Journal Issue
Share
Article

Djibouti: Staff Report for the 2019 Article IV Consultation—Debt Sustainability Analysis

Author(s):
International Monetary Fund. Middle East and Central Asia Dept.
Published Date:
October 2019
Share
  • ShareShare
Show Summary Details

Public Debt Coverage

1. The coverage of public debt used for this DSA is public and publicly guaranteed (PPG) debt. Specifically, the debt stock covers central government external debt—which accounted for 30 percent of total public debt at end-2018, excluding the loans contracted for the railway and the water pipeline—as well as government guarantees on state-owned enterprises’ (SOEs) external debt. 1 SOEs’ non-guaranteed external debt is not consolidated at the central government level and therefore not included in this DSA—according to the authorities, it would in any case likely be marginal and limited to short-term supplier credits. Information on domestic debt remains partial. Domestic debt contracted by the central government appears to be covered by the authorities’ data (and hence included in this DSA) and remains small. SOEs domestic debt is not captured. As of December 2018, consolidated data from the monetary survey indicate gross domestic credit to public nonfinancial corporations amounted to some 3.1 percent of GDP. It is however unclear whether this is guaranteed by the central government and the debt service is unknown, as neither the Ministry of Finance nor the Ministry of Budget track these debts. This is also not captured in the debt statistics.

Subsectors of the public sectorSub-sectors covered
1Central governmentX
2State and local governmentX
3Other elements in the general government
4o/w: Social security fund
5o/w: Extra budgetary funds (EBFs)
6Guarantees (to other entities in the public and private sector, including to SOEs)X
7Central bank (borrowed on behalf of the government)
8Non-guaranteed SOE debt
1The country’s coverage of public debtThe central, state, and local governments, government-guaranteed debt
DefaultUsed for the analysisReasons for deviations from the default settings
2Other elements of the general government not captured in 1.0 percent of GDP0
3SoE’s debt (guaranteed and not guaranteed by the government) 1/2 percent of GDP2
4PPP35 percent of PPP stock15.43Based on the World Bank’s PPP database.
5Financial market (the default value of 5 percent of GDP is the minimum value)5 percent of GDP5
Total (2+3+4+5) (in percent of GDP)22.4

2. Data on stocks and flows of private external debt are not available —except for a loan contracted in 2017 by a company related to the Djibouti Free Trade Zone (FTZ) and expected to be guaranteed by the government when it is refinanced in 2019–20. This and the marginal size of central government domestic debt result in external and public debt to largely overlap in this DSA over the projection period. External debt is defined based on residency, and there is a material difference with the criterion of currency denomination, as domestic financial institutions can issue credits in foreign currency—mainly U.S. dollars—to local counterparts. There is, however, no domestic debt market open to foreign investors.

3. The debt management capacity could be strengthened in Djibouti, as noted in the last World Bank’s Debt Management Performance Assessment (DeMPA). Strengthening the coverage of public debt, including by recording and monitoring all SOE debt, is crucial for a better risk assessment. The authorities plan on establishing a debt committee in 2019 which should contribute to improving coordination between government entities that contract and manage debt and addressing shortcomings in SOEs’ debt oversight.

4. The contingent liability stress test accounts for risks arising from public and private partnerships (PPPs) and financial markets. The stress test involves a one-off increase in the debt-to-GDP ratio in the second year of projection. The shock is made of two components: (i) a starting value of 5 percent of GDP, representing the average cost to the government of a financial crisis in a low-income country; and (ii) an additional value to capture PPPs contingent liabilities. PPPs are limited to a few projects in Djibouti, mainly related to the ports and more recently to a submarine broadband cable. This has been incorporated in the DSA, raising its contribution to the total contingent liability tailored shock by about 15½ percent of GDP, which corresponds to 35 percent of the total PPP stock as reported in the World Bank’s Private Participation in Infrastructure Database.2

Background

A. Debt and Budgetary Developments

5. The rapid debt accumulation since 2013 reflects mainly loan disbursements to finance three large infrastructure projects. In 2013, the government contracted two large loans to finance the construction of the Addis Ababa–Djibouti railway and a water pipeline from Ethiopia in the amount of US$814 million.3 In 2016, another government-guaranteed loan of US$345 million was signed to finance the construction of a multipurpose port. These three loans were extended by the Export-Import Bank of China (Exim Bank of China). During 2014–18, related disbursements reached a cumulative US$1.1 billion (around 71 percent of total disbursements over the period). As a result, total PPG debt increased from 34 percent of GDP in 2013 to 72 percent of GDP at end-2018 and external debt increased to 76 percent of GDP (Table 1). The share of government-guaranteed SOE debt—including debt contracted for the railway and the water pipeline—rose from 30 percent to 65 percent of external debt over the same period. The share of central government domestic debt in total public debt has been declining steadily and is now marginal.

Table 1.Djibouti: External and Public and Publicly-Guaranteed Debt, 2013 and 2018 1/
20132018
Millions of US$Percent of GDPPercent of external debtMillions of US$Percent of GDPPercent of external debt
Total (External + Domestic)71435.02,2537 7.1
External debt69434.0100.02,2317 6.3100.0
Central government debt4822 3.669.51,3534 6.360.6
Multilateral3111 5.244.83431 1.715.4
IMF311 .54.5160 .50.7
IDA1346 .519.21535 .26.8
Other multilateral1467.221.11756 .07.9
Official bilateral1718.424.71,0093 4.545.2
Paris Club653.29.4471 .62.1
China---78526.835.2
Railway---46315.820.7
Water pipeline---32211.014.4
Other non-Paris Club1065.215.31776 .17.9
Publicly-guaranteed debt2121 0.430.56712 3.030.1
China251 .23.64031 3.818.1
Other creditors1879.227.02689 .212.0
Other external debt---1444 .96.5
Stock of external arrears0-0.0642 .22.8
Domestic central government debt201 .0220 .8
In DF billions41.040 .8
Sources: Djibouti authorities; IMF staff calculations.

This table presents the data for the debt covered in this DSA and described in paragraph 1.

Sources: Djibouti authorities; IMF staff calculations.

This table presents the data for the debt covered in this DSA and described in paragraph 1.

6. Disbursements under the three large loans described above increased the share of U.S. dollar-denominated debt in total external debt and reinforced exposure to global financing conditions. In 2014, Djibouti’s external debt was mainly denominated in Kuwaiti dinar (22 percent), U.S. dollar (20 percent), and euro (10 percent). The disbursements related to the three large projects have increased the dollar-denominated debt share to 60 percent in 2018, thereby lowering valuation risks associated with movements in U.S. dollar exchange rate.4 However, the share of external debt with variable interest rate terms has also increased, as interest rates on loans for the construction and electrification of the railway are tied to the LIBOR, and therefore sensitive to changes in U.S. monetary policy.

7. The average grant element on new external borrowing has shrunk over the past few years. The average grant element on disbursements is estimated to have reached 31½ percent at end-2018. It is expected to increase temporarily in the medium-term—peaking at 37 percent in 2021—owing mainly to significant concessional financing from IDA and the Saudi Fund for Development and as projects financed on non-concessional terms reach completion and related disbursements stop. The grant element of new borrowing is then projected to decrease to 31 percent and remain stable in the medium term.

8. There have been increasing pressures on central government spending over the past few years. Budget support grants declined from 2.9 to 0.9 percent of GDP during 2015–18. Reflecting large tax expenditure—special tax regimes and exemptions for free zones, military bases, new investments—tax revenues decreased from 14.2 to 13.3 percent of GDP over the same period. On the spending side, given that the central government has been servicing the loans contracted for the railway and the water pipeline projects, and since the interest rate on the railway loan is indexed to the LIBOR, the central government interest bill has picked up significantly over the past few years. Against this backdrop, domestically-financed capital expenditure has been cut by close to 4 percent of GDP.

9. The authorities have recently reached an understanding on the restructuring of a large loan. Delays in operationalizing the railway project have resulted in lower revenues than expected to service the related loan. The latter, which represents about 16 percent of GDP, is being serviced by the government and was due to start amortizing in 2020. Against this backdrop, the authorities conducted discussions with Exim Bank of China to restructure the loan. They indicated that a memorandum of understanding had been signed to extend the grace period (by 5 years) as well as the maturity of the loan (by 10 years) and to reduce the interest rate (to LIBOR + 210 bps, from LIBOR + 300 bps). The Djiboutian authorities also indicated that arrears on interest payments on this loan that had been accumulated during the restructuring discussion (1.2 percent of GDP) had been restructured—they have been integrated to the principal and are expected to be repaid over the extended amortization period. Pending administrative and legal processes, they expected the agreement to be finalized within the next few months. The new terms of the railway loan have reduced the PV of debt-to-GDP ratio by 4 percentage points and smoothed total debt service. The latter is now expected to peak at 4.5 percent of GDP in 2025 against 5.3 percent of GDP in 2022 before the restructuring.

10. The authorities indicated that, as of end-June 2019, external debt arrears on PPG external debt (1.6 percent of GDP) reflected delays in finalizing conversion and cancellation agreements and also included arrears of technical nature or reflecting diplomatic issues. According to the authorities, arrears had accumulated in recent years pending the formalization of conversion and cancelation agreements or due to ongoing discussions with creditors for debt conversion (0.8 percent of GDP). Arrears related to diplomatic disputes amounted to 0.5 percent of GDP. Finally, notwithstanding efforts to clear technical arrears in recent months, the authorities noted the latter amounted to 0.2 percent of GDP. Technical arrears reflect mainly limited capacity to manage treasury cash-flow pressures, and the authorities do not consider them to be indicative of a fundamental payment incapacity as these are generally cleared after a short period of time. IMF staff has recommended to quickly address capacity issue in this area, including through technical assistance on cash management. On this basis, and in accordance with paragraph 90 of the LIC DSF guidance note, external debt arrears excluding arrears related to technical or diplomatic issues stand at 0.9 percent of GDP and are therefore assessed to be below the de minimis threshold of 1 percent of GDP.

B. Macroeconomic Statistics, Outlook, and Risks

11. This DSA is based on newly revised national accounts and balance of payment statistics. The new national accounts reflect expanded information on the activity of the ports and key enterprises in the FTZs, allowing for a better reflection of the large value added generated by the trade and transport sectors. The new data point to an upward revision of nominal GDP by close to 38 percent compared to the data used by staff for the last DSA (2017). Consistent with the general trade system, the authorities have also taken into consideration the large trade flows that are channeled through Djibouti’s FTZs for compiling external trade statistics, national accounts and balance of payment statistics. Given the central role of the FTZs in Djibouti and the large size of re-exports, both imports and exports have been revised up significantly (3.5- and 6-fold, respectively). For the analysis in the present document, debt-burden and debt-servicing capacity indicators as well as international reserve adequacy are assessed based on ratios that exclude re-exports from total exports and imports of goods and services. Similarly, for the purpose of analyzing external sector developments, staff uses an underlying current account balance that adjusts for the large changes in imports and exports related to re-export activities (and hence for changes in FTZs inventories related to these activities), which have led to large swings in the actual current account balance on the historical period.

12. Baseline macroeconomic projections assume a significant reduction in debt-financed infrastructure projects. With core infrastructure in place, the baseline does not incorporate any additional mega-projects, implying a significant reduction in foreign-financed public capital outlays. It also envisages a broadly stable government balance excluding foreign-financed spending, as further erosion of the revenue base relative to GDP—reflecting the large tax expenditure related to fast-growing trade and transport sectors—is offset through continued efforts to re-prioritize spending. In light of this, the baseline envisages a primary fiscal deficit and an overall public sector primary deficit—i.e. including SOEs—of about ¼ and 1⅓ percent of GDP respectively on average over the medium-term. Nominal PPG debt is projected to decline gradually to about 60 percent of GDP by 2024.

13. Growth is projected to pick up modestly this year and remain robust over the medium term. Notwithstanding the reduction in public infrastructure investment, the rapid recovery in Ethiopian trade flows is expected to underpin a continued recovery this year. Overall, growth is projected to remain at about 6 percent over the medium term, driven by strong export growth—as the new transport and logistics infrastructure position the country well to leverage Ethiopia’s strong growth and expand transshipment—and a pickup in FDI. Higher export growth combined with lower investment-related imports would lead to an improvement in the underlying current account balance to about 1⅓ percent of GDP on average in the medium term.

14. In the long term (2025–39), the baseline projects sustained growth driven by improved net exports and FDI inflows. The baseline assumes a successful leverage of Djibouti’s position to serve as the main entry-point for goods into East Africa but also as a regional platform for logistics, banking and information technology services. The momentum generated by recent improvements in the business climate5 is assumed to generate long-term dividends translating into sustained FDI inflows. Consequently, growth is projected to hover at about 5 percent in the long-term. The primary fiscal balance and the overall public sector primary balance are assumed to improve to slightly below zero percent of GDP. Finally, the current account surplus is assumed to improve further and average 2½ percent of GDP in the long term.

Table 2.Djibouti: Evolution of Selected Macroeconomic Indicators, 2017–39(in percent of GDP, unless otherwise indicated)
2017–20182019–20242025–2039
Real GDP growth (percent change)
Current DSA5.36.05.3
Previous DSA, February 20176.86.36.0
Primary fiscal balance
Current DSA-2.5-0.3-0.1
Previous DSA, February 20171.4-0.2-1.6
Public sector primary balance 1/
Current DSA-6.6-1.40.5
Previous DSA, February 2017---
Current account balance 2/
Current DSA-0.51.32.6
Previous DSA, February 2017-24.6-16.2-12.2
External PPG debt (nominal)
Current DSA71.065.535.7
Previous DSA, February 201785.968.654.1
Source: IMF Staff.

Including SOEs.

Adjusted for re-export activities.

Source: IMF Staff.

Including SOEs.

Adjusted for re-export activities.

15. Debt service is set to increase over the medium term. Notwithstanding the agreement to lengthen the debt service profile of the railway loan, and with water pipeline loan set to start amortizing in 2022, debt service is projected to increase over the next few years. The baseline assumes that SOEs will generate the cash flow necessary to repay the debt contracted for the projects they manage.

16. The baseline is subject to significant uncertainty, with risks tilted to the downside:

  • On the domestic front, delays in implementing reforms to ensure debt sustainability—notably if the authorities engage in new debt-financed mega-projects or fail to raise returns on infrastructure and SOEs profitability—could worsen debt prospects, as there is limited space for consolidation at the level of the central government. A slowdown in structural reforms or slower progress in improving governance, the business climate, and external competitiveness, could reduce FDI and export growth, entail lower growth and weaken the international reserves coverage.
  • Externally, regional geopolitical tensions and a slowdown in international or Ethiopia’s trade would have adverse effects on Djibouti’s exports, investment, economic growth, and fiscal prospects. In this context, lower exports and a widening current account deficit would increase external financing needs. The cancellation of DP World contract could entail reputational risks (affecting investment prospects) as well as fiscal costs. A worsening of global financial conditions could increase interest costs and worsen external financing prospects. Renewed pressures on CBRs could create financing disruptions. Finally, higher global oil prices than in the baseline scenario could weigh on the budget and the external position.

17. Financing assumptions have been updated to integrate more granularity on financing terms. Debt service projections are based on the financing terms of the loans for which disbursements have started and loans for which borrowing agreements have already been signed. Financing terms are aggregated by creditors, allowing for more granularity in the assessment of concessionality. The discount factor remains fixed at 5 percent, as approved by the IMF and World Bank Executive Boards in October 2013. New disbursements for the central government and the SOEs through 2039 are broadly in line with past disbursements excluding the three recent mega projects—the Doraleh multipurpose port, the railway and the water pipeline—averaging about 3 percent of GDP annually. The terms of these new disbursements are also assumed to be broadly consistent with past loans, excluding the loans for the large infrastructure projects. As such, the average grant element of new borrowing is projected to stabilize at 31 percent in the medium term.

18. The realism tools added to the new LIC DSA framework signal risks around the baseline projections. The projected three-year primary balance adjustment stands at 3.5 percentage points of GDP. It lies within the top quartile of past adjustments of primary balances in LIC countries under Fund-supported programs. The realism tools flag optimism on GDP growth projections in the baseline given the magnitude of the fiscal adjustment. Past forecast errors have been very large (as illustrated by Figure 3), also entailing additional risks. However, baseline growth projections assume that, while the decrease in capital expenditure will weigh on growth, the underlying fiscal multiplier is low due to the substantial import content of the large projects. In addition, the development in trade and transport infrastructure in recent years positions the country well to leverage the rapid growth in Ethiopia’s trade. Combined with an improved business climate, this is expected to give a more prominent role to private investment, including FDI, and exports in driving growth over the medium term.

C. Country Classification and Stress Test Scenarios

19. Djibouti’s debt carrying capacity is classified as medium.6 The new debt sustainability framework encompasses a broader set of indicators to define debt carrying capacity and applicable thresholds. The underlying composite indicator accounts for the CPIA—as in previous DSA vintages—but also real GDP growth rate, import coverage of reserves–adjusted for reexports–remittances and world economic growth. Under the new framework, Djibouti’s debt carrying capacity is classified as medium with the largest contributions from import coverage of reserves and the CPIA to the composite indicator.

Calculation of the CI Index
ComponentsCoefficients (A)10-year average values (B)CI Score components (A*B) = (C)Contribution of components
CPIA0.3852.9881.1541%
Real growth rate (in percent)2.7196.2100.176%
Import coverage of reserves (in percent)4.05247.4691.9268%
Import coverage of reserves^2 (in percent)-3.99022.533-0.90-32%
Remittances (in percent)2.0220.0000.000%
World economic growth (in percent)13.5203.5790.4817%
CI Score2.83100%
CI ratingMedium
FinalClassification based on current vintageClassification based on the previous vintage
MediumMedium

2.83
Medium

2.74
External Debt burden Thresholds
PV of debt in % of
Exports180
GDP40
Debt service in % of
Export15
Revenue18
Public Debt Burden Thresholds
PV of total public debt in percent of GDP55

20. This DSA applies the standard stress tests, as well as additional tailored stress tests on contingent liabilities. The standard tests account for shocks on real GDP growth, the primary balance, exports and other external financing flows. A shock on the exchange rate, consisting of a one-time 30 percent nominal depreciation is also considered. The tailored stress test on contingent liabilities encompasses a shock from a financial market crisis and the materialization of a PPPs contingent liability as discussed above. The natural disaster, commodity price and market financing tailored stress tests do not apply to Djibouti.

Debt Sustainability

A. External Debt Sustainability

21. Based on the guidance note of the Bank-Fund debt sustainability framework for low-income countries (LIC DSF), the results of the external DSA indicate that Djibouti is at high risk of debt distress. This reflects the fact that the present value (PV) of the external debt-to-GDP ratio breaches its threshold during 2019–26 and the debt service-to-revenue ratio increases and stays above its threshold from 2022 onward.

22. The DSA also concludes that debt is sustainable as the authorities are expected to be able to service their debt under current terms. Although the debt-service-to-revenue liquidity ratio for the external debt stands above its threshold after 2022—reflecting an increase in debt service over the next few years (as amortizations on the two loans contracted for the railway and water pipeline projects start falling due) and suggesting that liquidity risk to the debt path could re-assert themselves in the coming years—the recent understanding on the restructuring of the railway loan has considerably eased the debt service constraint in the short term, giving more time for the project to generate the necessary resources. 7

23. The debt dynamics are sensitive to shocks. The bound tests indicate that the most extreme shock for the PV of external debt-to-GDP is the contingent liabilities which would raise the PV of the external debt-to-GDP ratio to 70 percent. The most extreme shock for the debt service-to-revenues is the primary balance. Finally, a shock to export revenues, which could stem from a slowdown in economic growth in Ethiopia or a slowdown in international trade, would be the most extreme for both the PV of debt-to-exports and the debt service-to-exports ratio, although in both cases the ratios would remain below their thresholds.

B. Public Debt Sustainability

24. The dynamics of total public debt reflect the large share of PPG external debt in total public debt. Under the baseline, reflecting the sharp rise in PPG external debt, overall public debt is projected to increase from 34 percent of GDP in 2013 to 72 percent at end-2018. The PV of debt-to-GDP stands below the 55 percent benchmark throughout the projection horizon. Simulations show that under the historical scenario, public debt dynamics worsen as the primary fiscal deficit is wider than in the baseline. According to stress tests, solvency indicators are most vulnerable to combined contingent liabilities, while the liquidity indicator is most vulnerable to a one-time depreciation shock. Given the Djibouti franc’s exceptional stability vis-à-vis the dollar under the currency board, the DSA exchange rate stress scenario should be considered a tail risk event. While a bilateral movement of the U.S. dollar exchange rate vis-à-vis other major currencies could contribute to some adverse valuation effects, this should be limited given the currency board linked to the U.S. dollar and the recent increase in the share of U.S. dollar-denominated debt. A tightening of global financing conditions could, however, pose important risks, as a sizeable share of external debt is linked to the LIBOR—particularly the debt related to the railway project.

C. Risk Rating and Vulnerabilities

25. Overall, this debt sustainability analysis indicates that Djibouti is at high risk of external and overall debt distress. This rating is unchanged compared to the 2016 DSA. It reflects the fact that a memorandum of understanding on the restructuring of the railway loan has been signed, and external arrears have been cleared (except for de minimis arrears). The restructuring has eased potential liquidity pressures over the next few years. However, under the baseline scenario, the PV of external debt-to-GDP ratio and the debt service-to-revenue ratio stand above their respective thresholds for prolonged periods of time. The DSA also shows that Djibouti’s debt dynamics is particularly vulnerable to adverse contingent liabilities and overall public sector primary balance shocks.

26. Debt is considered sustainable, as the authorities are expected to be able to service their debt under current terms, but a multifaceted policy and reform approach is also important to further support debt sustainability. The authorities should implement a combination of policies aimed at reducing the scope for new borrowing, especially on non-concessional terms. Policies should include strict limits on guarantees and on-lending to contain contingent liabilities. These limits should be embedded in an explicit debt strategy that lays out high-level policy priorities and anchor macroeconomic management. The introduction of a ceiling on PPG debt would also help. Adopting a medium-term fiscal framework would support the authorities in targeting medium-term objectives that are consistent with debt sustainability. Reforms to reduce tax expenditure and enhance spending prioritization to generate fiscal space for poverty-reduction spending and debt reduction are also critical. Improving public investment management and efforts to raise returns on existing and new investments are also important to allow for additional resources to repay the debt. Strengthening the governance and oversight of state-owned enterprises is key both to control the accumulation of debt and improve the return on investments. In this context, the authorities should adopt a framework to identify, monitor and manage fiscal risks from contingent liabilities of SOEs. They should also prioritize strengthening debt and cash management to bring Djibouti’s debt back to a sustainable trajectory and avoid the incurrence of arrears. Finally, accelerating reforms to improve the business climate and foster high private-sector led growth can contribute to re-establish debt sustainability.

D. Authorities’ Views

27. The authorities broadly agreed with staff assessment. They recognized the importance of reducing debt vulnerabilities and were broadly in agreement with staff’s recommendations. The authorities expressed confidence about their capacity to service the debt contracted for various projects by mobilizing resources from the SOEs managing them. They highlighted that the restructuring of the railway project had contributed to ease pressures on the budget and were confident that the project would be able to rapidly generate the revenues necessary to service the related debt. They did not see the debt ceiling as an immediate priority, given the importance of strengthening inter-agency coordination first. While acknowledging that reducing debt and debt service is important, the authorities also felt that the DSA thresholds were overly conservative for a small economy with large investment needs, many of which relate to projects aimed at fostering regional integration, with large potential economic returns.

Table 3.Djibouti: External Debt Sustainability Framework, Baseline Scenario, 2016–2039(In percent of GDP, unless otherwise indicated)
Sources: Country authorities; and staff estimates and projections.1/ Includes both public and private sector external debt.2/ Derived as [r – g – ρ(1+g) + εα (1 + r)]/(1+g+ρ+gρ) times previous period debt ratio, with r = nominal interest rate; g = real GDP growth rate, ρ = growth rate of GDP deflator in U.S. dollar terms, Ɛ=nominal appreciation of the local currency, and α= share of local currency-denominated external debt in total external debt.3/ Includes exceptional financing (i.e., changes in arrears and debt relief); changes in gross foreign assets; and valuation adjustments. For projections it also includes contribution from price and exchange rate changes.4/ Residual debt-creating flows and large negative gross external financing need arise because of the large and systematically negative errors and omissions in the historical data of the balance of payment. Both the residual and gross external financing need remain substantial in projections reflecting the “Other investment” item of the balance of payment. The latter helps address the issue of error and omissions in the projection period, but is not accounted for in this table.5/ Current-year interest payments divided by previous period debt stock6/ Defined as grants, concessional loans, and debt relief.7/ Grant-equivalent financing includes grants provided directly to the government and through new borrowing (difference between the face value and the PV of new debt).8/ Assumes that PV of private sector debt is equivalent to its face value.9/ Historical averages are generally derived over the past 10 years, subject to data availability, whereas projections averages are over the first year of projection and the next 10 years.
Sources: Country authorities; and staff estimates and projections.1/ Includes both public and private sector external debt.2/ Derived as [r – g – ρ(1+g) + εα (1 + r)]/(1+g+ρ+gρ) times previous period debt ratio, with r = nominal interest rate; g = real GDP growth rate, ρ = growth rate of GDP deflator in U.S. dollar terms, Ɛ=nominal appreciation of the local currency, and α= share of local currency-denominated external debt in total external debt.3/ Includes exceptional financing (i.e., changes in arrears and debt relief); changes in gross foreign assets; and valuation adjustments. For projections it also includes contribution from price and exchange rate changes.4/ Residual debt-creating flows and large negative gross external financing need arise because of the large and systematically negative errors and omissions in the historical data of the balance of payment. Both the residual and gross external financing need remain substantial in projections reflecting the “Other investment” item of the balance of payment. The latter helps address the issue of error and omissions in the projection period, but is not accounted for in this table.5/ Current-year interest payments divided by previous period debt stock6/ Defined as grants, concessional loans, and debt relief.7/ Grant-equivalent financing includes grants provided directly to the government and through new borrowing (difference between the face value and the PV of new debt).8/ Assumes that PV of private sector debt is equivalent to its face value.9/ Historical averages are generally derived over the past 10 years, subject to data availability, whereas projections averages are over the first year of projection and the next 10 years.
Table 4.Djibouti: Public Sector Debt Sustainability Framework, Baseline Scenario, 2016–2039(In percent of GDP, unless otherwise indicated)
Sources: Country authorities; and staff estimates and projections.1/ Coverage of debt: The central, state, and local governments, government-guaranteed debt. Definition of external debt is Residency-based.2/ The underlying PV of external debt-to-GDP ratio under the public DSA differs from the external DSA with the size of differences depending on exchange rates projections.3/ Debt service is defined as the sum of interest and amortization of medium and long-term, and short-term debt.4/ Gross financing need is defined as the primary deficit plus debt service plus the stock of short-term debt at the end of the last period and other debt creating/reducing flows.5/ Defined as a primary deficit minus a change in the public debt-to-GDP ratio ((-): a primary surplus), which would stabilizes the debt ratio only in the year in question.6/ Historical averages are generally derived over the past 10 years, subject to data availability, whereas projections averages are over the first year of projection and the next 10 years.
Sources: Country authorities; and staff estimates and projections.1/ Coverage of debt: The central, state, and local governments, government-guaranteed debt. Definition of external debt is Residency-based.2/ The underlying PV of external debt-to-GDP ratio under the public DSA differs from the external DSA with the size of differences depending on exchange rates projections.3/ Debt service is defined as the sum of interest and amortization of medium and long-term, and short-term debt.4/ Gross financing need is defined as the primary deficit plus debt service plus the stock of short-term debt at the end of the last period and other debt creating/reducing flows.5/ Defined as a primary deficit minus a change in the public debt-to-GDP ratio ((-): a primary surplus), which would stabilizes the debt ratio only in the year in question.6/ Historical averages are generally derived over the past 10 years, subject to data availability, whereas projections averages are over the first year of projection and the next 10 years.

Figure 1.Djibouti: Indicators of Public and Publicly Guaranteed External Debt Under Alternatives Scenarios, 2019–2029

Sources: Country authorities; and staff estimates and projections.

1/ The most extreme stress test is the test that yields the highest ratio in or before 2029. Stress tests with one-off breaches are also presented (if any), while these one-off breaches are deemed away for mechanical signals. When a stress test with a one-off breach happens to be the most exterme shock even after disregarding the one-off breach, only that stress test (with a one-off breach) would be presented.

2/ The magnitude of shocks used for the commodity price shock stress test are based on the commodity prices outlook prepared by the IMF research department.

Figure 2.Djibouti: Indicators of Public Debt Under Alternative Scenarios, 2019–2029

* Note: The public DSA allows for domestic financing to cover the additional financing needs generated by the shocks under the stress tests in the public DSA. Default terms of marginal debt are based on baseline 10-year projections.

Sources: Country authorities; and staff estimates and projections.

1/ The most extreme stress test is the test that yields the highest ratio in or before 2029. The stress test with a one-off breach is also presented (if any), while the one-off breach is deemed away for mechanical signals. When a stress test with a one-off breach happens to be the most exterme shock even after disregarding the one-off breach, only that stress test (with a one-off breach) would be presented.

Table 5.Djibouti: Sensitivity Analysis for Key Indicators of Public and Publicly Guaranteed External Debt, 2019–2029(In percent)
Projections
20192020202120222023202420252026202720282029
PV of debt-to GDP ratio
Baseline52.853.153.451.549.246.543.540.437.735.032.5
A. Alternative Scenarios
A1. Key variables at their historical averages in 2019–2039 1/52.851.751.248.846.644.241.137.934.931.828.8
B. Bound Tests
B1. Real GDP growth52.855.057.455.352.850.046.743.440.537.634.9
B2. Primary balance52.861.068.666.263.259.956.352.649.246.143.1
B3. Exports52.855.761.058.856.152.949.345.742.539.436.5
B4. Other flows 2/52.857.762.460.157.354.050.346.643.340.137.1
B6. One-time 30 percent nominal depreciation52.866.864.562.159.356.252.648.945.642.539.4
B6. Combination of B1-B552.861.564.762.359.456.052.248.545.141.838.7
C. Tailored Tests
C1. Combined contingent liabilities52.870.069.767.264.361.157.654.050.847.744.8
C2. Natural disastern.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
C3. Commodity pricen.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
C4. Market Financingn.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
Threshold4040404040404040404040
PV of debt-to-exports ratio
Baseline113.3115.1117.1115.8113.5109.9102.795.589.082.776.7
A. Alternative Scenarios
A1. Key variables at their historical averages in 2019–2039 1/113.3112.3112.3109.8107.5104.397.089.482.475.268.1
B. Bound Tests
B1. Real GDP growth113.3115.1117.1115.8113.5109.9102.795.589.082.776.7
B2. Primary balance113.3132.3150.5148.8146.0141.6132.9124.2116.3108.8101.7
B3. Exports113.3129.9154.7152.9149.8144.5134.7124.9116.1107.799.7
B4. Other flows 2/113.3125.1136.8135.2132.3127.4118.7110.0102.294.887.7
B6. One-time 30 percent nominal depreciation113.3115.1112.3111.0108.8105.498.691.885.679.673.9
B6. Combination of B1-B5113.3130.8129.8141.2138.1133.3124.3115.3107.299.592.1
C. Tailored Tests
C1. Combined contingent liabilities113.3151.8152.9151.1148.5144.4136.0127.6119.9112.7105.8
C2. Natural disastern.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
C3. Commodity pricen.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
C4. Market Financingn.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
Threshold180180180180180180180180180180180
Debt service-to-exports ratio
Baseline6.27.97.59.08.99.39.910.19.69.19.0
A. Alternative Scenarios
A1. Key variables at their historical averages in 2019–2039 1/6.27.97.48.98.69.09.59.89.28.78.5
B. Bound Tests
B1. Real GDP growth6.27.97.59.08.99.39.910.19.69.19.0
B2. Primary balance6.27.98.010.010.711.912.312.411.811.311.1
B3. Exports6.28.58.911.011.212.312.813.012.311.611.4
B4. Other flows 2/6.27.97.89.69.910.811.211.410.810.210.0
B6. One-time 30 percent nominal depreciation6.27.97.58.88.79.09.59.89.38.88.7
B6. Combination of B1-B56.28.38.510.310.711.311.812.011.410.810.6
C. Tailored Tests
C1. Combined contingent liabilities6.27.98.510.09.910.310.811.010.49.99.7
C2. Natural disastern.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
C3. Commodity pricen.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
C4. Market Financingn.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
Threshold1515151515151515151515
Debt service-to-revenue ratio
Baseline13.116.616.218.818.719.520.921.720.819.719.5
A. Alternative Scenarios
A1. Key variables at their historical averages in 2019–2039 1/13.116.716.118.618.218.820.121.020.118.918.4
B. Bound Tests
B1. Real GDP growth13.117.317.420.220.121.022.523.322.421.220.9
B2. Primary balance13.116.617.320.822.524.926.026.725.724.524.1
B3. Exports13.116.716.619.920.422.223.424.123.121.921.5
B4. Other flows 2/13.116.616.820.020.922.723.924.523.422.221.8
B6. One-time 30 percent nominal depreciation13.121.020.423.323.223.625.426.525.424.123.8
B6. Combination of B1-B513.117.818.421.422.423.524.925.724.623.322.9
C. Tailored Tests
C1. Combined contingent liabilities13.116.618.521.020.921.622.923.622.621.521.1
C2. Natural disastern.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
C3. Commodity pricen.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
C4. Market Financingn.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
Threshold1818181818181818181818
Sources: Country authorities; and staff estimates and projections.

Variables include real GDP growth, GDP deflator (in U.S. dollar terms), non-interest current account in percent of GDP, and non-debt creating flows.

Includes official and private transfers and FDI.

Sources: Country authorities; and staff estimates and projections.

Variables include real GDP growth, GDP deflator (in U.S. dollar terms), non-interest current account in percent of GDP, and non-debt creating flows.

Includes official and private transfers and FDI.

Table 6.Djibouti: Sensitivity Analysis for Key Indicators of Public Debt, 2019–2029(In percent)
Projections
20192020202120222023202420252026202720282029
PV of Debt-to-GDP Ratio
Baseline53.353.353.451.549.246.543.540.437.735.032.5
A. Alternative Scenarios
A1. Key variables at their historical averages in 2019–2039 1/53.356.559.560.661.762.763.764.565.366.166.8
B. Bound Tests
B1. Real GDP growth53.355.859.258.457.055.252.950.448.246.244.2
B2. Primary balance53.361.268.766.363.560.256.552.849.546.343.3
B3. Exports53.355.860.758.555.852.649.045.542.239.236.3
B4. Other flows 2/53.357.962.460.157.354.050.346.643.340.137.1
B6. One-time 30 percent nominal depreciation53.368.366.562.057.553.048.243.539.235.231.4
B6. Combination of B1-B553.359.059.550.448.245.742.739.737.034.431.9
C. Tailored Tests
C1. Combined contingent liabilities53.370.269.867.364.561.457.954.351.048.045.0
C2. Natural disastern.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
C3. Commodity pricen.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
C4. Market Financingn.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
Public debt benchmark55.055.055.055.055.055.055.055.055.055.055.0
PV of Debt-to-Revenue Ratio
Baseline208.0221.5231.1221.9220.0212.5201.2189.3178.3165.7153.7
A. Alternative Scenarios
A1. Key variables at their historical averages in 2019–2039 1/208.0234.8257.3260.9275.7285.9294.2301.7308.7312.4315.8
B. Bound Tests
B1. Real GDP growth208.0231.3254.6250.1253.5250.6243.1234.6226.9217.1207.8
B2. Primary balance208.0254.4297.3285.9284.0274.9261.6247.5234.3219.2204.8
B3. Exports208.0231.9262.4252.0249.6240.1226.8212.8200.0185.4171.6
B4. Other flows 2/208.0240.7269.9259.1256.3246.4232.5218.2204.9189.9175.7
B6. One-time 30 percent nominal depreciation208.0286.4289.6269.0259.1243.4224.5205.0186.8167.7149.5
B6. Combination of B1-B5208.0246.1257.9217.7216.2209.0197.9186.2175.3162.9151.0
C. Tailored Tests
C1. Combined contingent liabilities208.0291.8301.9290.3288.8280.4267.7254.2241.6227.0213.1
C2. Natural disastern.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
C3. Commodity pricen.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
C4. Market Financingn.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
Debt Service-to-Revenue Ratio
Baseline12.316.015.417.317.318.119.320.019.218.217.9
A. Alternative Scenarios
A1. Key variables at their historical averages in 2019–2039 1/12.316.115.918.119.221.023.425.425.926.227.3
B. Bound Tests
B1. Real GDP growth12.316.616.518.718.920.221.923.022.421.721.7
B2. Primary balance12.316.016.419.220.723.024.024.623.722.622.2
B3. Exports12.316.015.718.218.720.421.522.121.120.019.7
B4. Other flows 2/12.316.016.018.419.320.922.022.621.620.520.1
B6. One-time 30 percent nominal depreciation12.317.920.022.622.423.324.925.924.723.423.1
B6. Combination of B1-B512.315.615.217.017.017.819.019.718.917.917.7
C. Tailored Tests
C1. Combined contingent liabilities12.316.017.519.319.220.021.121.820.819.819.4
C2. Natural disastern.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
C3. Commodity pricen.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
C4. Market Financingn.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.
Sources: Country authorities; and staff estimates and projections.

Variables include real GDP growth, GDP deflator and primary deficit in percent of GDP.

Includes official and private transfers and FDI.

Sources: Country authorities; and staff estimates and projections.

Variables include real GDP growth, GDP deflator and primary deficit in percent of GDP.

Includes official and private transfers and FDI.

Figure 3.Djibouti: Drivers of Debt Dynamics – Baseline Scenario

1/ Difference between anticipated and actual contributions on debt ratios.

2/ Distribution across LICs for which LIC DSAs were produced.

3/ Given the relatively low private external debt for average low-income countries, a ppt change in PPG external debt should be largely explained by the drivers of the external debt dynamics equation.

Figure 4.Djibouti: Realism Tools

1The government guarantees on SOE external debt stem from loans contracted by the government that have been on-lent (under loan agreements, accords de rétrocession) to public enterprises. Under the lending agreement, the public enterprise assumes responsibility to pay the debt service falling due on the loan. But this debt is treated in this DSA as being guaranteed by the government because the latter remains the borrower of record and would be the payer of last resort to the creditor in the event of a default by the public enterprise.
2The World Bank’s Private Participation in Infrastructure Database is available at: http://ppi.worldbank.org
3These loans are still to be on-lent under the mechanism described in footnote 1.
4Under the currency board arrangement, the Djibouti franc is pegged to the U.S. dollar.
5Djibouti is ranked 99th in the 2019 World Bank’s Doing Business from 154th in 2018 and 171th in 2017.
6The composite indicator is calculated based on the data published in the World Economic Outlook from April 2019.
7Consistent with paragraph 22 of the Guidance Note on the Bank-Fund LIC DSA framework, the debt-service-to-revenue liquidity ratio is calculated as the ratio of total public sector interests and amortizations to central government revenues augmented by the net income that SOEs can use for debt service. Given that data limitations do not allow for a detailed cash flow analysis of SOEs, the net income that SOEs can use for debt service is assumed to be equal to the total debt service due by SOEs. This assumption may entail an under-estimation of SOEs’ capacity to service debt, to the extent that they generate higher cash flows that could be used for debt service. However, while data on gross operating balances (a concept somewhat equivalent to the EBITDA) are available for most large SOEs, using those to estimate net income that SOEs can use for debt service would lead to a considerable over-estimation of SOEs’ capacity to service debt. This is because these data do not account for the cash flows used by SOEs to self-finance large investment programs. In addition, aggregating these surpluses would entail considering that revenue from SOEs are fungible—i.e. that these are resources that can be used to service any debt—which is not the case.

Other Resources Citing This Publication