of openness to international trade of a given donor-country and the size of the donor-country’s population. The dummy variable ‘COLDWAR’ aims to capture donors’ behaviour in terms of aid supply before and after the Cold War era.

μ_i are countries’ fixed effects. The disturbance term ε_it is assumed to be independently and identically distributed. The description and source of variables used in model (1) are provided in Table A1. Standard descriptive statistics on these variables are reported in Table A2.

Discussion on the Expected Impact of Control Variables in Model (1)

‘FACTOSP’: Given the way the ‘De Facto Fiscal Space’ indicator is calculated, we expect a negative sign of the coefficient of this variable in order to reflect a positive impact on the ‘ODA’ variable considered.

‘GDPC’: Dudley (1979) has argued that ODA can be considered as a luxury good demanded only when more basic needs are fulfilled. According to Mosley (1985), the willingness of a country’s taxpayers to pay for ODA, and hence the size of the aid budget increases with this country’s average per capita income. Nonetheless, the empirical outcome on the impact of real per capita income in donor-countries on their aid effort is inconclusive. For example, a positive impact has been obtained by Boschini and Olofsgård (2007), Round and Odedokun (2003, 2004) and Jones (2015), whereas Mosley (1985), Faini (2006) and Lundsgaard, Breunig, and Prakash (2007) have obtained no significant impact of income per capita on donors’ aid effort. At the same time, Bertoli et al. (2008), Dreher and Fuchs (2011) and Frot (2009) have obtained a mixed positive effect. Overall, we expect a positive or a negative impact of real per capita income on donor-countries’ aid supply.

‘UR’: we expect higher unemployment rates in donor-countries to be negatively related to these countries’ aid expenditures. Indeed, Beenstock (1980) and Mosley (1985) showed that apart from the fiscal policy variables, unemployment rate is the other most important factor that helps explain donors’ aid supply. This is because there may be obvious incentives to cut aid expenditures and redirect funds towards domestic expenditures in times of employment problems in a donor-country.

‘EXPEND’: Based on the discussion laid out in Section 2.2, we expect that a rise in expenditure for other items than aid could be associated with a rise in aid expenditures.

‘OPEN’: The importance of trade openness for aid allocation has been emphasized in the empirical literature: aid can be used as an instrument by donor-countries to influence the economic policies of recipients, especially their openness to international trade (Alesina & Dollar, 2000; Heron, 2008). Hence, donor-countries that rely more on trade may see foreign aid as a useful tool to promote trade and hence increase their aid effort. From the empirical perspective, the literature remains inconclusive on the impact of trade openness on donors’ foreign aid supply. For example, Tingley (2010) suggested that trade-dependent donor-countries—with trade dependency measured by the sum of exports and imports relative to gross domestic product (GDP)—tend to be more ‘generous’ in terms of aid supply than countries that are less dependent on trade. However, he does not find empirically a significant effect of this variable on donors’ aid supply. Lundsgaarde et al. (2007) showed that the share of donors’ imports from developing countries to donors’ GDP is significantly and negatively associated with their aid effort. In other words, increased imports in a donor country from developing countries are associated with diminishing aid budgets. However, this impact is ‘loosely’ significant, that is, significant at only the 10 per cent level. Nevertheless, Fuchs et al. (2014) provided empirical support to this outcome, that is, by obtaining a negative and statistically significant (at 5% level) impact of imports from developing countries on donors’ aid effort.

‘POP’: As ODA has a public good character, it is expected that governments of larger donor-countries supply more ODA than those of small countries since the latter could free-ride on the aid effort of the former (e.g., Dudley, 1979; Mosley, 1985). At the same time, it could be expected that by exploiting economies of scale, including, for example, with regard to administrative costs of aid, larger donor-countries (i.e., those with higher size of population) can achieve specified objectives with relatively less aid effort (see Round & Odedokun, 2004). In contrast, small countries might relatively supply greater aid if the supply of the latter is only cost-effective in achieving its goals beyond a certain threshold (e.g., Fuchs et al., 2014). Finally, Bertoli et al. (2008) argued that the presence of less social cohesion on larger and more heterogenous countries compared to smaller countries could lead the former to be less inclined than the latter to supply greater aid. Most of the available empirical evidence has shown a negative impact of donor-countries’ population on their aid effort (e.g., Fuchs et al., 2014).

‘COLDWAR’: This variable aims to capture certain key miscellaneous qualitative time-related factors that influence donors’ aid supply. In particular, it aims at capturing the Cold War period compared to the more recent past. The emergence of Eastern European countries from the early 1990s has created competition for aid with the conventional developing countries and provides greater freedom to donors to reduce aid on the basis of concerns about governance issues, something to which they had to turn a blind eye during the Cold War era (see Hjertholm & White, 2000; Round & Odedokun, 2004). Boschini and Olofsgard (2007) also found that military expenditures of the Warsaw Pact (until 1990) and the Russian Federation (since 1991) had influenced positively Western aid efforts during the Cold War era. However, they obtain that these expenditures did not lead to a rise in Russian military expenditures in the 1990s. Recent studies have also shown that the ‘War on Terror’ could matter for donors’ aid budgets. For example, Fleck and Kilby (2010) found that the USA have significantly increased their aid expenditure after the terror attacks of 11 September 2001. Dreher and Fuchs (2011) observed for 22 donor-countries that their aid efforts rose during the War on Terror period but did not respond to the actual number of terror events. Overall, according this short literature review, aid effort has declined after the end of the Cold War, but increased again after the onset of the War on Terror. Figure 1 presents the scatter plot between each of the Aid variables and the variable FACTOFS. All figures suggest a negative correlation pattern between each Aid variable and the indicator FACTOFS. In other words, there is a positive correlation pattern between aid variables and De Facto Fiscal Space.

Figure 1.

Source: The author.

Empirical Methodology and Interpretation of Empirical Results

We estimate model (1) by means of the within fixed effects estimator (denoted ‘FE-DK’), where the standard errors are corrected by means of the Driscoll and Kraay (1998) technique, which takes into account the possible cross-sectional dependence along with the eventual heteroscedasticity and serial correlation in residuals. Note that this technique has been developed by Hoechle (2007) in Stata software for unbalanced panels. The results of the estimations of model (1) specifications where the dependent variable is alternatively ‘NAT’, ‘ODAALLSECT’, ‘ODATRADE’ and ‘ODANONTRADE’, are reported in Table 1.

Table 1.

Impact of ‘De Facto Fiscal Space’ on Donors’ Aid Supply

Variables	NAT	ODAALLSECT	ODATRADE	ODANONTRADE
Variables	(1)	(2)	(3)	(4)
One-year Lag of the dependent variable	0.868***	0.586***	0.584***	0.562***
	(0.0267)	(0.0655)	(0.0342)	(0.0806)
FACTOSP	−0.00527**	−0.00563*	0.00255	−0.00799**
	(0.00211)	(0.00310)	(0.00213)	(0.00307)
Log(GDPC)	0.0507**	0.151***	0.00861	0.147***
	(0.0223)	(0.0319)	(0.0179)	(0.0278)
UR	−0.00180*	−0.00142	−0.00114*	2.03e-05
	(0.000989)	(0.00100)	(0.000654)	(0.00113)
EXPEND	−9.51e-06	7.24e-05	0.000766**	−0.00119*
	(0.000622)	(0.000615)	(0.000334)	(0.000573)
OPEN	3.82e-05	−0.000366*	−1.60e-05	−0.000399**
	(0.000125)	(0.000192)	(0.000114)	(0.000168)
Log(POP)	−0.0427	−0.117**	0.0189	−0.131***
	(0.0248)	(0.0506)	(0.0257)	(0.0377)
COLDWAR	0.0144	0.0416**	0.0381***	0.00490
	(0.00942)	(0.0175)	(0.0108)	(0.0105)
Constant	0.248	0.535	−0.414	0.842
	(0.317)	(0.835)	(0.417)	(0.660)
Observations	921	848	825	825
Number of vountries	22	22	22	22
Within R²	0.8128	0.4374	0.5398	0.4418

Source: The author.

Notes: In each column, the variable ‘EXPEND’ is calculated as the difference between the total Government expenditure (% GDP) and the concerned Aid variable. Hence, ‘EXPEND’ is across columns [1] to [4] respectively ‘EXPEND1’, ‘EXPEND2’, ‘EXPEND3’ and ‘EXPEND4’. *p-value < 0.1; **p-value < 0.05; ***p-value < 0.01. Robust Standard Errors are in parenthesis.

Table 2 reports the outcome of the estimations of two specifications of model (1) where, for the dependent variables ‘ODATRADE’ and ‘ODANONTRADE’, we interact the ‘De Facto Fiscal Space’ indicator with the real per capita income in order to check whether the impact of fiscal space on donors’ aid supply depends on donors’ level of economic wealth. This is particularly relevant because not all donors could be considered as having the same level of economic wealth, even though they are all considered as developed countries (or high-income countries according to the World Bank classification of countries).

Table 2.

Does the impact of ‘DE FACTO FISCAL SPACE’ on ODA-RELATED TO TRADE and NON-TRADE SECTORS depends on Donor-countries’ Level of real Per Capita Income?

Variables	ODATRADE	ODANONTRADE
Variables	(1)	(2)
One-year lag of the dependent variable	0.582***	0.558***
	(0.0347)	(0.0790)
FACTOSP	−0.149**	0.304**
	(0.0538)	(0.124)
Log(GDPC)	−0.00625	0.179***
	(0.0190)	(0.0287)
FACTOSP*Log(GDPC)	0.0143**	−0.0293**
	(0.00509)	(0.0115)
UR	−0.000855	−0.000581
	(0.000595)	(0.00114)
EXPEND	0.000797**	−0.00124**
	(0.000338)	(0.000562)
OPEN	−5.47e-05	−0.000323*
	(0.000114)	(0.000178)
Log(POP)	0.0128	−0.119***
	(0.0268)	(0.0361)
COLDWAR	0.0386***	0.00437
	(0.0107)	(0.0109)
Constant	−0.154	0.307
	(0.454)	(0.616)

Observations	825	825
Number of countries	22	22
Within R²	0.5419	0.4449

Source: The author.

Notes: In each of the two columns, the variable ‘EXPEND’ is calculated as the difference between the total Government expenditure (% GDP) and the concerned Aid variable. Hence, ‘EXPEND’ is across columns [1] and [2] respectively ‘EXPEND3’ and ‘EXPEND4’. *p-value < 0.1; **p-value < 0.05; ***p-value < 0.01. Robust Standard Errors are in parenthesis.

Results across the four columns of Table 1 confirm the findings of previous studies that development aid follows a state dependence path, that is, aid in the previous year is positively and significantly associated with the aid of the current year. Results in column [1] particularly suggest that higher fiscal space exerts a positive and significant impact on donors’ NAT. This is because the coefficient associated with the variable ‘FACTOSP’ is negative and statistically significant at the 5 per cent level of statistical significance. In other words, as lower values of the variable ‘FACTOSP’ indicate greater fiscal space, we conclude from the results in column [1] that a decline by 1 year of the tax years that a donor would take to repay the public debt would induce a 0.0053 percentage point increase in donors’ NAT (% GDP). The other important positive drivers of donors’ NAT are lower unemployment rates and higher real per capita income. Other control variables do not appear to influence significantly donors’ NAT.

Turning to the results reported in columns [2]–[4], we note that greater fiscal space induces higher donors’ ODA allocated to all sectors in the recipient-countries (although the coefficient associated with the variable ‘FACTOSP’ is statistically significant at only the 10% level). At the same time, there is no significant impact of fiscal space on donors’ ODA related to the trade sector, but there appears to be a significant impact (at the 5% level) of greater fiscal space on ODA allocated by donors to recipient-countries’ non-trade sector. In particular, a decline by 1 year of the average tax-years that a donor would take to repay its public debt is associated with a 0.0056 percentage point increase in the total sectoral ODA (% GDP)allocated by this donor and a 0.008 percentage point increase in the ODA allocated to the non-trade sector. It needs to be noted that the positive drivers of the total sectoral ODA includes higher real per capita income, lower size of the population and lower trade openness, although for the latter, the coefficient is statistically significant at only the 10 per cent level. The total sectoral ODA appears to have increased in the aftermaths of the Cold War compared to the period before it. The same result is obtained for the impact of the Cold War on ODA allocated by donors to the trade sector of recipient-countries, although with a slightly different coefficient. Incidentally, lower unemployment rates and higher government expenditures (excluding ODA expenditures related to the trade sector) drive positively ODA related to trade sector. The other factors (apart from the fiscal space) that matter for ODA associated with the non-trade sector include higher real per capita income, lower government expenditure (excluding ODA related to non-trade sector expenditure), lower trade openness and a lower size of the population.

Let us now interpret the results displayed in Table 2. We are particularly interested in the coefficient associated with the variable ‘FACTOSP’ as well as the interaction variable ‘FACTOSP*Log(GDPC)’. It could be noted from column [1] that the coefficient relating to the variable ‘FACTOSP’ is negative and statistically significant at the 5 per cent level, whereas the interaction term of the interaction variable is positive and statistically significant at the 5 per cent level. This means that there is a threshold for donor-countries’ real per capita income above which the impact of fiscal space on ODA allocated to the trade sector changes sign. A graphical representation of the marginal impact of donor- countries’ de Facto Fiscal Space on their ODA related to the trade sector would provide a better picture on how fiscal space in donor-countries affects this type of ODA. In that respect, we provide in Figure 2, at the 95 per cent confidence intervals, the evolution of the marginal effect of ‘De Facto Fiscal Space’ on Donors’ ODATRADE for different donor-countries’ levels of economic development (proxied by real per capita income). For the interpretation of this graph, the statistically significant effects at the 95 per cent confidence intervals are those encompassing only the upper and lower bounds of the confidence interval that are either above (or below) the zero line. It could be observed in this figure that donor-countries whose real per capita income ranges between US$5,594.7 [= exponential (8.629572)] and US$20,242.2 [= exponential (9.915523)] supply higher ODATRADE to recipient-countries when they enjoy a greater fiscal space. In contrast, donors whose real per capita income is higher than US$46,825.8 [= exponential (10.75419)] tend to supply lower ODATRADE to recipient-countries when they experience greater fiscal space. In other words, these donors, which include wealthier developed countries, tend to allocate less ODA related to trade to recipients, probably at the benefit for ODA related to other sectors (as shown in Figure 3). Finally, for countries whose real per capita income is comprising between US$20,242.2 and US$46,825.8, there is no significant impact of their fiscal space on their ODA related to trade supply.

Figure 2:

Source: The author.

Figure 3:

Source: The author.

Concerning results in column [2] of Table 2, we note that the coefficient of the variable ‘FACTOSP’ is positive and statistically significant at the 5 per cent level, whereas the interaction term of the interaction variable is negative and statistically significant at the 5 per cent level. This means that the pattern of the impact of Donors’ De Facto Fiscal Space on their ODANONTRADE expenditure is likely the inverse of the pattern of results observed in column [1] of Table 2. From a graphical perspective, Figure 3 shows, at the 95 per cent confidence intervals, the evolution of the marginal effect of Donors’ De Facto Fiscal Space on their ODANONTRADE for different countries’ levels of economic development (proxied by real per capita income). This graph suggests that for donor-countries whose real per capita income is lower than US$7,824.7 [= exponential (8.965037)], greater fiscal space is associated with lower ODANONTRADE expenditures (i.e., for these donor-countries, there is a negative and significant impact of greater fiscal space on the supply of ODANONTRADE). For countries whose real per capita income is higher than US$39,594.7 [= exponential (10.58645)], there is a positive and significant impact of greater fiscal space on their supply of ODA related to non-trade sector to recipient-countries. For other countries, that is, those whose real per capita income is higher than US$7,824.67 and lower than US$39,594.7, there is no significant impact of fiscal space on their supply of ODA related to non-trade sector.

Conclusion

This article assesses the impact of donors’ fiscal space, that is, here their ‘De Facto Fiscal Space’—measured by the inverse of the average tax-years that they would take to repay the public debt—on their supply of development aid (ODA). Various indicators of aid have been used to carry out the analysis. They include donors’ total NAT, their ODA allocated to all sectors of recipient-economies as well as two major components of the latter, namely the ODA allocated to the trade sector in recipient-countries and ODA allocated to the non-trade sector in recipient-countries. The analysis covers 22 donor-countries over the period 1964–2015. The empirical results suggest evidence that greater fiscal space in donor-countries induces higher donors’ NAT and higher donors’ ODA allocated to non-trade sector in recipient-countries. The impact is also positive on ODA allocated to all sectors, though this impact is statistically significant at only the 10 per cent level. At the same time, greater fiscal space in donor-countries does not significantly influence donors’ ODA allocated to the trade sector in recipient-countries. Interestingly, the impact of fiscal space in donor-countries on their ODA allocated to the trade and non-trade sectors appears to be dependent on the level of economic wealth of donor-countries. Richer donor-countries tend to supply higher ODA relating to the non-trade sector when they experience greater fiscal space, whereas less rich donor-countries provide greater ODA relating to the trade sector when they experience greater fiscal space. Overall, fiscal space significantly matters for donors’ aid supply, including their aid to the trade and non-trade sectors in recipient-countries.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author received no financial support for the research, authorship and/or publication of this article.

Appendix

Table A2.

Standard Descriptive Statistics

Variable	Observations	Mean	Standard Deviation	Minimum	Maximum
NAT	1,006	0.365	0.237	−0.249	1.446
ODAALLSECT	922	0.301	0.185	0.004	1.413
ODATRADE	892	0.086	0.078	0.000	0.443
ODANONTRADE	892	0.221	0.148	0.009	1.294
UR	1,025	6.063	3.940	0.046	27.466
OPEN	1,119	62.481	31.572	9.201	216.243
FACTOSP	1,144	1.467	0.952	0.065	7.932
EXPEND1	1,006	41.970	10.026	8.280	70.878
EXPEND2	922	43.127	9.297	8.653	71.087
EXPEND3	892	43.455	9.262	8.666	71.545
EXPEND4	892	43.320	9.233	8.664	71.262
DEBT	1,144	53.973	34.285	1.595	249.114
GDPC	1,103	33806.580	14743.130	5594.682	91593.670
POP	1,144	3.63e + 07	5.68e + 07	188983	3.21e + 08

Source: The author.

Notes

References

Adedeji

Ahokpossi

Battiati

& Farid

(2016). A probabilistic approach to fiscal space and prudent debt level: Application to low-income developing countries (IMF Working Paper, WP/16/163). Washington, DC: International Monetary Fund.

Aizenman

& Jinjarak

(2010). De facto fiscal space and fiscal stimulus: Definition and assessment (NBER Working Paper No. 16539). Cambridge, MA: The National Bureau of Economic Research.

Aizenman

& Jinjarak

(2011). The fiscal stimulus of 2009–10: Trade openness, fiscal space and exchange rate adjustment (NBER Working Paper No. 17427). Cambridge, MA: The National Bureau of Economic Research.

Alesina

& Dollar

(2000). Who gives foreign aid to whom and why? Journal of Economic Growth, 5(1), 33–63.

Beenstock

(1980). Political econometry of official development assistance. World Development, 8(2), 137–144.

Bertoli

Giovanni

A. C.

& Manaresi

(2008). Aid effort and its determinants: A comparison of the Italian performance with other OECD donors (Working Papers Series wp2008-11). Universita’ degli Studi di Firenze Dipartimento di Scienze Economiche, Firenze, Italy.

Blanchard

Chouraqui

J.-C.

Hagemann

& Sartor

(1990). The sustainability of fiscal policy: New answers to an old questions. OECD Economic Studies, 15, 7–36.

Bohn

(1998). The behavior of US public debt and deficits. Quarterly Journal of Economics, 113(3), 949–963.

Bohn

(2008). The sustainability of fiscal policy in the United States. In Neck

& Sturm

J.-E.

(Eds.), Sustainability of public debt (pp. 15–49). Cambridge, MA: MIT Press.

10.

Boschini

& Olofsgård

(2007). Foreign aid: An instrument for fighting communism? Journal of Development Studies, 43(4), 622–648.

11.

Botev

Fournier

& Mourougane

(2016). A re-assessment of fiscal space in OECD countries (OECD Economics Department Working Papers No. 1352). Paris: OECD Publishing. Retrieved from https://dx-doi-org.web.bisu.edu.cn/10.1787/fec60e1b-en

12.

Brun

J.-F.

Chambas

Combes

J.-L.

Dulbecco

Gastambide

Guerineau

& Rota Graziosi

(2006). Fiscal space in developing countries (Concept Paper). New York, NY: UNDP, Bureau for Development Policy, Poverty Group.

13.

Cheng

H. W. J.

& Pitterle

(2018). Towards a more comprehensive assessment of fiscal space (DESA Working Paper No.153, ST/ESA/2018/DWP/153). New York, NY: United Nations Department of Economic & Social Affairs.

14.

Dang

H.-A.

Knack

& Rogers

(2013). International aid and financial crises in donor countries. European Journal of Political Economy, 32, 232–250. Retrieved from https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/pii/S0176268013000657

15.

Development Committee (2006). Fiscal policy for growth and development—An interim report. Paper presented at the Development Committee Meeting, Washington, DC.

16.

Dreher

& Fuchs

(2011). Does terror increase aid? Public Choice, 149(3–4), 337–363.

17.

Driscoll

J. C.

& Kraay

A. C.

(1998). Consistent covariance matrix estimation with spatially dependent panel data. Review of Economics and Statistics, 80(4), 549–560.

18.

Dudley

(1979). Foreign aid and the theory of alliances. Review of Economics and Statistics, 61(4), 564–571.

19.

Escolano

(2010). A practical guide to public debt dynamics, fiscal sustain- ability, and cyclical adjustment of budgetary aggregates (Technical Notes and Manuals No. 2010/2). Washington, DC: International Monetary Fund.

20.

European Commission (EC) (2007). Public finances in the EMU 2007 (European Economy, No. 3). Brussels: European Commission.

21.

Faini

(2006). Foreign aid and fiscal policy (CEPR Discussion Paper No. 5721). London, UK: CEPR.

22.

Fleck

R. K.

& Kilby

(2010). Changing aid regimes? U. S. foreign aid from the cold war to the war on terror. Journal of Development Economics, 91(2), 185–197.

23.

Fournier

J. M.

& Fall

(2015). Limits to government debt sustainability (OECD Economics Department Working Papers No. 1229). Paris: OECD Publishing. Retrieved from https://dx-doi-org.web.bisu.edu.cn/10.1787/5jrxv0fctk7j-en

24.

Frot

(2009). Aid and the financial crisis: Shall we expect development aid to fall? Stockholm Institute of Transition Economics, Stockholm School of Economics. Retrieved from http://www.papers.ssrn.com/sol3/papers.cfm?abstract_id=1402788

25.

Fuchs

Dreher

& Nunnenhamp

(2014). Determinants of donor generosity: A survey of the aid budget literature. World Development, 56, 172–199.

26.

Ghosh

A. R.

Kim

J. I.

Mendoza

E. G.

Ostry

J. D.

& Qureshi

M. S.

(2013). Fiscal fatigue, fiscal space and debt sustainability in advanced economies. Economic Journal, 123(566), 4–30. Retrieved from https://onlinelibrary.wiley.com/doi/pdf/10.1111/ecoj.12010

27.

Gnangnon

S. K.

(2013). The consequences of fiscal episodes in OECD DAC countries for aid supply. The Quarterly Review of Economics and Finance, 53(3), 302–313.

28.

Heller

(2005). Understanding fiscal space (IMF Policy Discussion Paper No. PDP/05/4). Washington, DC: International Monetary Fund.

29.

Heron

(2008). Aid for trade: Towards a new development assistance paradigm for small states? (IPEG Papers in Global Political Economy No.37). Newcastle, UK: IPEG.

30.

Hjertholm

& White

(2000). Foreign aid in historical perspective: Background and trends. In Tarp

(Ed.), Foreign aid and development: Lessons learnt and directions for the future (pp. 80–102). London, UK and New York, NY: Routledge.

31.

Hoechle

(2007). Robust standard errors for panel regressions with cross-sectional dependence. The Stata Journal, 7(3), 281–312.

32.

Jones

(2015). Aid supplies over time: Addressing heterogeneity, trends, and dynamics. World Development, 69, 31–43.

33.

Lundsgaarde

Breunig

& Prakash

(2007). Trade versus aid: Donor generosity in an era of globalization. Policy Sciences, 40(2), 157–179.

34.

Mario

(2013). Budgeting for fiscal space and government performance beyond the great recession. OECD Journal on Budgeting, 13(2), 9–47.

35.

Mosley

(1985). The political economy of foreign aid: A model of the market for a public good. Economic Development and Cultural Change, 33(2), 373–393.

36.

Nerlich

& Reuter

(2013). The design of national fiscal frameworks and their budgetary impact (ECB Working Paper Series No.1588, pp. 1–30). Frankfurt am Main: European Central Bank.

37.

Ostry

Ghosh

A. R.

Kim

J. I.

& Qureshi

(2010). Fiscal space (IMF Staff Position Note SPN/10/11). Washington, DC: International Monetary Fund.

38.

Ostry

Atish

Ghosh

& Espinoza

(2015). When should public debt be reduced (IMF Staff Discussion Note No. 15/10). Washington, DC: International Monetary Fund.

39.

Park

(2012). Quantifying impact of aging population on fiscal space (IMF Working Paper No. 12/164). Washington, DC: Fiscal Affairs Department, IMF.

40.

Round

J. I.

& Odedokun

(2003). Aid effort and its determinants (WIDER Discussion Paper 2003/03). Helsinki: United Nations University, World Institute for Development Economics Research.

41.

Round

J. I.

& Odedokun

(2004). Aid effort and its determinants. International Review of Economics and Finance, 13(3), 293–309.

42.

Roy

Heuty

& Letouzé

(2006). Fiscal space for public investment: Towards a human development approach. Paper prepared for the G24-Meeting in Singapore, September 2006, UNDP, New York.

43.

Roy

Heuty

& Letouzé

(2007). Fiscal space for what? Analytical issues from a human development perspective. Paper prepared for the G20-Workshop on Fiscal Policy, June–July 2007, UNDP, New York.

44.

Saxegaard

(2014). Safe debt and uncertainty in emerging markets: An application to South Africa (IMF Working Paper No. 14/231). Washington, DC: International Monetary Fund.

45.

Schick

(2009). Budgeting for fiscal space. OECD Journal on Budgeting, 9(2). Retrieved from http://www.oecd.org/gov/budgetingandpublicexpenditures/45361746.pdf

46.

Tingley

(2010). Donors and domestic politics: Political influences on foreign aid effort. The Quarterly Review of Economics and Finance, 50(1), 40–49.

47.

Wildavsky

(1964). The politics of the budgetary process. Boston, MA: Little Brown.