I. Introduction

The COVID-19 pandemic has shown how developing countries are vulnerable to shocks, and even more so than developed countries. The vulnerability of developing countries at the macroeconomic level is not a new issue in the economic development literature (e.g., Barrot et al., 2018; Briguglio et al., 2009; Dabla-Norris and Gündüz, 2014; Essers, 2013; Guillaumont, 2009; 2017; Harjoto et al., 2020; Keefe, 2021; Montalbano, 2011; Lee, 2018).

Recognizing the greater extent of macroeconomic vulnerability experienced by least developed countries1 (LDCs) among developing countries, the United Nations Committee for Development Policy (UN-CDP) has developed the concept of “structural economic vulnerability.” Structural economic vulnerability is the structural component of a country’s overall level of economic vulnerability, the latter being “the risk of a (poor) country seeing its development hampered by environmental or natural shocks as well as external shocks” (Guillaumont, 2009). Therefore, structural economic vulnerability indicates a country’s extent of exposure to exogenous shocks as well as the size and frequency of these shocks. It is important to note that the conjunctural component of the overall economic vulnerability is referred to as “economic resilience,” reflecting the country’s capacity to react to shocks, as measured through the policies that it implements (Guillaumont, 2009).

In addition to being exposed to a high degree of structural economic vulnerability, developing countries, and in particular LDCs among them, suffer from low or insufficient levels of productive capacities that could enable them to reduce their exposure to negative shocks and mitigate the adverse effects of such shocks on their economies. According for example to the United Nations Conference on Trade and Development (UNCTAD, 2006, p.61), “productive capacities” refers to “the productive resources, entrepreneurial capabilities and production linkages which together determine the capacity of a country to produce goods and services, and enable it to grow and develop.” To help researchers undertake policy analyses and make appropriate policy recommendations concerning countries’ performance outcomes with regard to their productive capacities, UNCTAD launched in February of 2021 a comprehensive index of productive capacities (UNCTAD, 2020). This indicator helps to fill a void in the literature given that an indicator of productive capacities that could help with comparative analyses across countries did not exist. Many analyses have emphasized the importance of productive capacities for promoting economic growth and development as well as enhancing economic resilience in developing countries (e.g., Cornia and Scognamillo, 2016; Gnangnon, 2022; Shiferaw, 2017; UN, 2017; UNIDO, 2001).

While development aid inflows, remittances inflows and foreign direct investment inflows could help build productive capacities in developing countries, these countries should rely first and foremost on their own financial resources as a sustainable means of financing their development needs, including the strengthening of their productive capacities. Thus, securing greater fiscal space is an ultimate objective for governments in developing countries. At the same time, the volatility of fiscal policy, likely reflecting greater fiscal space volatility, is a source of major concern to policymakers, as it can significantly undermine economic growth (e.g., Afonso and Jalles, 2012; Fatás and Mihov, 2013; Fernández-Villaverde et al., 2015). For example, public spending volatility, which could lead to greater fiscal space volatility, heightens output volatility and hampers economic growth (e.g., Afonso and Furceri, 2010; Afonso and Jalles, 2012; Fatás and Mihov, 2013; Fernández-Villaverde et al., 2015). Likewise, with the exacerbation of the instability of both public investment and government consumption, the instability of tax revenue becomes detrimental to economic growth (e.g., Bleaney et al., 1995; Ebeke and Ehrhart, 2012).

How do productive capacities and structural economic vulnerability affect fiscal space volatility in developing countries? How does the strengthening of productive capacities affect fiscal space volatility in developing countries that face a higher degree of structural economic vulnerability? The present paper aims to address these two issues.

There are several definitions of the concept of fiscal space in the literature2 (e.g., Botev et al., 2016; Gnangnon, 2019a; 2019b; Nerlich and Reuter, 2016; Roy et al., 2007; Schick, 2009). For example, Heller (2005) considers 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. More practically, Aizenman and Jinjarak (2010; 2011) define fiscal space as the ratio of public debt to public revenue. In other words, the authors propose the indicator of “de facto fiscal space3” as the ratio of outstanding public debt to the ‘de facto tax base’. De facto fiscal space is the number of years of tax revenues needed for a country to repay its debt. In the present analysis, we define fiscal space in order to facilitate the interpretation of empirical outcomes: fiscal space is the ratio of the current total public revenue to the outstanding public debt. It reflects for a given country the ability and willingness of the country to fund fiscal expenditures and transfers using the current public revenues.

To investigate the effects of productive capacities and structural economic vulnerability on fiscal space volatility in developing countries and to examine how both factors interact in influencing fiscal space volatility in these countries, we rely on the indicator of fiscal space defined above, the indicator of productive capacities proposed by the UNCTAD (2020), and on the indicator of structural economic vulnerability as defined by the United Nations.

While a number of studies have explored the determinants of fiscal space (e.g., Botta et al., 2023; Gnangnon, 2018; 2019b; Gnangnon and Brun, 2020; Nerlich and Reuter, 2016), studies of the determinants of fiscal space volatility are scarce. This may be due to the lack of consensus among economists on how to measure “fiscal space.” To the best of our knowledge, one of the few studies of the factors underpinning fiscal space volatility is that by Gnangnon (2020b), who used the indicator of fiscal space defined above to examine the effect of export product diversification on fiscal policy volatility through the avenue of economic growth volatility. He found that export product concentration enhances fiscal space volatility in countries that face greater economic growth volatility.

From a theoretical perspective, we argue, on the one hand, that by reducing economic growth volatility (e.g., Gnangnon, 2021) and enhancing economic resilience (e.g., Cornia and Scognamillo, 2016; Gnangnon, 2022; Shiferaw, 2017), the strengthening of productive capacities would help dampen the effects of shocks on economies and hence reduce fiscal space volatility. On the other hand, it can be intuitive to consider that an increase in structural economic vulnerability, which reflects an increase in the level of exposure to shocks and/or a higher extent of shocks, is likely to result in greater volatility of fiscal space in developing countries. In addition, we expect that the development of productive capacities would dampen the heightening effect of structural economic vulnerability on fiscal space volatility on developing countries. Specifically, productive capacities would exert a greater negative effect on fiscal space volatility in countries that experience a higher degree of structural economic vulnerability.

The empirical analysis has confirmed these hypotheses. It used the feasible generalized least squares (FGLS) estimator and relied on a panel dataset of 116 developing countries over the period of 2000 to 2018.

The remainder of the paper is organized around five sections. Section II presents a theoretical discussion of the effects of productive capacities and structural economic vulnerability on fiscal space volatility. Section III lays down the empirical strategy. Section IV interprets the empirical results, and Section V concludes the paper.

II. Theoretical discussion of the effects of productive capacities and structural economic vulnerability on fiscal space volatility

On the one hand, we argue that by reducing economic growth volatility (e.g., Gnangnon, 2021) and enhancing economic resilience (e.g., Cornia and Scognamillo, 2016; Gnangnon, 2022; Shiferaw, 2017), the strengthening of productive capacities would help dampen the effects of shocks on economies and hence reduce fiscal space volatility. Thus, we formulate the following hypothesis.

On the other hand, it is intuitive to expect that an increase in structural economic vulnerability, which reflects an increase in the level of exposure to shocks and/or a higher extent of shocks, is likely to result in greater volatility of fiscal space in developing countries. In fact, the indicator of structural economic vulnerability has two main components, which are the exposure sub-index and the shocks sub-index (see for example Feindouno and Goujon, 2016). The former has five component indexes, while the latter encompasses three component indexes.

The five component indexes of the exposure sub-index (with their weights in brackets) are as follows: population size (25%); remoteness from world markets (25%); export product concentration (12.5%); share of agriculture, forestry, and fishery in GDP (12.5%); and the share of population living in low elevated coastal zones (25%). Thus, it is likely that a higher degree of exposure to shocks will enhance fiscal space volatility. For example, Gnangnon (2020b) found that an increase in the level of export product concentration results in higher fiscal space volatility. Similarly, countries whose production structure reflects a high share of agriculture, forestry, and fishery in their GDP are exposed to shocks, especially environmental and external economic and financial shocks. Such shocks would adversely affect these economies and heighten the fiscal space volatility.

The three component indexes of the shocks sub-index (with their weights in brackets) are as follows: victims of natural disasters (25%), instability in agricultural production (25%), and instability in exports of goods and services (50%). It can be straightforward to expect that an increase in the extent of shocks faced by a country will heighten their fiscal space volatility. In other words, countries that face higher magnitudes of shocks will experience greater fiscal space volatility than countries that experience lower magnitudes of shocks.

On another note, Gnangnon (2021) showed that higher structural economic vulnerability is associated with greater economic growth volatility in developing countries. Therefore, we formulate the following hypothesis.

In light of hypotheses 1 and 2, we can postulate that through its positive economic resilience effect, the strengthening of productive capacities is likely to dampen the positive effect of structural economic vulnerability on fiscal space volatility. In light of the potential positive effect of economic growth volatility on the volatility of fiscal space, this theoretical expectation is further exemplified by the findings of Gnangnon (2021), who showed that the development of productive capacities contributes to dampening economic growth volatility in countries that face a higher level of structural economic vulnerability.

Therefore, we can postulate hypothesis 3, as follows.

The next sections will test empirically each of these hypotheses.

III. Empirical Strategy

This section includes three sub-sections. First, we present the baseline model specification that helps address the questions at the heart of the analysis (sub-section III.A). Second, we briefly present some data analysis, notably concerning the key variables of interest in the analysis, specifically fiscal space volatility, productive capacities, and structural economic vulnerability (sub-section III.B). Third, we present the estimator used to carry out the empirical analysis and explain the different variants of the baseline model that will be estimated using this estimator (see sub-section III.C).

B. Data Analysis

Before turning to the estimation method employed to conduct the empirical analysis, we find it useful to provide some insights into the developments of our key variables of interest (i.e., fiscal space volatility, productive capacities, and structural economic vulnerability) over the full sample. Figure 1 shows how these variables have evolved over time over the full sample. We observe that fiscal space volatility rose from 2001 to 2006, reached its peak in 2006, and then declined up to 2012. It then rebound from 2012 to 2016 and subsequently declined from 2016 to 2018. In the meantime, Figure 1 shows a declining trend of structural economic vulnerability over time and an increasing trend of productive capacities, on average, over the full sample.

FIGURE 1.

PRODUCTIVE CAPACITY, STRUCTURAL ECONOMIC VULNERABILITY AND FISCAL SPACE VOLATILITY OVER THE FULL SAMPLE

Source: Author.

Figure 2 presents the correlation pattern (in the form of a scatter plot) between productive capacities and fiscal space volatility on the one hand and between structural economic vulnerability and fiscal space volatility on the other hand, over the full sample. It shows that the indicator of productive capacities is negatively correlated with fiscal space volatility while the indicator of structural economic vulnerability is positively correlated with fiscal space volatility. In addition, we note the presence of outliers concerning instances where the values of the indicator of fiscal space volatility exceed 0.5. We take into account these outliers in the regression by introducing the dummy outlier “ DUMOUT ,” which takes a value of 1 for these outliers, and 0 otherwise. The same patterns are observed for the sub-samples of LDCs and non-LDCs (see Figure 3). The empirical analysis will provide further guidance as to whether these correlation patterns indeed reflect causality.

FIGURE 2.

CROSS PLOT BETWEEN PRODUCTIVE CAPACITY, STRUCTURAL ECONOMIC VULNERABILITY AND FISCAL SPACE VOLATILITY OVER THE FULL SAMPLE

Source: Author.

FIGURE 3.

CROSS PLOT BETWEEN PRODUCTIVE CAPACITY, STRUCTURAL ECONOMIC VULNERABILITY AND FISCAL SPACE VOLATILITY OVER THE SUB-SAMPLES OF LDCS AND NON-LDCS

Source: Author.

C. Econometric Approach

To test hypotheses 1 to 3, we estimate the baseline model (1) by means of the feasible generalized least squares (FGLS) estimator. This estimator generates more efficient estimates than those obtained from the ordinary least squares estimator, notably in the presence of heteroskedasticity, as well as serial and cross-sectional correlations (e.g., Bai et al., 2021; Zellner, 1962).

First, we test hypotheses 1 and 2 (notably the effects of productive capacities and structural economic vulnerability on fiscal space volatility) by estimating the baseline model (1) over the full sample and the sub-samples of LDCs and non-LDCs. The outcomes of these estimations are presented in Table 1.

TABLE 1

EFFECTS OF PRODUCTIVE CAPACITY AND STRUCTURAL ECONOMIC VULNERABILITY ON FISCAL SPACE VOLATILITY (ESTIMATOR: FGLS (WITH PANEL-SPECIFIC FIRST-ORDER AUTOCORRELATION))

Note: 1) *p-value<0.1, **p-value<0.05, ***p-value<0.01; 2) Robust standard errors are in parenthesis, as they are clustered at the country level; 3) The Pseudo R² is calculated as the correlation coefficient between the dependent variable and its predicted values; 4) Time dummies are included in the FGLS-based regressions.

Next, we examine how the effects of productive capacities and structural economic vulnerability vary across countries in the full sample. To that effect, we estimate in the first instance a variant of model (1) in which we introduce the interaction between the indicator of productive capacities and the real per capita income. The outcomes of this regression are presented in column [1] of Table 2. We then estimate another variant of model (1) that includes the interaction variable between the indicator of structural economic vulnerability and the real per capita income. The estimates arising from this regression are reported in column [2] of Table 2.

TABLE 2

EFFECT OF PRODUCTIVE CAPACITY AND STRUCTURAL ECONOMIC VULNERABILITY ON FISCAL SPACE VOLATILITY FOR VARYING LEVELS OF REAL PER CAPITA INCOME OVER THE FULL SAMPLE (ESTIMATOR: FGLS (WITH PANEL-SPECIFIC FIRST-ORDER AUTOCORRELATION))

Note: 1) *p-value<0.1, **p-value<0.05, ***p-value<0.01; 2) Robust standard errors are in parenthesis; 3) The Pseudo R² is calculated as the correlation coefficient between the dependent variable and its predicted values; 4) Time dummies are included in the FGLS-based regressions.

We test hypothesis 3 by estimating another variant of model (1) in which we interact the variables “ PCI ” and “ EVI .” The outcomes of this estimation are presented in Table 3.

TABLE 3

INTERACTION EFFECT OF PRODUCTIVE CAPACITY AND STRUCTURAL ECONOMIC VULNERABILITY ON FISCAL SPACE VOLATILITY OVER THE FULL SAMPLE (ESTIMATOR: FGLS (WITH PANEL-SPECIFIC FIRST-ORDER AUTOCORRELATION))

Note: 1) *p-value<0.1, **p-value<0.05, ***p-value<0.01; 2) Robust standard errors are in parenthesis; 3) The Pseudo R² is calculated as the correlation coefficient between the dependent variable and its predicted values; 4) Time dummies are included in the FGLS-based regressions.

Finally, we carry out a robustness check of the outcomes reported in column [1] of Tables 1 and 3 by measuring fiscal space volatility using four-year rolling windows and five-year rolling windows. In other words, the first other measure of fiscal space volatility is denoted as “ FSVOL4 ” and is computed as the ratio of the standard deviation of the indicator of fiscal space (over four-year rolling windows, that is, from t - 3 to t ) to the mean of the indicator of fiscal space over four-year rolling windows. The second alternative measure of fiscal space volatility (denoted as “ FSVOL5 ”) is computed as the ratio of the standard deviation of the indicator of fiscal space (over five-year rolling windows, that is, from t - 5 to t ) to the mean of the indicator of fiscal space over five-year rolling windows. Higher values of these two indicators of fiscal space volatility reflect greater volatility of fiscal space.

The results in columns [1] and [3] of Table 4 are obtained by estimating the baseline model (1), where the dependent variables are respectively “ FSVOL4 ” and “ FSVOL5 .” Likewise, the results in columns [2] and [4] of Table 4 are uncovered by estimating the specifications of model (1) that include the interaction between the variables “ PCI ” and “ EVI ,” and where the dependent variables are respectively “ FSVOL4 ” and “ FSVOL5 .”

TABLE 4

INTERACTION EFFECT OF PRODUCTIVE CAPACITY AND STRUCTURAL ECONOMIC VULNERABILITY ON FISCAL SPACE VOLATILITY OVER THE FULL SAMPLE (ESTIMATOR: FGLS (WITH PANEL-SPECIFIC FIRST-ORDER AUTOCORRELATION))

Note: 1) *p-value<0.1, **p-value<0.05, ***p-value<0.01; 2) Robust standard errors are in parenthesis; 3) The Pseudo R² is calculated as the correlation coefficient between the dependent variable and its predicted values; 4) Time dummies are included in the FGLS-based regressions.

IV. Estimation Outcomes

We note across columns [1] to [3] of Table 1 that at the 1% level, productive capacities reduce fiscal space volatility and that structural economic vulnerability heightens it, respectively, over the full sample, as well as the sub-samples of LDCs and non-LDCs. These findings confirm hypotheses 1 and 2 set out above. Interestingly, the magnitude of productive capacities exerts a stronger negative effect on fiscal space volatility in non-LDCs than in LDCs. Concurrently, structural economic vulnerability exerts a stronger positive effect on fiscal space volatility in LDCs than in non-LDCs. In terms of magnitude, we find that over the full sample, an increase in the value of the index of productive capacities by one point is associated with a reduction of fiscal space volatility by 0.011 points. In other words, an increase in the index of productive capacities by one standard deviation is associated with a reduction of fiscal space volatility of 0.065 points (= 6.052*0.0107). Similarly, over the full sample, an increase in the value of the index of structural economic vulnerability by one point is associated with a rise in fiscal space volatility by 0.00188 points. For LDCs and non-LDCs, the magnitude of the effect of productive capacities on fiscal space volatility amounts respectively to -0.00687 and -0.012. Likewise, for LDCs and non-LDCs, the magnitude of the effect of structural economic vulnerability on fiscal space volatility amounts to 0.00194 and 0.00147, respectively for LDCs and non-LDCs.

Regarding the control variables, we find over the full sample a positive effect of the real per capita income on fiscal space volatility, at the 1% level. Put differently, developing countries with higher incomes tend to exhibit higher fiscal space volatility than developing countries with relatively lower incomes. Trade openness and volatility of the inflation rate are not significantly associated with fiscal space volatility, at the 10% level. Finally, and without surprise, we find that fiscal space volatility is higher for outlier countries than for non-outlier countries, as the coefficient of the indicator “ DUMOUT ” is positive and significant at the 1% level across the three columns of Table 1. These findings concerning the control variables are confirmed in Tables 2 to 4.

Turning to the outcomes in Table 2, we find from column [1] of this table that the coefficient of the variable “ PCI_t-3 ” is negative and significant at the 1% level, while the interaction term associated with the variable “ [PCI_t-3]*[Log(GDPC)_t-3] ” is positive and significant at the 1% level. These outcomes tend to suggest that, on average, over the full sample, productive capacities negatively affect fiscal space volatility but only up to a level of the real per capita income; beyond that level, the effect of productive capacities on fiscal space volatility becomes positive. This level of real per capita income above which the effect of productive capacities on fiscal space volatility changes amounts to US$ million 3.372 [= exponential (0.0242/0.00161)]. The latter is far higher than the maximum value of real per capita income in the full sample, which is US$ 69679.1. We conclude that on average, over the full sample, productive capacities always exert a negative effect on fiscal space volatility (that is, regardless of the countries’ real per capita income), but the magnitude of this negative effect is higher withthe lower the real per capita income levels. In other words, less developed countries among developing countries experience a stronger negative effect of productive capacities on fiscal space volatility compared to relatively advanced developing countries. These findings are confirmed in Figure 4, which displays, at 95% confidence intervals, the marginal impact of productive capacities on fiscal space volatility, conditioned on real per capita income. This figure shows that productive capacities always negatively affect fiscal space volatility, with the magnitude of this negative effect becoming lower as countries experience higher real per capita income.

FIGURE 4.

MARGINAL IMPACTS OF “PCI” ON “FSVOL3” FOR VARYING LEVELS OF REAL PER CAPITA INCOME

Note: The variable “Log(GDPC)” is considered at year t-3.

Source: Author.

We now turn to affect outcomes in column [2] of Table 2. These results indicate that the coefficient of the variable “ EVI ” is positive and significant at the 1% level, while the interaction term of the variable (“ [EVI_t-3]*[Log(GDPC)_t-3] ”) is negative and significant at the 1% level. We therefore conclude that over the full sample, the effect of structural economic vulnerability on fiscal space volatility is negative for countries whose real per capita incomes are lower than US$ 12891 [= exponential (0.0106/0.00112)] and positive for countries whose real per capita incomes exceed US$ 12891. Figure 5 tends to confirm these findings. It shows at the 95% confidence intervals the marginal impact of structural economic vulnerability on fiscal space volatility, conditioned on real per capita income. It appears that this marginal impact is positive for countries whose real per capita incomes are lower than US$ 8481.25 [= exponential(9.045613)], with the magnitude of the positive effect of structural economic vulnerability on fiscal space volatility increasing as real per capita income decreases. At the same time, countries whose real per capita incomes range from US$ 8481.25 to US$ 28707.1 [= exponential(10.2649)] experience no significant effect of structural economic vulnerability. Note that the numbers “9.045613” and “10.2649” are obtained when constructing Figure 5 using the software Stata. They represent respectively the minimum and maximum values of the variable “ Log(GDPC) ” at which the marginal impact of structural economic vulnerability on fiscal space volatility becomes statistically nil at the 95% confidence interval. Finally, for countries whose real per capita incomes exceed US$ 28707.1, the effect of structural economic vulnerability on fiscal space volatility is negative, and the higher the real per capita income is (i.e., above US$ 28707.1), the greater the magnitude is of the negative effect of structural economic vulnerability on fiscal space volatility. This latter outcome suggests that more advanced developing countries are better equipped (for example in terms of financial and human capital resources, as well as capital stock) than relatively less advanced countries (including LDCs) to cope with the adverse economic and social effects of structural economic vulnerability that would translate into greater fiscal space volatility.

FIGURE 5.

MARGINAL IMPACTS OF “EVI” ON “FSVOL3” FOR VARYING LEVELS OF REAL PER CAPITA INCOME

Note: The variable “Log(GDPC)” is considered at year t-3.

Source: Author.

We now consider the outcomes reported in Table 3. To recall, these outcomes serve primarily to test the hypothesis 3 set out in section 2, that is, to examine the extent to which productive capacities affect fiscal space volatility for varying degrees of structural economic vulnerability. It appears from this table that the coefficients of the variables “ PCI_t-3 ” and “ [PCI_t-3]*[EVI_t-3] ” are negative and significant at the 1% level. These estimates suggest that productive capacities always influence negatively and significantly fiscal space volatility, regardless of the degree of structural economic vulnerability. In addition, the higher the level of structural economic vulnerability is, the greater is the magnitude of the negative effect of productive capacities on fiscal space volatility. These findings are confirmed in Figure 6, which presents, at 95% confidence intervals, the marginal impact of productive capacities on fiscal space volatility, conditioned on the degree of structural economic vulnerability. It appears from this figure that this marginal impact is always negative and significant and that it decreases as the level of structural economic vulnerability rises.

The key message conveyed by the outcomes reported in Table 3 (along with Figure 6) is that productive capacities exert a stronger negative effect on fiscal space volatility in countries that face a higher degree of structural economic vulnerability than in counties with a relatively low level of structural economic vulnerability.

FIGURE 6.

MARGINAL IMPACTS OF “PCI” ON “FSVOL3” FOR VARYING DEGREES OF STRUCTURAL ECONOMIC VULNERABILITY

Note: The variable “EVI” is considered at year t-3.

Source: Author.

These findings confirm hypothesis 3.

Finally, we take up the results in Table 4 regarding the robustness of the outcomes in column [1] of Tables 1 and 3. We note from columns [1] and [3] of Table 4 that over the full sample, productive capacities exert a negative and significant effect (at the 1% level) on fiscal space volatility, while structural economic vulnerability exerts a positive and significant effect (also at the 1% level) on fiscal space volatility. These findings confirm hypotheses 1 and 2 and align with those obtained in column [1] of Table 1. Interestingly, the magnitudes of these effects are higher in terms of absolute values for “ FSVOL5 ” than for “ FSVOL4 ,” and then for “ FSVOL3 .”

Furthermore, in both columns [2] and [4] of Table 4, we obtain findings that align with those in Table 3. Specifically, we find that the coefficients of the variables “ PCI_t-4 ” and “ [PCI_t-4]*[EVI_t-4] ” are all negative and significant at the 1% level, as shown in column [2] of Table 4. Similarly, the estimates associated with the variables “ PCI_t-5 ” and “ [PCI_t-5]*[EVI_t-5] ” are also all negative and significant at the 1% level, as shown in column [4] of Table 4. We therefore reach conclusions identical to those derived from Table 3, whereby productive capacities always influence negatively and significantly fiscal space volatility, with the magnitude of this negative effect increasing as the degree of structural economic vulnerability rises. Figures 7 and 8 confirm these findings. These two figures present at 95% confidence intervals the marginal impact of productive capacities respectively on fiscal space volatility indicators FSVOL4 and FSVOL5, conditioned on the degree of structural economic vulnerability. In both figures, the marginal impact of productive capacities respectively on fiscal space volatility is always negative and significant, and decreases as the degree of structural economic vulnerability rises. In other words, highly structurally vulnerable countries tend to experience a stronger negative effect of productive capacities on fiscal space volatility than relatively less structurally vulnerable countries.

FIGURE 7.

MARGINAL IMPACTS OF “PCI” ON “FSVOL4” FOR VARYING DEGREES OF STRUCTURAL ECONOMIC VULNERABILITY

Note: The variable “EVI” is considered at year t-4.

Source: Author.

FIGURE 8.

MARGINAL IMPACTS OF “PCI” ON “FSVOL5” FOR VARYING DEGREES OF STRUCTURAL ECONOMIC VULNERABILITY

Note: The variable “EVI” is considered at year t-5.

Source: Author

V. Conclusion

This article investigated the effects of productive capacities and structural economic vulnerability on fiscal space volatility in developing countries using a panel dataset of 116 countries over the period of 2000 to 2018. The results indicate that the development of productive capacities is associated with lower fiscal space volatility, while higher structural economic vulnerability is associated with greater fiscal space volatility. While the strengthening of productive capacities exerts a stronger negative effect on fiscal space volatility in non-LDCs than in LDCs, the increase in structural economic vulnerability induces greater fiscal space volatility in LDCs than in non-LDCs. The analysis of the extent to which the effects of productive capacities and structural economic vulnerability on fiscal space volatility vary across countries in the full sample provides a better picture of these effects. We found that regardless of countries’ real per capita income, the development of productive capacities is always associated with lower fiscal space volatility. However, the lower the real per capita income is, the higher the magnitude of the negative effect of productive capacities on fiscal space volatility becomes. These findings do not contradict those observed over LDCs versus non-LDCs, as the latter represent average effects over each of these sub-samples while the former are marginal effects for varying levels of real per capita income. We also find that for less developed countries, i.e., those whose real per capita incomes are lower than US$ 8481.25 (this set of countries includes LDCs), structural economic vulnerability enhances fiscal space volatility, while countries whose real per capita incomes are higher than US$ 28707.1 tend to experience a negative effect of structural economic vulnerability on fiscal space volatility.

Finally, the analysis reveals that highly structurally vulnerable countries tend to experience a stronger negative effect of productive capacities on fiscal space volatility than relatively less structurally vulnerable countries.

From a policy perspective, these findings have shown that if highly vulnerable countries are to reduce their fiscal space volatility, they need to foster their productive capacities. Strengthening productive capacities in developing countries would surely require efforts by both national policymakers and international institutions, in a coordinated manner.

We recognize that the concept of “fiscal space” and hence that of “fiscal space volatility” are complex and difficult to measure. The present study aims to shed the first light on the effect of productive capacities and structural economic vulnerability on fiscal space volatility by relying on a simple measure of fiscal space. An avenue for future research could involve using other possible indicators of fiscal space (and hence of fiscal space volatility) and eventually other indicators of productive capacities to conduct such an analysis.