Abstract
This paper reviews the calibration of regional debt and deficit ceilings considering the macroeconomic trade-offs and risks involved. Overall, it seems advisable to maintain the two ceilings at their previous levels. The 70 percent of GDP debt ceiling seems to strike the right balance between growth and fiscal prudence considerations. Model-based simulations show that raising the deficit ceiling from 3 to 4 percent of GDP would likely undermine domestic and external stability, except in very specific circumstances, that is, if access to external financing is unconstrained and significant progress in fiscal transparency is achieved.
An Analysis of Medium-Term Regional Fiscal Targets in the Waemu1
This paper reviews the calibration of regional debt and deficit ceilings considering the macroeconomic trade-offs and risks involved. Overall, it seems advisable to maintain the two ceilings at their previous levels. The 70 percent of GDP debt ceiling seems to strike the right balance between growth and fiscal prudence considerations. Model-based simulations show that raising the deficit ceiling from 3 to 4 percent of GDP would likely undermine domestic and external stability, except in very specific circumstances, that is, if access to external financing is unconstrained and significant progress in fiscal transparency is achieved.
A. Introduction
1. In the aftermath of the deep crisis caused by the COVID-19 pandemic, the WAEMU authorities are revisiting the regional fiscal framework that will guide policy over the medium-term. The fiscal framework includes, among other features, convergence criteria at the regional level for debt (ceiling of 70 percent of GDP) and the overall deficit (ceiling of 3 percent of GDP).2 These criteria, which are supposed to be complied with after a convergence period, have been suspended in the context of the crisis. The WAEMU commission is expected to make a proposal to revise the current framework of fiscal rules by the end of 2021. In addition, in June 2021, a decision was taken at the ECOWAS level (15 countries, including the 8 WAEMU countries) to maintain the current debt and deficit numerical targets (of 70 percent of GDP and 3 percent of GDP), but adopt a more gradual convergence path, with a majority of countries expected to reach the 3 percent deficit ceiling by 2024.3 The reflection on a possible revision of the fiscal rules will need to balance a number of objectives, including weighing large development needs and associated demands for higher public spending with fiscal prudence to preserve sustainability.
2. A revision of the fiscal rules framework can encompass several dimensions, but this chapter focuses exclusively on the calibration of the debt and deficit ceilings. Other dimensions including the selection of rules as well as supporting arrangements and institutions such as enhancements in public financial management and monitoring and enforcement mechanisms were discussed in detail in last year’s Article IV consultation (Nguyen-Duong and Selim, 2021). Moreover, the chapter does not examine the duration of the convergence period, but rather focuses on the fiscal targets once the convergence is achieved.
3. The paper is divided in two main sections. In the first section, the paper discusses the calibration of the debt ceiling following different approaches. It also analyzes whether there is scope to prudently increase debt levels in the region. The second section of the paper considers the calibration of the deficit ceiling and the macroeconomic implications of possible revisions of the ceiling using a structural macroeconomic model. One important caveat to bear in mind is that the chapter focuses mostly on the regional implications of revisions to fiscal rules. Country-specific considerations—outside the scope of the paper—are also crucial and should also be taken into account if the rules were to be revised.
4. This paper does not discuss the possible differentiation of member states’ ceilings within the single regional ceiling. The analysis presented in this chapter focuses on supranational rules from a regional perspective with a view to preserving debt sustainability, supporting growth and fostering external stability at the union level. It is possible to envisage some differentiation of the ceilings across countries. Although this exercise is out of the scope of the chapter, some considerations should be kept in mind. There is little historical precedent to rule differentiation within a currency union or federation. In fact, uniform rules across member states facilitate fiscal coordination, which is essential to prevent negative spillover effects and preserve external sustainability at the regional level. In addition, in the context of WAEMU member states, it may be difficult to calibrate country-specific debt ceilings due to a number of constraints, not least data availability. Nevertheless, the analysis presented here does not preclude such differentiation. Member states could, for instance, pursue different debt anchors at the country-level, while maintaining an aggregate debt ceiling at 70 percent of GDP at the regional level. A more comprehensive analysis is needed to assess the benefits and risks of such differentiation in the case of the WAEMU.
B. Calibration of the Debt Ceiling
5. This chapter follows a sequential approach to calibrate fiscal rules in the WAEMU, which starts with obtaining estimates for the debt anchor and subsequently infer the associated fiscal deficit ceiling. This sequential approach to calibration is further explained in IMF (2018). The first step of the exercise is to estimate the debt anchor, which is the level of debt towards which countries are expected to converge in the medium to long term. To estimate the debt anchor, three approaches are used: i) one based on the calculation of a debt limit above which dynamics become explosive, from which the anchor is derived given a safety margin or buffer; ii) the second approach focuses on the debt level that would maximize growth in the context of a simple model; iii) in the final approach, the debt anchor is obtained based on the current IMF’s debt sustainability framework (IMF, 2018). These approaches try to capture the fundamental tradeoff between the need to keep debt levels in check to foster economic resilience and market confidence versus the competing objective of allowing space for borrowing to finance the large development needs.
Approach 1: “Safe” Debt Estimate
6. The first approach uses a method based on precautionary considerations to calibrate the debt anchor for WAEMU countries. The approach is presented in detail in IMF (2018). The “safe” debt level is defined as the debt-to-GDP ratio that ensures that debt dynamics remain under control even if adverse economic shocks occur. The approach is based on two main principles. Firstly, we assume that there is a point beyond which debt dynamics can spiral out of control, which we call a maximum debt limit. The second principle is that the fiscal framework should aim at keeping debt well below this limit. Because countries are vulnerable to significant macroeconomic and fiscal shocks, including changes in market sentiment, there must be a sufficient safety margin between the debt anchor and the debt limit.
7. In that context, the debt anchor is measured as the difference between the debt limit and the safety buffer. Therefore, the analysis entails three separate steps: (i) first, an estimate of the maximum debt limit is obtained; (ii) then, we proceed to estimate the required safety margin; (iii) and finally the debt anchor is inferred as the debt limit minus the safety margin. More specifically, the required safety margin is estimated through stochastic simulations. To do so, the distribution of macroeconomic and fiscal shocks faced by the country in the past is estimated. Subsequently, future debt trajectories under these shocks are simulated over a 6-year horizon. This creates a fan chart of debt realizations, which allows for the calibration of the debt anchor and calculation of the probability that public debt would exceed the maximum debt limit in the medium-term.
Estimating a Regional Debt Limit for WAEMU Countries
8. One approach to calculate such debt limit is based on the principle of “fiscal fatigue” (Gosh et al., 2013) and consists in estimating the limit above which debt cannot be stabilized in times of fiscal stress. Policymakers cannot realistically claim that they will do “whatever it takes” to generate primary surpluses sufficiently large to stabilize debt in an unfavorable macroeconomic environment. This may be because very high primary surpluses are politically unacceptable or economically unachievable (due, for instance, to Laffer curve effects). The maximum debt level is reached when negative macroeconomic conditions, measured by the interest-growth differential and the exchange rate, create an upward pressure on debt, but the government cannot increase further the primary balance to offset this pressure. We approximate the maximum debt level that could be stabilized by calculating the ratio of the maximum achievable primary balance to the interest rate-economic growth differential (r-g) under stress:
9. Using historical evidence for WAEMU countries, we estimate a debt limit of about 80 percent of GDP using this approach. Based on data from the April 2021 vintage of the WEO database over the period 2000–20, we estimate that the 95th percentile of the distribution of the interest growth differential (based on effective interest rates on debt) for countries in the region stands at around 3.3 percent. Regarding the maximum achievable primary balance, using historical evidence from the WEO dataset since 1996, we calculate the 95th percentile of the distribution of primary balances for WAEMU countries and find a value of 2.7 percent of GDP. This larger maximum primary surplus would result in a debt limit of 80 percent of GDP.
10. An alternative approach to estimate the debt limit is based on the principle of “debt-carrying capacity.” More specifically, the approach focuses on the ratio of interest expenses to revenues (excluding grants), as an indicator of the ability to repay debt. Arguably, for sub-Saharan African countries where revenue mobilization is low, a measure of this ability to repay debt is highly relevant for sustainability. In fact, empirical evidence shows that the interest to revenue ratio is tightly linked to fiscal stress (Bentum, David, Slavov, and Sode, 2021). In addition, econometric models applied to emerging markets and developing economies suggest that thresholds for the ratio that would signal upcoming fiscal stress would range from 16 to 19 percent. Using these thresholds for the interest to revenue ratio, denoted τ it is possible to obtain the associated debt limit level beyond which the risk of fiscal stress increases significantly by using the following relationship:
Using the average revenue (excluding grants) to GDP ratio and average effective interest rates over 2015–2019 for the WAEMU from the WEO database, one would obtain the debt limits outlined in Table depending on the threshold used.
11. Overall, the results indicate that a debt limit of around 80 percent of GDP for the WAEMU seems a reasonable estimate. The approaches based on fiscal fatigue and debt servicing capacity yield broadly similar results. One important caveat to these calculations of the debt limit is that key inputs such as the maximum primary balance or the interest-growth differentials were estimated based on historical information. They are subject to uncertainty and may change in the near future in manners that differ from historical patterns (see robustness checks below).
Estimating the Safety Buffer
12. Having computed a debt limit, we now turn to estimate a required safety margin based on the history of macroeconomic and fiscal shocks that countries in the region tend to face. The buffer reflects two factors: i) the history of macroeconomic shocks for countries in the region, ii) contingent liabilities, estimated at 3 percent of GDP every 6-years based on the evidence discussed in Bova et al. (2019). We start by estimating the distribution of macroeconomic and fiscal shocks facing WAEMU countries derived from a multivariate normal distribution based on annual data at the regional level for key macroeconomic and fiscal variables (namely, real GDP growth, the primary balance, real interest rates and the real exchange rate).4 The exchange rate variable is particularly important since it can drive large and sudden changes in debt ratios. These shocks are subsequently used to perform simulations of future debt trajectories based on the standard debt dynamics equation and a fiscal reaction function. The resulting debt paths are presented in a fan chart (see Figures below). Each trajectory represents the evolution of debt under a certain shock scenario. The debt anchor is the initial point of the different simulations presented below and it is calibrated, so that the fan chart stays below the debt limit over a 6-year horizon with a high probability.
13. The simulations point to a debt anchor of around 70 percent of GDP if policymakers are willing to accept a 10 percent probability of breaching the maximum debt limit in the medium-term. An anchor of around 70 percent of GDP is considered a “safe” level of debt, ensuring that countries in the region can withstand negative shocks without breaching the debt limit (assumed to be 70 percent of GDP) by the 6th year with very high probability.
14. The anchor would be slightly lower if one accounts for the possible realization of contingent liabilities. Certain types of transactions may lead to a disconnect between the evolution of deficits and debt in the standard debt dynamics equation levels (the so-called stock flow adjustments or SFAs). These include the likely realization of contingent liabilities (e.g. arising from the recapitalization of a bank or state-owned enterprise); off budget operations; or other large operations in financial assets that are recorded below the line. Bova et al. (2019) based on a comprehensive dataset of contingent liability realizations in advanced and emerging markets for the period 1990–2014 find that on average a country would be expected to have experienced a contingent liability realization every 12 years or so with a fiscal cost of 6.1 percent of GDP per episode. In the context of our simulations, if we consider that the expected realization of contingent liabilities is 3 percent of GDP over 6 years, the implied debt anchor would be 68 percent of GDP.
WAEMU: Fiscal Anchors and Safety Buffers
Citation: IMF Staff Country Reports 2022, 068; 10.5089/9798400204579.002.A002
Source: IMF Staff Estimates.15. Overall, the analysis suggests a debt anchor of around 70 percent of GDP, but there is uncertainty around that level. This corresponds to a debt limit of 80 percent of GDP minus a safety buffer of 10 percent of GDP, deemed appropriate for the region given the history of shocks and expected realization of contingent liabilities. Nevertheless, it is important to bear in mind that the size of the safety buffer was estimated based on historical information. Going forward, there might be important risks and structural trends that would call for a wider safety margin, such as for example monetary policy normalization in Advanced Economies pushing up global interest rates. Robustness checks show that a 100 basis points increase in interest rates relative to recent levels would increase the size to the buffer to over 16 percent of GDP (bringing the debt anchor below 65 percent of GDP if a debt limit of 80 percent of GDP is considered). Nonetheless, there is also uncertainty about some elements of the calculation of the debt limit, which may support a higher debt limit than 80 percent of GDP. For example, if efforts to mobilize additional domestic revenues materialize and the revenues excluding grants to GDP ratio used as an input in Table 1 increase from 14.8 to 16 percent of GDP, the debt limit would be revised upward to 95 percent of GDP, keeping other things equal.
WAEMU: Estimates of the Debt Limit
WAEMU: Estimates of the Debt Limit
| Threshold | ||
| 16% | 19% | |
| Revenue (excluding grants)/GDP | 14.8% | 14.8% |
| Effective Interest Rate | 3.2% | 3.2% |
| Debt Limit | 74.0% | 87.9% |
WAEMU: Estimates of the Debt Limit
| Threshold | ||
| 16% | 19% | |
| Revenue (excluding grants)/GDP | 14.8% | 14.8% |
| Effective Interest Rate | 3.2% | 3.2% |
| Debt Limit | 74.0% | 87.9% |
Approach 2: Growth-Maximizing Debt Level
16. Another approach to determine the appropriate debt anchor attempts to estimate a debt ratio that maximizes growth. This approach differs from the one discussed in the previous section, as it places greater emphasis on increasing debt to finance development needs rather than focusing on limiting debt to foster resilience and fiscal prudence. This chapter uses a theoretical model constructed by Checherita-Westphal and others (2014) to derive the level of public sector debt beyond which debt starts to have a negative effect on growth, even when considering the positive effect of public investment (financed through additional debt) on GDP.
17. The optimal debt-to-GDP ratio depends on how additional public capital translates to higher GDP levels (the elasticity of the public capital stock). The level of public sector debt level that maximizes output growth is derived in an infinite horizon model with flexible prices and wages with a production function that includes labor (L); private capital (K); and public capital (Kg). Output is given by the production function below, where α is the output elasticity of the public capital stock:
Assuming that public debt is used exclusively for public capital financing (“golden rule”), the optimal debt to GDP ratio (D*) is given by the following expression and depends crucially on the output elasticity of the public capital stock (α).
18. The elasticity is estimated econometrically using pooled data for WAEMU countries over the period 1960–2015. We use data from the Penn-World Tables version 10.0 (Feenstra and others, 2015) and from the IMF’s Investment and Capital Stock Dataset to estimate the parameter α.5 We follow Checherita-Westphal and others (2014) and use two different specifications. In the first model, output and labor are expressed as a share of the private capital stock. In the second model variables are expressed in per capita terms (except for the ratio of public capital to private capital). The results are presented in Table 2.
19. Using this “optimal debt” approach, the debt anchor would range between 65 and 84 percent of GDP. All regressions yield point estimates of around 0.31, which would imply an optimal debt to GDP ratio target of about 74 percent. Considering a 90 percent confidence interval around the point estimate (that is, an upper bound of 0.34 and a lower bound of 0.28) would yield optimal debt to GDP ratios ranging from 65 percent to 84 percent.
WAEMU: Estimates of the Output Elasticity of Public Capital
WAEMU: Estimates of the Output Elasticity of Public Capital
| (1) | (2) | (3) | (4) | |
|---|---|---|---|---|
| ln(Y/K) | ln(Y/L) | ln(Y/K) | ln(Y/L) | |
| ln(Kg/K) | 0.312*** [0.016] |
0.312*** [0.016] |
0.307*** [0.016] |
0.307*** [0.016] |
| ln(L/K) | 0.366*** [0.021] |
0.377*** [0.021] |
||
| ln(K/L) | 0.634*** [0.021] |
0.623*** [0.021] |
||
| Trend | 0.003*** [0.001] |
0.003*** [0.001] |
||
| Constant | 0.380*** [0.016] |
0.380*** [0.016] |
0.273*** [0.041] |
0.273*** [0.041] |
| Observations | 480 | 480 | 480 | 480 |
| R-squared | 0.738 | 0.659 | 0.743 | 0.664 |
WAEMU: Estimates of the Output Elasticity of Public Capital
| (1) | (2) | (3) | (4) | |
|---|---|---|---|---|
| ln(Y/K) | ln(Y/L) | ln(Y/K) | ln(Y/L) | |
| ln(Kg/K) | 0.312*** [0.016] |
0.312*** [0.016] |
0.307*** [0.016] |
0.307*** [0.016] |
| ln(L/K) | 0.366*** [0.021] |
0.377*** [0.021] |
||
| ln(K/L) | 0.634*** [0.021] |
0.623*** [0.021] |
||
| Trend | 0.003*** [0.001] |
0.003*** [0.001] |
||
| Constant | 0.380*** [0.016] |
0.380*** [0.016] |
0.273*** [0.041] |
0.273*** [0.041] |
| Observations | 480 | 480 | 480 | 480 |
| R-squared | 0.738 | 0.659 | 0.743 | 0.664 |
Approach 3: Thresholds Extracted from the Debt Sustainability Analysis
20. The Bank-Fund’s debt sustainability framework (DSF) for Low Income Countries can also provide insights on the debt anchor for WAEMU countries. The framework includes the estimation of indicative thresholds and benchmarks for different debt burden indicators (for debt levels and debt service), which vary depending on whether the country is classified as having weak, medium, or strong debt carrying capacity. In turn, the classification of debt carrying capacity is based on a composite measure that includes indicators of institutional strength (CPIA index), economic growth, reserve coverage, remittances, and world growth.
21. Although the LIC DSF does not include explicit thresholds for total nominal debt ratios, it is possible to infer such limits based on the past ratio of present value to nominal debt. Debt stock indicators used in the LIC DSF methodology are expressed in net present values (NPV) to reflect the grant element, or concessionality, of the debt (IMF 2018). The NPV of public debt is calculated as the discounted sum of future debt repayments (principal and interests). This can be proxied as the product of the nominal stock of debt and a discount factor. Using this approximation, it is possible to multiply the 2020 ratio of nominal to NPV debt by the applicable LIC-DSF benchmark to obtain an implicit threshold for a country’s nominal debt ratios (Table 3). Such a value provides an estimate of the maximum level of nominal debt above which a country would very likely breach the LIC-DSF benchmarks, given current assumptions concerning the concessionality of the debt.
WAEMU: Implied Nominal Debt Limits
As included in the staff report.
As reported in the LIC-DSF Guidance note.
WAEMU: Implied Nominal Debt Limits
| Benin | Medium | 55 | 45.0 | 39.5 | 1.14 | 63 |
| Burkina Faso | Medium | 55 | 46.1 | 38.8 | 1.19 | 65 |
| Cote d’Ivoire | Medium | 55 | 44.7 | 40.8 | 1.10 | 60 |
| Guinea-Bissau | Weak | 35 | 78.4 | 65.7 | 1.19 | 42 |
| Mali | Medium | 55 | 44.1 | 38.4 | 1.15 | 63 |
| Niger | Medium | 55 | 41.7 | 35.0 | 1.19 | 66 |
| Senegal | Strong | 70 | 67.6 | 59.5 | 1.14 | 80 |
| Togo | Medium | 55 | 67.1 | 61.4 | 1.09 | 60 |
| Average implied upper limit on nominal public debt (weighted by share of 2020 WAEMU GDP) | 65 | |||||
As included in the staff report.
As reported in the LIC-DSF Guidance note.
WAEMU: Implied Nominal Debt Limits
| Benin | Medium | 55 | 45.0 | 39.5 | 1.14 | 63 |
| Burkina Faso | Medium | 55 | 46.1 | 38.8 | 1.19 | 65 |
| Cote d’Ivoire | Medium | 55 | 44.7 | 40.8 | 1.10 | 60 |
| Guinea-Bissau | Weak | 35 | 78.4 | 65.7 | 1.19 | 42 |
| Mali | Medium | 55 | 44.1 | 38.4 | 1.15 | 63 |
| Niger | Medium | 55 | 41.7 | 35.0 | 1.19 | 66 |
| Senegal | Strong | 70 | 67.6 | 59.5 | 1.14 | 80 |
| Togo | Medium | 55 | 67.1 | 61.4 | 1.09 | 60 |
| Average implied upper limit on nominal public debt (weighted by share of 2020 WAEMU GDP) | 65 | |||||
As included in the staff report.
As reported in the LIC-DSF Guidance note.
22. Under simplifying assumptions, the public sector debt benchmarks from the DSF suggest a debt threshold in nominal terms of around 65 percent of GDP for the WAEMU region. Most countries in the region are classified as having “medium” debt carrying capacity with an associated public sector debt benchmark in net present value terms of 55 percent of GDP, except for Guinea-Bissau (which is classified as having weak debt carrying capacity) and Senegal (classified as having strong debt carrying capacity). Converting the public sector debt thresholds from net present value to nominal terms for each country and aggregating by using the countries’ shares in regional GDP results in a nominal debt benchmark of 65 percent of GDP (while individual country limits range from 42 to 80 percent of GDP in nominal terms).
Putting All These Approaches Together
23. Overall, changing the regional debt ceiling presently embedded in the fiscal framework does not seem to be warranted. Considering the simple average and median of estimates of different approaches, the 70 percent of GDP ceiling for the region is appropriate and strikes the right balance between fiscal prudence and growth considerations.
24. While the analysis presented in this chapter focuses on supranational rules with a regional perspective, it could be possible to envisage some differentiation of the ceilings across countries. The analysis shows that a 70 percent of GDP debt ceiling is appropriate at the aggregate level. But this does not necessarily preclude a differentiation of ceilings across countries within this aggregate. This differentiation could be motivated by the fact that WAEMU countries have different capacity to repay debt and different initial conditions. Recently, some proposals have been made in this direction in the context of the Euro Area (Martin, Pisani-Ferry, and Ragot, 2021). Although this exercise is out of the scope of the chapter, two considerations should be kept in mind. First, there could be difficulties in implementing such differentiation in practice. There is little historical precedent to rule differentiation within a currency union. It could make fiscal policy coordination and monitoring more complex. Second, in the context of WAEMU member states, it may be difficult to calibrate country-specific debt ceilings due to a number of constraints, not least data availability. For instance, when estimating the debt limit, it could be difficult to ascertain how maximum primary balances and interest-growth differentials under stress vary across countries.
C. Calibration of the Deficit Ceiling
From Debt Anchor to Overall Deficit Ceiling
25. Absent stock-flow adjustments, the current deficit ceiling of 3 percent of GDP would stabilize debt at around 40 to 50 percent of GDP. The overall deficit (OB) that would make debt converge in the absence of shocks can be obtained by using the standard relationship linking the debt as a share of GDP (D) and nominal GDP growth (θ):
Assuming that potential medium-term real growth ranges between 5 to 6 percent for countries in the region in line with current IMF projections and taking the BCEAO’s objective of an inflation rate of 2 percent (+/- 1 percent band), would yield an estimate of nominal potential growth (θ) rate of between 6 and 9 percent.6 This implies that the 3 percent of GDP ceiling embedded in the current framework would stabilize debt between 40 to 50 percent of GDP, which is close to current debt levels for several countries in the WAEMU.
26. A modestly higher deficit ceiling of 4 percent of GDP may be compatible with a debt ceiling ranging from 50 to 70 percent of GDP. Given the previously discussed assumptions about potential nominal growth, a 4 percent deficit ceiling would stabilize debt at around 70 percent of GDP if a nominal growth rate of 6 percent is assumed. Considering a nominal growth rate of 8 percent, a 4 percent of GDP deficit would stabilize debt at 54 percent of GDP.
27. The rest of the chapter analyzes the macroeconomic implications of raising the regional deficit ceiling to 4 percent of GDP. International experience with deficit ceilings suggests that they tend to become a “focal” point with countries converging to the ceiling over time (from initial levels above and below it). Therefore, the shift to a 4 percent of GDP fiscal deficit ceiling would likely increase the deficit path in the region (the Annex discusses the consequences of a tax-financed increase in expenditure). The macroeconomic implications can be grouped in three main categories:
External stability. Permanently wider fiscal deficits could have implications for the sustainability of the currency peg in the WAEMU, given that they would likely constitute a net drain on international reserves. Larger deficits would have a direct impact on reserves through higher imports, but could also have adverse effects on competitiveness through real exchange rate appreciation.
Absorptive capacity of the regional market and crowding-out effects. At least part of the wider fiscal deficits would likely need to be accommodated by the regional market. Given the lack of liquidity in the regional market for sovereign bonds (IMF 2021), additional borrowing requirements could create financial pressures, particularly if banks (by far the most important players in the market) cannot accommodate them. This could result in a tightening of financial conditions and crowding-out of financing for the private sector. Staff analysis indicates that a 1 percent increase in government securities held by regional banks as share of M2 is on average associated with a 3.6 percent decline in the subscription rates of sovereign bonds’ issuances (Figure 2). This suggests that large volumes of additional domestic borrowing, if they materialize, could test the absorption capacity of regional markets.
Effect on debt dynamics. Higher deficits are likely to result in more elevated debt ratio trajectories, although the effect may also depend on how the deficit is used and whether it finances growth-enhancing spending. Another important consideration is whether the fiscal deficit, as defined by the rule, captures all the debt creating flows. A number of fiscal operations, so-called Stock-Flow Adjustments (SFAs), create a gap between “above the line” deficits and the evolution of debt, beyond the role played by automatic drivers of debt such as growth rates, interest rates, and the exchange rate.7 These SFAs include off-budget and/or “below the line” operations (acquisitions of financial assets) as well as the realization of contingent liabilities. On average, Staff estimates indicate that SFAs for the WAEMU region typically stood around 1 percent of GDP between 2013–2019 (Figure 2). Preliminary estimates for 2020 suggests that SFAs were historically high in part due to accounting issues (such as the timing of disbursements of investment projects).8 IMF staff’s baseline projections typically assume convergence to 3 percent deficit ceiling without large SFAs (on average they would amount to 0.2 percent of GDP over 2021–2016).
WAEMU: Stock-Flow Adjustments and Regional Market Absorptive Capacity
Citation: IMF Staff Country Reports 2022, 068; 10.5089/9798400204579.002.A002
Source: IMF Staff Estimates.28. The rest of the analysis will center around two policy options. These two options are assessed using a structural macroeconomic model (Box 1). The difference between these two options pertains to how the deficit is measured:
Option 1: 4 percent deficit ceiling measured “above the line”. The first option considers an increase in the deficit ceiling to 4 percent of GDP measured “above the line” (as revenue minus expenditure). This implies that the definition of the deficit would remain the same as the one prevailing under the current rule, but there is simply an increase in the ceiling from 3 percent of GDP to 4 percent of GDP. Relative to IMF staff baseline projections, this option would imply higher financing needs of, at least, 1 percent of GDP.
Option 2: 4 percent deficit ceiling measured “below the line”. The second option focuses on an increase in the deficit ceiling to 4 percent of GDP, but with the deficit redefined “below the line”—as the change in financial liabilities minus the change in financial assets.9
Option 1: raising the fiscal deficit ceiling to 4 percent of GDP “above the line”
29. A structural model is used to simulate four scenarios that illustrate the effects of the 4 percent of GDP deficit target under alternative assumptions regarding the composition of financing and spending.10 The scenarios differ in three main dimensions: i) the composition of spending (government consumption versus investment or spending to increases potential output); ii) whether the additional deficit is financed by external resources of whether it is financed in the regional market iii) whether stock flow adjustments are contained or whether 1 percent of GDP in SFAs are added. A summary of the assumptions is presented in Table 4:
The first scenario assumes a gradual convergence to an overall deficit of 4 percent of GDP by 2024 and the deficit would stay at that level permanently thereafter (1 percent of GDP higher than the baseline deficit ceiling of 3 percent of GDP). Moreover, half of the permanent increase in the deficit is devoted to an increase in public investment, which would boost medium-term potential output. Under this first scenario, the total increase in the deficit would be financed by external resources (additional donor funds and access to international markets).
The second scenario considers the same convergence path to a permanent increase in the deficit to 4 percent of GDP, externally financed, but with the additional spending geared entirely towards current expenditures.
The third scenario considers the same convergence path to a permanent increase in the deficit to 4 percent of GDP with the additional spending geared towards current expenditures, but mostly domestically financed with an ensuing increase in domestic interest rates (a risk premium increase of 50 basis points relative to baseline).
Finally, the fourth scenario, envisages a convergence to a deficit ceiling of 4 percent of GDP by 2024, but also takes into account the stock-flow adjustments and therefore the “actual” deficit would de facto increase by 2 percent of GDP relative to the baseline deficit ceiling of 3 percent of GDP. The increase would be entirely devoted to additional public consumption and a large share of financing in the regional market, with an ensuing increase in the domestic rates (a risk premium increase of 80 basis points relative to the baseline).
WAEMU: Key Assumptions Under Different Scenarios for an Increase in the Deficit Ceiling Under Option 1
WAEMU: Key Assumptions Under Different Scenarios for an Increase in the Deficit Ceiling Under Option 1
| Deficit | Expenditure composition | Financing | |
|---|---|---|---|
| Scenario 1: No stock-flow adjustments (SFAs); high quality spending; external financing | Convergence to 4 percent of GDP by 2024 (instead of 3 percent in baseline) | Half of the increase in the deficit relative to baseline is due to higher public investment. | Higher deficit relative to baseline financed fully externally, no effect on domestic rates. |
| Scenario 2: No stock-flow adjustments; lower quality spending; external financing | Convergence to 4 percent of GDP by 2024 | Does not increase relative to baseline. | Higher deficit relative to baseline financed fully externally, no effect on domestic rates. |
| Scenario 3: No stock-flow adjustments; lower quality spending; domestic financing | Convergence to 4 percent of GDP by 2024 | Does not increase relative to baseline. | Recourse to domestic financing leads to increase in interest rates (risk premium) by 50 basis points. |
| Scenario 4: Includes stock-flow adjustments of 1 percent of GDP; lower quality spending; domestic financing | Convergence to 4 percent of GDP, but SFAs bring the de facto deficit to 5 percent of GDP. | Does not increase relative to baseline. | Recourse to domestic financing leads to increase in interest rates (risk premium) by 80 basis points. |
WAEMU: Key Assumptions Under Different Scenarios for an Increase in the Deficit Ceiling Under Option 1
| Deficit | Expenditure composition | Financing | |
|---|---|---|---|
| Scenario 1: No stock-flow adjustments (SFAs); high quality spending; external financing | Convergence to 4 percent of GDP by 2024 (instead of 3 percent in baseline) | Half of the increase in the deficit relative to baseline is due to higher public investment. | Higher deficit relative to baseline financed fully externally, no effect on domestic rates. |
| Scenario 2: No stock-flow adjustments; lower quality spending; external financing | Convergence to 4 percent of GDP by 2024 | Does not increase relative to baseline. | Higher deficit relative to baseline financed fully externally, no effect on domestic rates. |
| Scenario 3: No stock-flow adjustments; lower quality spending; domestic financing | Convergence to 4 percent of GDP by 2024 | Does not increase relative to baseline. | Recourse to domestic financing leads to increase in interest rates (risk premium) by 50 basis points. |
| Scenario 4: Includes stock-flow adjustments of 1 percent of GDP; lower quality spending; domestic financing | Convergence to 4 percent of GDP, but SFAs bring the de facto deficit to 5 percent of GDP. | Does not increase relative to baseline. | Recourse to domestic financing leads to increase in interest rates (risk premium) by 80 basis points. |
The Structural Model Used for the Simulations
A structural general equilibrium model, akin to the one described in Andrle and others (2015), is used to assess the macroeconomic effects of the higher deficit scenarios relative to the IMF WEO baseline. Simulations are produced using the African version of the Flexible System of Global Models (FSGM) calibrated to the WAEMU region. In the model, inflation, imports and GDP are determined by aggregated demand and potential output. Private consumption depends mainly on current income (liquidity constraint households) and, to a lesser extent, on the interest rate. Private investment depends on the gap between current and desired capital stock (a function of financial conditions and potential output). Domestic financial conditions are set by the central bank in the form of a domestic nominal interest rate, equal to the euro area interest rate plus an interest premium. Potential output depends on labor supply, private capital stock, and a productivity term that increases with public capital stock.
In addition to standard model features, the simulations also assume that monetary policy adjusts the domestic interest rate to the level of official reserves. To guarantee the fixed nominal exchange rate regime, official reserves changes match any discrepancy between the current account and net private capital outflows. To ensure the sustainability of the exchange rate regime, monetary policy adjusts the domestic interest rate to the level of official reserves. Therefore, the interest premium in the model captures policy decisions to mitigate changes in official reserves. For instance, when net capital outflows or a more deteriorated trade balance reduce official reserves, the central bank increases the nominal interest rate to induce capital inflows in the short run and restore the trade balance in the medium run (import compression). Based on historical events, the simulations assume that the domestic interest rate increases by one percent for each reduction in official reserves representing one month of imports.
The model captures important general equilibrium effects that are not included in simpler reduced-form frameworks. As discussed above, the model features the monetary policy response to larger deficits, the reaction of capital flows, crowding-out effects of private spending, as well as the effects of public investment on potential output. Nonetheless, there are a number of caveats that should be considered when analyzing the model simulations. The model assumes that countries in the region do not face difficulties in getting access to external financing (either in the form of Eurobonds, external loans, or grants) at favorable rates. Rollover risk on external financing is not taken into account and it is assumed that access to external markets is maintained even with permanently higher deficits. Moreover, the estimates of effect of public investment on productivity are based on evidence from advanced economies. Finally, the import of content is assumed to be the same for consumption, private investment, and government investment.
30. The macroeconomic implications differ significantly across scenarios. The various scenarios can be compared depending on the macroeconomic variable of interest, namely: GDP growth, the debt to GDP ratio, the trade balance to GDP ratio, or reserve coverage (in months of imports):
Growth. The comparison of scenarios shows that, except when the deficit is financed externally and used to finance capital expenditure, the output effects are worse than under the baseline in the medium term. Under the first scenario as half of the additional public spending is devoted to public investment, prospects for potential output are enhanced, which generates a positive externality on private productivity and helps to attract foreign investment, allowing the central bank to build foreign reserves and ease domestic financial conditions. Overall, real GDP would increase by over 0.5 percent relative to the baseline over the near to medium-term, but long-term effects would be larger with GDP increasing by 0.7 percent by 2031 (Figure 3). In contrast, under the second scenario, assuming that the additional public spending is devoted to public consumption only, with no positive externalities on private productivity, the medium-term increase in GDP would be negligeable. Under the scenarios 3 and 4 that rely on domestic financing, increases in the deficit fail to stimulate growth because of crowding-out and limited effects on potential output11 Deteriorated growth prospects reduce confidence and capital inflows, and the central bank tightens financial conditions to avoid depleting foreign reserves.12 Output declines relative to the baseline would be substantial in the near-term reaching close of 1.5 percent in 2022 and 2023 in the scenario 4 that includes SFAs (Figure 3).
Debt dynamics. Convergence to a 4 percent of GDP deficit level could lead to significant debt increase if it is financed domesticaiiy and there are SFAs. All scenarios would lead to higher debt levels as a share of GDP relative to the baseline given that they assume higher deficit levels and any positive effects on growth would be insufficient to fully offset the adverse effects on debt dynamics. The increase in debt levels would be particularly stark under the case where SFAs are included and the increase in the deficit is financed domestically. In scenario 4, debt would be about 8 percent of GDP higher than under the baseline by 2026 and 15 percent of GDP higher over the long run.
Trade balance. Higher fiscal deficits would modestly deteriorate the trade balance in the case where the deficit is financed externaiiy, but under the scenarios with domestic financing the trade balance would improve due to import compression. In scenarios 1 and 2, the trade balance would deteriorate as higher government expenditures and private investment generate larger imports (Figure 4).13 Under the domestically financed scenarios (3 and 4), the trade balance would improve due to the lasting reduction of domestic demand, as the fixed exchange rate regime prevents swift expenditure switching and monetary policy is tightened to protect reserves.
Foreign exchange reserves. Reserve coverage would fall significantly in the near term if the fiscal deficit is financed from the regional market. When the permanent increase in the deficit is financed externally (scenarios 1 and 2), reserve coverage improves relative to the baseline by about 0.8 months of imports, because of strong capital inflows. Over the long term, as the trade balance deteriorates, reserve accumulation moderates (Figure 4). In the case where the permanent increase in the deficit to 4 percent of GDP is financed in the regional market without SFA (scenario 3), there is a reduction in net capital inflows in part because of diminished confidence, and consequently reserves fall. To mitigate it, the central bank tightens domestic credit conditions, such that the reserve coverage reduction is limited to about 1.5 months of imports over 2022–2024. Over the medium, as the trade balance improves so does reserve coverage. By comparison, in the case where the deficit is permanently increased to 4 percent of GDP and SFAs are taken into account (scenario 4), similar mechanisms are at play and the deterioration in reserve coverage relative to the baseline is even larger reaching 2.5 months of imports by 2023–24 and possibly undermining external stability, as reserves would fall well below the 3-months of imports reserve adequacy benchmark.
WAEMU: Effects of Different Scenarios on Growth and Debt
Citation: IMF Staff Country Reports 2022, 068; 10.5089/9798400204579.002.A002
Source: IMF Staff estimates.WAEMU: Effects of Different Scenarios on the Trade Balance and Reserves
Citation: IMF Staff Country Reports 2022, 068; 10.5089/9798400204579.002.A002
Source: IMF Staff estimates.31. Overall, the various scenarios suggest that two conditions are key for an increase in the deficit to lead to favorable regional outcomes: (1) reliance on (and unconstrained access to) external financing and (2) sound public financial management to reduce the size of SFAs. External financing mitigates crowding-out effects and contributes to maintaining reserve coverage. Improvements in public financial management and transparency limit off-budget and below-the-line operations, ensuring that the de facto deficit ceiling is close to the “above the line” one and may also improve the quality of spending. Although the simulations show that these two conditions are the most essential, other conditions can also lead to more favorable outcomes. In particular, measures to ensure that the composition of additional spending is geared towards items that enhance potential output like investment would also create positive outcomes relative to the baseline. Table 5 presents an overview of the macroeconomic implications of the four different scenarios discussed above, as well as three additional scenarios, two of which will be presented in detail below as they usefully highlight the importance of the two conditions (reliance on external financing and sound PFM).14
WAEMU: Overview of Implications of Higher Deficit Ceiling
WAEMU: Overview of Implications of Higher Deficit Ceiling
| 4 percent of GDP deficit de facto | 5 percent of GDP deficit de facto (with SFAs of 1 percent of GDP) | ||||||
|---|---|---|---|---|---|---|---|
| Type of spending/financing | High quality/ ext. finance | Low quality/ ext. finance | Low quality/ dom. finance | High quality/ dom. finance | High quality/ ext. finance | Low quality/ ext. finance | Low quality/ dom. finance |
| Scenario | Scenario 1 | Scenario 2 | Scenario 3 | Scenario 5 (not shown) | Scenario 6 (not shown) | Scenario 7 (not shown) | Scenario 4 |
| Policy option | Options 1 or 2 | Options 1 or 2 | Options 1 or 2 | Options 1 or 2 | Option 2 exclusively | Option 2 exclusively | Option 2 exclusively |
| GDP Growth | |||||||
| Reserve Coverage | |||||||
| Debt | |||||||
WAEMU: Overview of Implications of Higher Deficit Ceiling
| 4 percent of GDP deficit de facto | 5 percent of GDP deficit de facto (with SFAs of 1 percent of GDP) | ||||||
|---|---|---|---|---|---|---|---|
| Type of spending/financing | High quality/ ext. finance | Low quality/ ext. finance | Low quality/ dom. finance | High quality/ dom. finance | High quality/ ext. finance | Low quality/ ext. finance | Low quality/ dom. finance |
| Scenario | Scenario 1 | Scenario 2 | Scenario 3 | Scenario 5 (not shown) | Scenario 6 (not shown) | Scenario 7 (not shown) | Scenario 4 |
| Policy option | Options 1 or 2 | Options 1 or 2 | Options 1 or 2 | Options 1 or 2 | Option 2 exclusively | Option 2 exclusively | Option 2 exclusively |
| GDP Growth | |||||||
| Reserve Coverage | |||||||
| Debt | |||||||
32. If these two conditions are not in place simultaneously, the higher deficit target of 4 percent of GDP could have adverse macroeconomic consequences. In this context, it is useful to consider two “uncooperative” equilibria. The first one considers that the possibility that the international community delivers on additional external assistance and more external financing is available, but authorities do not advance on PFM reforms and maintain low quality composition of expenditure. The second case considers that the authorities deliver on PFM reforms and high-quality spending, but access to external financing is constrained.
In the case where additional external finance is available, but there is no improvement in PFM (to eliminate SFAs) and spending remains of low-quality, public debt would be significantly higher with no visible long-term growth benefits. Such scenario would lead to modestly higher GDP in the medium-term (less than 1 percent increase relative to the baseline), but the longer terms effects would be small and negative. Crucially, debt levels would be 14 percent of GDP higher than under the baseline (Figure 5).
In the case where PFM reforms are implemented (eliminating SFAs) and the additional fiscal deficit is used for high-quality spending, but access to external financing is constrained, the simulations indicate adverse near-term effects on output, but positive effects over the longer term. This result is due to the effects of additional investment on potential GDP, while the near-term output losses reflect crowding-out of private spending. In the short-run, one would observe a decline in reserve coverage (by less than 1 month of imports) as the risk premium increases (even if the increase is more moderate than in other scenarios) and capital flows decrease, which is attenuated over the medium and long-terms (Figure 6).
WAEMU: Unconstrained External Finance, but No Improvements in PFM
Citation: IMF Staff Country Reports 2022, 068; 10.5089/9798400204579.002.A002
Source: IMF Staff estimates.Note: In the context of the model, “no improvement in PFM” means that SFAs amounts to 1 percent of GDP. “Low-quality spending” means that the additional deficit finances current expenditure.WAEMU: Improvements in PFM and High-Quality Spending, but Constrained External Financing
Citation: IMF Staff Country Reports 2022, 068; 10.5089/9798400204579.002.A002
Source: IMF Staff estimates.Note: In the context of the model, “improvements in PFM” means that SFAs are eliminated. “High-quality spending” means that the additional deficit finances capital expenditure.33. In practice, meeting these two conditions simultaneously may prove very challenging in the WAEMU. The ability to finance additional deficits externally and transparently is likely to be limited by the following considerations:
Prospects for increased external financing are uncertain at the current juncture. Taking a long-term perspective, the ratio of overseas development assistance received to gross national income for the typical WAEMU country has declined by over 7 percentage points since the peak of the mid-1990s and by about 3 percentage points from the average ratio of the 2000s. More recently, some factors are also making access to external financing more difficult. For instance, some donors face political pressures to concentrate resources on domestic priorities after the COVID-19 pandemic. Concerns about debt sustainability may also constrain the support provided by international and bilateral institutions. Although Benin, Cote d’Ivoire, and Senegal have successfully tapped into Eurobond markets in 2020 and 2021 at favorable rates, changes in risk appetite by international investors and monetary policy normalization in advanced economies could lead to an increase in spreads in the near future.
In addition, large improvements in PFM are likely to take time to materialize. A set of directives issued in 2009 provides guidance to member states on the reform agenda, but in reality, the implementation of these directives has been mixed, both across member states and reform areas (Nguyen-Duong and Selim, 2021). In particular, the adoption of laws to improve internal expenditure controls and transparency of accounts has lagged behind.
WAEMU: Median Net Overseas Development Assistance
(percent of GNI)
Citation: IMF Staff Country Reports 2022, 068; 10.5089/9798400204579.002.A002
Source: World Banks’ WDI database.Option 2: raising the fiscal deficit ceiling to 4 percent of GDP “below the line”
34. Under the second policy option, the deficit target would also increase to 4 percent of GDP, but it would be measured “below-the-line”. More precisely, the deficit would be measured as the difference between the change in government’s financial liabilities and its financial assets (instead of revenue minus expenditure). It would thus include all the operations that account for SFAs in debt dynamics (and are not captured by the above the line deficit concept currently used in the fiscal rules framework). This redefinition of the deficit ceiling would not make much difference in terms of debt dynamics relative to the recent past, if SFAs amount to 1 percent of GDP (the historical average). Nonetheless, it would still differ from current WEO projections, which assume insignificant SFAs over the medium-term.
35. Some simulations presented above can illustrate the economic implications of this policy option. The outcomes of Scenarios 1 to 3 and Scenario 5 in Table 5 are compatible with the second policy option and the ensuing conclusions would apply, in particular in what regards the importance of external financing and the quality of spending to ensure more favorable outcomes relative to the baseline. Especially, as illustrated in Scenarios 3 and 5, in the case where the financing of the additional deficit would come from domestic sources (rather than external financing), one would observe higher debt (7 percentage points of GDP by the end of the projection period), and lower reserve coverage relative to the baseline. These scenarios would also imply a looser fiscal stance relative to the baseline embedded in IMF staff’s projections, which assume limited SFAs.
36. Some advantages of moving to a “below the line” definition of deficit targets are not directly captured by the model. Such move might lead to better control and monitoring of governments’ borrowing needs and debt dynamics. Moreover, it could enhance the transparency of fiscal rules framework and reduce incentives to move spending off-budget. Nonetheless, such change in definition might also introduce some difficulties in terms of measuring the deficit, particularly as far as the change in financial assets is concerned.
D. Conclusions
37. The analysis presented in this paper indicates that the 70 percent of GDP debt ceiling seems adequate, striking the right balance between growth and fiscal sustainability considerations. Raising the regional debt ceiling beyond 70 percent of GDP could create significant fiscal risks and expose countries to possible adverse debt dynamics. But even if the regional debt ceiling is kept at 70 percent of GDP, there is still room for a majority of WAEMU countries to accumulate more debt and the analysis does not preclude the possibility of differentiating individual country ceilings within the aggregated regional ceiling. In addition, the paper presents a range of estimates for the ceiling that vary not only according to the methodology used to compute the debt anchor, but also given uncertainty about the value of key parameters (such as interest rates). Finally, the results are very sensitive to the prospects for revenue mobilization. If countries succeed in making significant progress in mobilizing revenue, this would increase their debt carrying capacity and allow them to sustain higher debt levels.
38. A higher fiscal deficit ceiling of 4 percent of GDP could undermine domestic and external stability, unless very specific conditions are met. Overall, the simulations suggest that two conditions are essential for an increase in the “above the line” deficit to lead to more favorable outcomes relative to the baseline: (1) unconstrained access to external financing, and (2) significant progress in public financial management to pare down stock-flow adjustments. In practice, meeting these two conditions simultaneously may prove very challenging in the WAEMU, as prospects for increased external financing are uncertain and large improvements in the composition and quality of spending are likely to take time to materialize. If a decision is nonetheless taken to raise the deficit ceiling, there might be some advantages in measuring it “below the line” as this could, in principle, mitigate some of the negative outcomes depicted in the simulations and might foster fiscal transparency.
39. On balance, the analysis does not support a significant revision in WAEMU fiscal rule ceilings given existing challenges in public financial management and revenue mobilization as well as the still limited size/depth of the regional bond market. Various approaches described in this chapter show that the scope to revisit the debt and deficit ceilings is limited at the moment. No recalibration of fiscal targets would bring unequivocally superior economic outcomes relative to the current situation. The simulations highlight the policy trade-offs and risks associated with higher deficits and debt.
Annex I. Financing Additional Expenditures through Domestic Revenue Mobilization in Lieu of Higher Borrowing
1. The simulations presented in the main text depict the effects of a permanent increase in expenditure, assuming that taxes remain broadly unchanged. As illustrated in the figures above, the spending increase would be associated with a widening of fiscal deficits either financed by external or domestic resources. Nonetheless, domestic revenue mobilization is still falling short in the WAEMU when compared to other EMDEs and additional efforts in the area are urgently needed. Therefore, an alternative option for funding additional expenditure could be to mobilize more revenues rather than resorting to further debt accumulation.
2. This annex considers the macroeconomic effects of an increase of 1 percent of GDP in expenditures financed by an increase in consumption taxes (VAT). In this Annex, the fiscal deficit remains unchanged compared to the baseline. Expenditure is financed by higher consumption tax, which is assumed to have no incidence on investment. Two scenarios are considered regarding the additional spending financed by the tax increase: (1) in the first scenario, it is assumed that the increase in expenditures is concentrated in low quality spending, which means that the additional 1 percent of GDP is not spent on items that boost potential output; (2) in the second scenario, it is assumed that half of the additional spending will be devoted to items that increase potential output. A VAT-like tax is an efficient instrument to mobilize domestic revenue in the sense that it is less distortionary than other taxes, nevertheless its effects on growth cannot be analyzed without taking into account the use of the additional revenue.
3. A tax-financed increase in low quality expenditures could have a negative effect on GDP over the long run due to its distortionary effects on economic behavior, but the external position would modestly strengthen (Annex Figure). The dotted lines in the Annex Figure depict the variables for Scenario 3 in the main text for comparison purposes. In the short term the effects of the tax increase on output are muted because additional public expenditures broadly cancel out the reduction in private consumption caused by the tax increase. Nevertheless, the higher consumption tax also reduces labor supply in the long run with negative consequences on GDP.1 In fact, by 2031, GDP would be 0.6 percent lower than under the baseline. Given that the fiscal deficit does not widen relative to the baseline, the increase in the debt to GDP ratio is negligeable. The trade balance would improve in the medium run because of lower domestic demand. Import compression would cause an increase in the trade balance by 0.4 percentage points of GDP relative to the baseline by 2026, before gradually falling over the long run. In the short run, reserves would not be much affected by the tax increase, but reserve coverage would gradually increase by one month of imports over the long run.
4. A tax-financed increase in expenditures that boost potential GDP would lead to some gains in output over the long-run as well as a stronger external position relative to the baseline. Economic activity is slightly higher in the short run as private investment is stimulated because, in this Scenario, the additional public expenditure increases potential output (in the model private investment depends on the gap between current and desired capital stock, which is a function of financial conditions and potential output). Subsequently, as public and private capital stock increase, higher real wages compensate partially the negative effect of taxes on labor supply (which was discussed previously). As a result, GDP is somewhat higher in the long run whereas improvements in the trade balance and official reserves are more gradual.
5. Overall, even if a tax-financed increase in expenditures may, in some cases, lead to adverse effects relative to the baseline, it dominates the scenarios based on domestic financing of expenditure. The consumption tax plays the role of “compulsory saving” and therefore reduces the effect of additional public expenditures on the domestic interest rate compared to the scenarios that rely on domestic financing. Moreover, the negative effects of the tax increase on GDP could be reversed if the increase in expenditure is devoted to measures that enhance potential output rather than low-quality spending.
WAEMU: Simulation of a Tax-Financed Increase in Expenditures of 1 Percent of GDP
Citation: IMF Staff Country Reports 2022, 068; 10.5089/9798400204579.002.A002
Source: IMF Staff Estimates. The dotted lines refer to simulation results for Scenario 3 described in the main text.References
Andrle, M. P. Blagrave, P. Espillat, K. Honjo, B. Hunt, M. Kortelainen, R. Lalonde, D. Laxton, E. Mavroeidi, D. Muir, S. Mursula, and S. Snudden, 2015. “The Flexible System of Global Models FSGM”, IMF Working Paper 15/64. Washington, DC.
Barhoumi, K., and Towfighian, S., N., 2018. “Public Investment Efficiency in the WAEMU: An Empirical Assessment” West African Economic and Monetary Union Selected Issues Paper, IMF Country Report 18/107.
Bentum, Z., David, A. C., Slavov, S., and Sode, A., 2021 “A Simple Approach to Anchor Fiscal Policy in SSA Countries Post Covid” International Monetary Fund, mimeo.
Bova, E., Ruiz-Arranz, M., Toscani, F., and Ture, E., 2019. “The Impact of Contingent Liability Realizations on Public Finances” International Tax and Public Finance 26: 381–417.
Checherita-Westphal, C., A. H. Hallett, and P. Rother, 2014, “Fiscal Sustainability Using Growth-maximizing Debt Targets,” Applied Economics, 46: 638–647.
Eyraud, L., Baum, A., Hodge, A., Jarmuzek, M. Kim, Y., Mbaye, S., and Türe, E., 2018 “How to calibrate fiscal rules : a primer”. International Monetary Fund, Fiscal Affairs Department How to notes.
Feenstra, R. C., R. Inklaar, and M. P. Timmer, 2015, “The Next Generation of the Penn World Table,” available for download at www.ggdc.net/pwt
Ghosh, A., Jun I. Kim, Enrique G. Mendoza, Jonathan D. Ostry and Mahvash S. Qureshi, 2013. “Fiscal Space, Fiscal Fatigue, and Debt Sustainability in Advanced Economies” The Economic Journal 123: 4–30.
IMF, 2018. “Staff Guidance Note on Bank-Fund Debt Sustainability Framework for Low-Income Countries”. International Monetary Fund, Washington, DC.
IMF, 2021. “West African Economic and Monetary Union: Staff Report on Common Policies for Member Countries” IMF Country Report No. 21/49.
Martin, P., Pisani-Ferry, J., and Ragot, X., 2021. “Reforming the European Fiscal Framework” Les notes du conseil d’analyse économique, no 63, April.
Nguyen-Duong, A. and Selim, H., 2021. “Strengthening the Regional Fiscal Framework in the WAEMU” West African Economic and Monetary Union Selected Issues Paper, IMF Country Report 21/50.
Prepared by Antonio David (AFR), Benjamin Carton and Jared Bebee (all RES), and Giulio Lisi (SPR). We are extremely grateful for extensive guidance and comments by Annalisa Fedelino and Luc Eyraud as well as comments received from the WAEMU authorities.
As of 2020, all WAEMU countries had fiscal deficits above 3 percent of GDP. Only Guinea-Bissau’s debt exceeded the 70 percent of GDP ceiling.
WAEMU and ECOWAS processes do not necessarily need to be identical or aligned.
The estimation is based on data from the WEO database, covering the period 2000–20.
One shortcoming of using these data sources is that they do not necessarily reflect the rebasing of GDP that occurred in seven WAEMU countries since 2018, which may affect estimates of output, investment, and the capital stock. Nevertheless, long times series for the variables of interest constructed from the rebased GDP series are not available.
IMF staff estimates also indicate that in the absence of further mitigating policies, medium-term potential real GDP growth in the WAEMU could have been reduced by up to 1 percentage point due to the scarring effects of the pandemic (Selected Issues Paper).
SFAs may also be affected by data coverage issues (such as different perimeters in the coverage of deficits and debt statistics) and therefore relate to statistical compilation practices in the member countries.
Accounting issues related to the timing of disbursements of investment projects or in the case of Cote d’Ivoire, the use of cash accounting for the recording of debt, while the recording of deficits is done on an accrual basis are estimated to have amounted to 0.8 percent of the regional GDP in 2020. Moreover, below the line operations such as credit lines provided by governments are estimated to have reached about 0.3 percent of GDP in 2020. Other identified financial operations and accumulation of financial assets are estimated to amount to 0.6 percent of GDP.
This type of rule differs from the debt accumulation rule discussed in paragraph 22 of Nguyen-Duong and Selim (2021). The debt accumulation rule, popularized in the European framework as the 1/20th debt benchmark, is a rule that targets a certain annual reduction in the debt ratio relative to a target; complying with this rule can be difficult because (1) the debt ratio is highly sensitive to macroeconomic shocks (including GDP growth, inflation, and exchange rate), (2) the corresponding deficit targets change every year, and (3) this rule creates some confusion between the concepts of operational rule and anchor. By comparison, the fiscal deficit rule measured “below the line” is a more straightforward rule that targets a fixed fiscal deficit ratio that captures all changes in financial liabilities.
The scenarios focus on the macroeconomic implications of an increase in the deficit mainly translated as higher spending without significant changes relative to the baseline in terms of additional domestic revenue mobilization. The annex considers a set of simulations that include changes in taxes and their implications.
As financing of the deficits takes place domestically, banks direct domestic savings toward public bonds and deprive firms of funds for investment. Consequently, private investment declines.
The ensuing increase in interest rates attracts foreign capital and contracts private sector demand, including imports.
Nonetheless, when the increase in the deficit is devoted entirely to consumption under scenario 2, the trade balance deterioration in the short run is smaller because of the smaller effects of the higher deficit on private investment (in the absence of the positive externality of additional public investment on potential growth). In the long run, the tradable sector may shrink as factors of production move to the non-tradable sector because of Dutch disease (inflation is higher leading to a deterioration in competitiveness), and the trade deficit may increase again.
The scenario that is not presented in detail in the text is Scenario 6 of Table 5 that encompasses an increase in the deficit to 5 percent of GDP because of SFAs, but where external financing is readily available to cover the additional financing needs and most of the additional public spending is devoted to high quality expenditures that boost potential GDP. The figures related to this Scenario are not included to facilitate exposition of the arguments in the rest of the paper and because this Scenario is not considered to be the most likely. As indicated in the table, while growth and reserve coverage would be higher than under the baseline in this Scenario (because of the effect of public investment on potential output and because of additional external financing), the debt to GDP ratio would increase by over 14 percent of GDP by 2031.
Labor supply is affected because the increase in the consumption tax reduces the benefit of earning (and spending) income.
