Report Number 78485-NA REPUBLIC OF NAMIBIA Analysis and Options for Namibia’s Medium-Term Debt Strategy June 2013 Poverty Reduction Economic Management 1 Southern Africa Africa Region The World Bank The World Bank MEFMI . CURRENCY EQUIVALENTS (as of June 14, 2013) 9.8803 Namibian dollars = 1 U.S. dollar GRN FISCAL YEAR (April 1-March 31) ABBREVIATIONS AND ACRONYMS AfDB African Development Bank JIBAR Johannesburg Interbank Average Rate AIDS Acquired Immune Deficiency Syn- JSE Johannesburg Stock Exchange drome Libor London inter-bank offer rate BON Bank of Namibia MEFMI Macroeconomic and Financial Man- CDM Cash and Debt Management Division agement Institute CMA Common Monetary Area MOF Ministry of Finance CSD Central securities depository MTDS Medium-term debt strategy CS-DRMS Commonwealth Secretariat Debt Re- MTEF Medium-term expenditure framework cording and Management System NAD Namibian dollar DMO Debt management office NDP4 Fourth National Development Plan EUR European euro NPC National Planning Commission FX Foreign exchange PPP Public-private partnership FY Fiscal year SACU Southern African Customs Union GDP Gross domestic product SDMS Sovereign Debt Management Strategy GIPF Government Institutions Pension SOE State-owned enterprise Fund T-bills Treasury bill GRN Government of the Republic of Na- TIPEEG Targeted Intervention Program for mibia Employment and Economic Growth HIV Human Immuno-deficiency Virus USD United States dollar IMF International Monetary Fund ZAR South African rand Vice President: Makhtar Diop Country Director: Asad Alam Sector Manager: John Panzer Task Team Leader: Philip Schuler i Tables of Contents I. Introduction ................................................................................................................... 1 A. Updating Namibia’s Debt Management Strategy ..................................................... 1 B. Objective and Scope of the Report .......................................................................... 2 II. Current Debt Management Strategy and Debt Portfolio ................................................. 2 A. Current Debt Management Strategy ........................................................................ 2 B. Namibia’s Existing Public Debt Portfolio ................................................................ 4 C. Cost and Risk Characteristics of Existing Debt ........................................................ 6 III. Sources of Financing ..................................................................................................... 7 A. External Sources of Financing ................................................................................. 7 B. Domestic Sources of Financing ............................................................................... 8 IV. Baseline Macroeconomic Assumptions and Key Risk Factors ....................................... 8 A. Baseline Macroeconomic Assumptions ................................................................... 8 B. Key Vulnerabilities ............................................................................................... 10 C. Debt Sustainability ................................................................................................ 12 V. Cost-Risk Analysis of Alternative Debt Management Strategies .................................. 13 A. Baseline Interest and Exchange Rate Assumptions ................................................ 13 B. Description of Shock Scenarios ............................................................................. 16 C. Description of Alternative Debt Management Strategies ....................................... 16 D. Cost-Risk Analysis of Alternative Debt Management Strategies............................ 18 VI. Contingent Liabilities Stemming from Government Guarantees .................................. 22 A. Overall Governance Framework............................................................................ 23 B. Issuance rules and procedures by the MOF............................................................ 23 C. Risk Management ................................................................................................. 24 D. Collection of Receivables ...................................................................................... 25 E. Transparency and Accountability .......................................................................... 25 F. Public Private Partnerships ....................................................................................... 26 VII. Domestic Debt Market Development ........................................................................... 26 VIII.Legal, Institutional and Implementation Issues ............................................................ 30 A. Legal Framework .................................................................................................. 30 B. Institutional Arrangements .................................................................................... 30 C. Annual Borrowing Plan......................................................................................... 30 D. Implementing, Disseminating, Monitoring and Reviewing the Debt Strategy ........ 31 IX. Conclusions ................................................................................................................. 32 X. Annexes ...................................................................................................................... 34 Annex 1: Description of the MDTS Analytical Tool ........................................................ 34 Annex 2: External Debt ................................................................................................... 35 Annex 3: Main Functions and Required Skills in a Debt Management Office .................. 36 ii Figures Figure 1. Central Government Debt, 2008–2012 ................................................................... 4 Figure 2. External Debt by Currency, December 2012 ........................................................... 5 Figure 3. Redemption Profile of Existing Debt, 2013–2043 ................................................... 6 Figure 4. Fiscal Trends, 1996–2016..................................................................................... 12 Figure 5. Namibia: Pricing Assumptions for external instruments ....................................... 14 Figure 6. Namibia: Baseline Scenario for Exchange Rates................................................... 15 Figure 7. Namibia: Pricing Assumptions for the domestic instruments ................................ 15 Figure 8. Gross Borrowing by Instrument ............................................................................ 18 Figure 9. Cost-Risk Trade-offs ............................................................................................ 20 Figure 10. Publicly Guaranteed Debt Stock, 2008–2012 ...................................................... 22 Figure 11. Relative Sizes of Bond Markets, 2011 ................................................................ 26 Figure 12. Secondary Bond Market Trading, 2005–2012 ..................................................... 27 Tables Table 1: The 2005 SDMS Debt Thresholds ........................................................................... 3 Table 2. Namibia’s Public Debt Portfolio, December 2012.................................................... 5 Table 3. Cost and Risks of the Existing Debt Portfolio .......................................................... 6 Table 4. Namibia: Selected Economic Indicators used in the MTDS Analysis, 2007–17 ........ 9 Table 5. Strategies and Objectives ....................................................................................... 17 Table 6. Implied Net Borrowing as Share of GDP ............................................................... 18 Table 7. Risk Indicators of the strategies under the baseline scenario: 2012 vs 2017 ............ 19 Table 8. Debt stock to GDP as at end 2017, Baseline and Shock Scenarios.......................... 20 Table 9. Interest payments to GDP as at end 2017, Baseline and Shock Scenarios ............... 20 Table 10. Domestic Debt Securities Traded in Namibia, 2001 ............................................. 27 Table 11. Foreign Loans Outstanding, 2012 ........................................................................ 35 Boxes Box 1. Fiscal Rules vs Debt Management Objectives ............................................................ 3 Box 2. Strategic Benchmarks .............................................................................................. 21 iii Acknowledgements This report summarizes the analysis and findings of a Macroeconomic and Financial Man- agement Institute (MEFMI)-World Bank capacity-building mission conducted in November 2012 as part of the World Bank’s technical assistance to the Ministry of Finance on economic management (activity P133682). The team comprised Lekinyi Mollel of MEFMI and Cigdem Aslan, Rodrigo Cabral, Alvaro Manoel and Philip Schuler (task team leader) of the World Bank. Guilherme Pedras (IMF) backstopped the mission. The mission provided training in the application of the IMF-World Bank Medium-Term Debt Management Strategy frame- work to members of the Ministry of Finance–Bank of Namibia working group that is respon- sible for updating Namibia’s 2005 Sovereign Debt Management Strategy. The MEFMI-World Bank team wishes to thank members of the Ministry of Finance-Bank of Namibia working group, led by Deputy Director of Cash and Debt Management Marten Ashikoto of the Ministry of Finance, for their contributions during the training, and to Direc- tor of Asset Cash and Debt Management Angelina Sinvula of the Ministry of Finance for her overall direction of the capacity-building exercise. The team also acknowledges the com- ments, criticisms and suggestions provided by peer reviewer Ralph van Doorn (World Bank) and by Evans Makini Osano (IFC), Friska Parulian (IMF) and Gerson Kadhikwa (World Bank), and staff of the Ministry of Finance and Bank of Namibia. iv I. Introduction A. Updating Namibia’s Debt Management Strategy 1. Since gaining its independence 23 years ago, Namibia has established an enviable track record of political stability, prudent macroeconomic policies, moderate growth, poverty reduction, and natural resource conservation. The country has achieved these gains while fac- ing constraints imposed by geography, legacies of apartheid and colonialism, and the chal- lenges of constructing a national government. Daunting challenges remain, however. Namibia suffers from chronic high unemployment, the ravages of HIV/AIDS, and one of the world’s most skewed distributions of income. The structure of the economy has remained fundamentally unchanged since Independence: minerals and metals make up the majority of exports; the public sector remains the largest employer; and there has been little investment in labor-intensive manufacturing, which in many countries has absorbed low-skilled labor exit- ing traditional agriculture. 2. To address these challenges, Namibia is launching a number of new initiatives to transform the economy and thereby increase economic growth, create jobs, and reduce pov- erty and inequality. In 2011, the Ministry of Finance (MOF) tabled a bold spending package aimed at stimulating job creation that increased spending by 36 percent. The Government of the Republic of Namibia (GRN) subsequently launched a series of new strategies, starting with a new financial sector strategy and the first industrial policy, and culminating in the Fourth National Development Plan (NDP4), which was released in July 2012. These initia- tives call for increased investments in public infrastructure and targeted industries, an ex- panded use of public-private partnerships (PPPs), and new regulations governing investment and financial market operations. 3. These initiatives have also led to dramatic changes in the size and composition of sovereign debt, thus placing debt closer to the upper limits of Namibia’s fiscal targets and introducing new challenges for managing this debt. Central government debt almost doubled between 2010 and 2012: to 26.5 percent of GDP in the first quarter of 2012 from 14.1 percent in the second quarter of 2010. 1 The GRN issued its first bond in the international market in October 2011 with a US$500 million Eurobond, followed in November 2012 with a R850 million issue on the Johannesburg Stock Exchange (JSE). 4. MOF began updating its debt management framework in 2012 to better address this new environment. This report is one component of a technical assistance program requested by MOF to enhance economic management. The report summarizes findings of a November 2012 mission by staff of the Macroeconomic and Financial Management Institute of Eastern and Southern Africa (MEFMI) and the World Bank, which helped the authorities develop ca- pacity to analyze the costs and risks of alternative debt management strategies. The report follows a MEFMI-World Bank assessment of debt management procedures and institutions conducted in March 2012. 2 Expected future components include training in the analysis of 1 Bank of Namibia Quarterly Bulletins. 2 MEFMI and World Bank, “Debt Management Performance Assessment of Namibia,� March 2012. 1 debt sustainability and technical support for policies on sub-national finance and public- private partnerships (PPPs). B. Objective and Scope of the Report 5. This report uses the Medium-Term Debt Management Strategy (MTDS) framework developed by the International Monetary Fund (IMF) and the World Bank to analyze options facing the GRN as it prepares the new Sovereign Debt Management Strategy (SDMS). 3 This framework emphasizes the explicit analysis of relative costs and risks in a debt management strategy, the linkages between the debt strategy and other macroeconomic policies, and the strategy’s consistency with debt sustainability. 6. The report opens with a review of the GRN’s current debt management strategy, the sources of financing available to the government, and the macroeconomic environment. The report then applies the MTDS Analytical Tool to analyze costs and risks of alternative debt management strategies that were developed by MOF participants in the November 2012 ca- pacity-building exercise. 4 It also examines domestic debt market development and contingent liabilities arising from government guarantees—two issues of special concern to the GRN. Finally, it discusses institutional arrangements and implementation issues. II. Current Debt Management Strategy and Debt Portfolio A. Current Debt Management Strategy 7. The SDMS of 2005 guides public debt operations in Namibia. 5 The strategy’s objec- tives are to minimize the costs of government borrowing, consistent with an acceptable level of risk, and to develop the domestic debt market. In addition to these objectives, the strategy proposes changes in the institutional and legal framework for debt management, measures planned for the development of the domestic debt market, and the use of fiscal rules and debt management strategic benchmarks to guide government borrowing. The scope of the strategy is general government debt—all borrowing by the central government and parastatals, on- lending arrangements and government guarantees. 8. The SDMS also contains an analysis of debt sustainability, which tries to indicate lev- els of future public debt which could be considered sustainable in terms of government public finance. The SDMS lacks an evaluation of the cost and risk trade-offs of alternative strate- gies, including a sensitivity analysis of the main risks, such as the exchange rate and interest rate fluctuations, however. 9. The 2005 SDMS presents benchmarks for several indicators of the cost and risk of debt, which are summarized in Table 1. It should be noted that some—notably ratio of debt to exports or government revenues—are strictly speaking fiscal rules rather than debt manage- 3 Source documents are IMF and World Bank, “Guidelines for Public Debt Management,� March 2001, and World Bank and IMF, “Developing a Medium-Term Debt Management Strategy—Guidance Note for Country Authorities,� February 2009. 4 See Annex 1 for a description of the MTDS Analytical Tool. 5 The SDMS was meant to guide the debt management operations for five years, ending in 2010. Since then it has continued served as the de facto guide for the GRN’s debt management activities. 2 ment indicators (see Box 1). Currency risk is addressed implicitly by the separate limits on external and domestic debt relative to GDP. Table 1: The 2005 SDMS Debt Thresholds Benchmark Indicator Threshold Domestic Debt (including hedged debt) Domestic debt to GDP 20% Domestic debt to revenue 70% External Debt (unhedged) External debt to GDP 5% External debt to exports 10% Guarantees Outstanding guarantees to GDP 10% Total Debt Total debt to GDP 25% Total debt service to revnue 10% Debt falling due within 12 months 20% Fixed interest rate loans as a share of total 90% Average time to maturity (years) 5 years Source: Ministry of Finance, Sovereign Debt Management Strategy 10. A medium-term expenditure framework (MTEF) funding strategy complements the SDMS. MOF prepared a MTEF funding strategy in February 2012 that details Box 1. Fiscal Rules vs Debt Management Objectives the government’s financing composi- A fiscal rule defines how the government should re- tion in the medium term and provides spond to deviations from an expected fiscal position, such the basis for the annual borrowing plan. as keeping the public debt stock below a certain level or a The MTEF funding strategy is meant to proportion of GDP. Fiscal rules can be also more generic such as ceiling for payroll or other kind of expenditure as be reviewed annually. well as limit for taxation (in order to contain the size of 11. The government has generally the government). The traditional objective of debt management is to adhered to the 2005 SDMS bench- meet the government’s funding needs at the lowest possi- marks, particularly those related to total ble cost given a prudent level of risk. Additional debt debt, guarantees, and in designing the management objectives might be to support a well- annual domestic borrowing plan. Many functioning securities market or to reduce budgetary vola- of the institutional reforms proposed in tility by smoothing the maturity profile of government debt. the SDMS have not been implemented, Different and separate analytical tools are used to de- however. For example, it has been diffi- velop fiscal rules and debt management objectives, and cult to establish clear front-, middle-, these are generally presented in different government doc- and back-offices in the Ministry of Fi- uments. Fiscal rules on debt are often developed through a nance’s (MOF’s) Division of Cash and formal debt sustainability analysis and presented in a gov- ernment’s fiscal policy framework. Debt management Debt Management due to high turnover objectives are achieved through the debt management and hiring rigidities. In addition, the strategy. How well a strategy meets stated debt manage- 2011 Eurobond led to external debt ex- ment objectives is presented in the government’s medium- ceeding the SDMS’s threshold of 5 per- term debt strategy, which targets the composition of the cent of GDP. debt portfolio in terms of currencies, interest rate type and maturities 3 B. Namibia’s Existing Public Debt Portfolio 12. Central government debt has been accumulating rapidly since 2010, as shown in Fig- ure 1, and at the close of 2012 totaled N$26.3 billion (equal to 25 percent of GDP). The GRN relies on the domestic bond market to finance fiscal deficits, supplemented by a portfolio of external loans to finance development projects. As will be discussed below, the GRN has re- cently turned to international bond markets to expand its borrowing options. Figure 1. Central Government Debt, 2008–2012 30.0 Billions of Namibian dollars 25.0 20.0 15.0 10.0 5.0 0.0 Treasury bills Treasury bonds Foreign loans Foreign currency bonds Source: Bank of Namibia Quarterly Bulletins and 2012 Annual Report 13. Domestic debt: Treasury bonds (also known as internal registered stock or govern- ment bonds) maturing between 2014 and 2030 currently make up 53 percent of the GRN’s domestic debt (see Table 2). 6 Treasury bills make up the balance of domestic debt, and are issued in tenors of 91, 182, 273 and 364 days, with the 364 days accounting for 48 percent of total T-bills (as of December 2012). Auctions are held weekly. 14. External debt: Until 2011, the GRN’s external debt consisted of loans from bilateral and multilateral creditors to finance development projects. This picture changed dramatically in October 2011, when Namibia issued the US$500 million Eurobond, followed last year with the launch of the R3 billion medium-term note program on the Johannesburg Stock Ex- change. 7 The Eurobond and the R850 million JSE bond issued in November 2012 make up 57 percent of total external debt (Table 2). 8 15. A portfolio of around 50 loans from multilateral and bilateral creditors make up the balance. 9 Original maturities of these loans range from 9 to 50 years, most with 10-year grace periods. The lion’s share of foreign loans—88 percent of the Namibian dollar value of out- standing loans—have fixed interest rates (ranging from zero to 4.27 percent). Three loans 6 In July 2013 Namibia will begin issuing the GC35 bond, which will mature in 2035. 7 Rand-denominated debt is classified as external debt, although the authorities treat it as not having foreign exchange risk, due to the CMA’s arrangements on exchange rates and capital mobility. 8 Both are 10-year bonds. The Eurobond was priced at 5.5 percent; the JSE bond at 8.26 percent. 9 Loans from Japan International Cooperation Agency (JICA), Kreditanstalt für Wiederaufbau (KfW), African Development Bank (AfDB), and the Export-Import Bank of China account for 65 percent of total foreign loans. A list of outstanding external loans is presented in Table 11 on page 36 of the annex. 4 from the African Development Bank (AfDB) have interest rates based on the Johannesburg Interbank Average Rate (JIBAR) plus a margin of 50 basis points. Table 2. Namibia’s Public Debt Portfolio, December 2012 Domestic Debt Treasury Bills Amount Treasury Bonds Amount 91-day 700 GC14 (7.5% coupon) 1,530 182-day 1,972 GC15 (13.0% coupon) 1,647 273-day 1,550 GC17 (8.0% coupon) 858 365-day 3,820 GC18 (9.5% coupon) 1,743 Total T-Bills 8,042 GC21 (7.75% coupon) 784 GC24 (10.5% coupon) 2,077 GC27 (8.0% coupon) 307 GC30 (8.0% coupon) 289 Total Domestic Debt 17,278 Total T-Bonds 9,236 External Debt Category of Instrument Amount Eurobond (USD) 4,236 Bilateral, euro (all fixed rate) 1,637 Bilateral, other currencies (all fixed rate) 1,186 JSE Bond (ZAR) 850 Multilateral, fixed rate (various currencies) 769 Multilateral, floating rate (ZAR) 477 Total external debt 9,154 Sources: Bank of Namibia and Ministry of Finance Notes: Names of treasury bonds indicate the year when they mature. Amounts are expressed in millions of Namibian dollars. 16. As a result of the Eurobond issue, almost half of external debt is denominated in U.S. dollars (96 percent of which is the Eurobond). The euro previously was the largest single cur- rency represented in Namibia’s external debt portfolio. Figure 2 shows the distribution of ex- ternal debt by currency at the close of 2012. Figure 2. External Debt by Currency, December 2012 Euro Rand 21% 14% Yen 11% Other 16% Yuan U.S. dollar 5% 48% Kuwaiti dinar Swiss Franc 0.4% 0.3% Source: Bank of Namibia 5 C. Cost and Risk Characteristics of Existing Debt 17. The cost-risk characteristics of the existing central government debt portfolio are mostly consistent with the 2005 SDMS benchmarks. The ratios of total debt and domestic to GDP were 24.1 and 16.2 percent in October 2012, slightly below the SDMS’s thresholds of 25 and 20 percent, respectively. In the wake of the 2011 Eurobond issue, external public debt is now at 7.9 percent of GDP—above the SDMS’s threshold of 5 percent of GDP. Table 3. Cost and Risks of the Existing Debt Portfolio Risk Indicators External Domestic Total Nominal debt (share of GDP) 7.9 16.2 24.1 Present value of debt (share of GDP) 7.3 16.2 23.6 Cost of debt: Weighted average interest rate 3.9 7.8 6.5 Refinancing risk: Average time to maturity (years) 7.8 3.4 4.9 Debt maturing in 1 year (share of total) 4 49 34.3 Interest rate risk: Average time to refixing (years) 7.5 3.4 4.8 Debt refixing in 1 year (share of total) 9.2 49 34.3 Fixed rate debt (share of total) 94.2 100 98.1 Foreign currency risk: Foreign exchange debt (share of total) 32.7 Source: Staff calculations during the November 2012 mission, excludes the JSE bond 18. The cost of debt, as measured by the weighted average interest rate, is 7.8 percent for domestic debt and 3.9 percent for external debt.10 The relatively small share of external debt (32.7 percent) to total debt has reduced the portfolio’s exposure to foreign currency risk, but the large share of short-term domestic debt, particularly treasury bills, which account for about 31.8 percent of total debt, exposes the debt portfolio to both refinancing and interest rate risks. Table 3 summarizes costs and risks of the existing debt portfolio. Figure 3. Redemption Profile of Existing Debt, 2013–2043 9 8 Billions of Namibian Dollars 7 6 5 4 3 External Domestic 2 1 0 2013 2015 2017 2019 2021 2023 2025 2027 2029 2031 2033 2035 2037 2039 2041 Source: Ministry of Finance and team calculations 10 The cost for the existing debt is calculated based on the stock of debt as of end of 2012, which was USD 2,882 million. It represents expected cost for 2013 based on the cash flow data prepared by the Asset, Cash and Debt Management Directorate. One must compare these interest rates with caution as they reflect debt denomi- nated in different currencies. 6 19. The debt portfolio is exposed to large refinancing risks in 2013 and 2021. Figure 3 shows that maturities are bunched in these two years, reflecting dominance of T-bills—which account for around half of total domestic debt—and the Eurobond. It seems likely that the GRN will not encounter problems rolling over T-bills in 2013, given the nature of the domes- tic capital markets, which reduces the refinancing risk somewhat. Likewise, the good rating of the Namibian economy is expected to reduce the risk exposure when the Eurobond and JSE bonds are refinanced. Finally, MOF regularly transfers funds into a redemption funds at the BON to mitigate refinancing risk. 11 III. Sources of Financing A. External Sources of Financing 20. Although Namibia’s external debt portfolio contains some concessional debt, future sources of external financing are largely limited to debt on semi-concessional and commer- cial terms, given the country’s relatively high per capita income. Namibia has entered into few new external loans since the 2005 SDMS was released. 12 21. To diversify its sources of financing, the government has recently launched a ZAR 3 billion medium-term note program on the Johannesburg Stock Exchange, issuing an initial ZAR 850 million 10-year bond on November 14, 2012. 13 As Namibia is strongly linked to the South African economy and is a member of the Common Monetary Area (CMA), the Namibian government has a leeway in accessing the vibrant South African capital markets for financing. 14 22. Financing from multilateral and bilateral semi- and non-concessional sources (some of which are relatively concessional when compared to domestic sources) is still available. Nevertheless, the government is concerned about exposing the debt portfolio into foreign cur- rency risk, particularly at the juncture where the current level of international official reserves is relatively low. 23. Over the medium term, the government intends to continue its policy of minimizing external sources of financing. The Namibian Eurobond is trading favorably in the interna- tional capital markets with yields averaging 3.5 percent, well below the issuance coupon rate 11 As of December 2012, the Internal Registered Stock Redemption Account held N$2.6 billion (Bank of Na- mibia Annual Report 2012). Such redemption accounts are a common measure to mitigate risk in countries that rely on bond market financing. 12 Since 2005, Namibia has entered into loans with the Japanese International Cooperation Agency (JICA), Ex- port-Import Bank of China, and BNP-Paribas. 13 The program is being arranged by a consortium of banks, including Absa Bank, FNB of Namibia, First Rand Bank, Namibia Equity Brokers, and RMB Namibia. The initial offering was priced at 8.26 percent and was twice oversubscribed. This is the first sovereign bond issued on the JSE by an entity other than the South Afri- can National Treasury. See Government of the Republic of Namibia, “ZAR 3,000,000,000 Medium Term Note Program,� November 2, 2012, and Ministry of Finance press statement of November 14, 2012. 14 CMA members are Lesotho, Namibia, South Africa and Swaziland. CMA rules provide for unrestricted trans- fers of funds among member countries. Namibia and the other smaller CMA members fix their currencies to the rand at par and hold foreign reserves at least equivalent to local currency issued. The rand circulates as legal tender throughout the CMA. 7 of 5.5 percent, which reflects favorable perceptions of the economy. 15 Thus, although the government is not firm on whether or when the next entry into the capital markets will take place, the option is still feasible at least in the medium term. B. Domestic Sources of Financing 24. The domestic market remains the major source for financing the government deficit. The 2005 SDMS calls for maintaining domestic debt at around 80 percent of the total gov- ernment debt stock. Treasury bills and bonds have traditionally been the domestic sources of financing. 25. Treasury bills are expected to continue being an important source for meeting the government’s borrowing in the medium term. As of December 2012, 47 percent of the out- standing domestic debt of N$17.3 billion was treasury bills, with the balance being treasury bonds having tenors ranging from three years to twenty years (see Table 2 above on page 5). Commercial banks and other financial institutions prefer to invest in T-bills given their ex- cess liquidity and liability structures. The GRN requires institutional investors (e.g., pension funds and insurance companies) to invest at least 35 percent in the local market. This guaran- tees demand and reduces the government’s refinancing risk, consistent with the rollover poli- cy of the government. 26. Pension funds are expected to dominate the bond market. The Government Institu- tions Pension Fund (GIPF) is the major holder of long-term instruments and is expected to continue playing a leading role in the long-end market. 16 Medium- to long-term government bonds are particularly attractive to pension funds in light of the requirement to invest 35 per- cent of assets locally, given the limited availability of domestic investments IV. Baseline Macroeconomic Assumptions and Key Risk Factors 27. A central tenet of the IMF-World Bank MTDS framework is that a country’s debt management strategy should be underpinned by, and consistent with, the government’s over- all medium-term macroeconomic framework. The underlying macroeconomic assumptions and outlook should be regularly reviewed and updated as necessary. Although challenging, the authorities must understand the risks inherent in the macroeconomic projections, how the- se risks will impact the government’s financing needs, and what implications they have for the desired currency, interest rate, and maturity composition of public debt. In some instanc- es, significant changes in the macroeconomic environment will require a change in strategy. A. Baseline Macroeconomic Assumptions 28. The effects of the ongoing global economic slowdown on Namibia and spillovers from South Africa remain the key factors driving the current outlook. As a small open econ- 15 For example, Fitch reaffirmed its investment-grade rating of Namibian sovereign debt in December 2012, after downgrading South African debt. Moodys reaffirmed its investment-grade rating in January 2013. 16 GIPF is a defined-benefit pension fund whose members are employees of the GRN and other public institu- tions. As of September 31, 2012, its assets exceeded NAD 55 billion (The New Era, November 7, 2012). Mem- bers of the board of trustees are selected equally by the GRN, public employees unions, and the Public Service Commission. GIPF’s assets are managed by private asset management firms domiciled in Namibia. 8 omy, Namibia depends heavily on global economy performance. Given the expectations that the international crisis will continue for some time, and that the economy of South Africa may face economic and social difficulties in the coming years, Namibia’s baseline outlook for the medium-term should be moderate. For the analysis in this report, therefore, we assume GDP to grow at around 4.4 percent, slightly below growth in 2011 and 2012 (currently esti- mated at 4.9 and 5.0 percent, respectively), but in line with the authorities’ projections for the medium term. 17 The secondary industries will grow above average due to favorable perfor- mance in the construction sector (generated by rising government expenditures, port expan- sion, and energy projects). The primary sector is expected to grow below average. Trade will continue to be significant for Namibia, and exports will remain based on minerals. Table 4. Namibia: Selected Economic Indicators used in the MTDS Analysis, 2007–17 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 (millions of Namibian dollars) Revenue and Grants 20,672 23,447 24,017 23,678 29,962 32,439 36,344 39,439 43,118 46,944 47,897 Expenditure 17,541 21,898 25,712 28,142 36,743 38,212 39,637 44,310 49,072 54,854 61,813 of which interest payments 1,179 1,111 1,197 966 1,130 1,842 2,112 2,447 2,931 3,596 4,411 Primary Expenditures 16,362 20,787 24,515 27,176 35,613 36,370 37,525 41,863 46,141 51,258 57,402 Fiscal Balance 4,310 2,660 -498 -3,498 -5,651 -5,773 -3,293 -4,871 -5,954 -7,910 -13,916 (in percentage of GDP) Revenue and Grants 31.9% 31.8% 31.4% 28.3% 32.0% 31.9% 32.1% 31.2% 30.6% 29.9% 27.1% Expenditure 27.1% 29.7% 33.6% 33.7% 39.3% 37.6% 35.0% 35.1% 34.8% 34.9% 35.0% Fiscal Balance 6.7% 3.6% -0.7% -4.2% -6.0% -5.7% -2.9% -3.9% -4.2% -5.0% -7.9% Memorandum Items GDP (millions of current NAD) 64,798 73,629 76,587 83,562 93,559 101,710 113,190 126,206 140,832 156,985 176,608 Total debt to GDP 18.4% 18.2% 15.6% 16.1% 26.5% 26.4% 26.6% 27.7% 29.1% 31.1% 27.6% Gross international reserves (end of period) Months of imports 3.0 3.8 4.1 2.6 3.3 3.1 3.0 2.9 3.0 3.0 3.1 Ratio of reserves to short-term 1.3 4.0 4.6 2.1 4.2 4.1 3.7 3.6 4.2 4.4 4.4 Sources: Namibian authorities, IMF country reports, team calculations Notes: GDP is presented on a fiscal year basis; future projected debt/GDP ratios come from debt sustainability analysis pro- jections 29. One important assumption used in MTDS AT training and maintained in this report is that the government fiscal stimulus initiated in 2011 will not come to an end soon and the current expenditures as proportion of GDP will remain at the same level as 2012. This means that for the projections using the MTDS Analytical Tool, the fiscal space won’t be exhausted until FY2016–17, although capital investments, which were increased during the fiscal stimu- lus period, will de-accelerate slowly during the medium-term. This baseline scenario differs somewhat from MOF’s February 2012 medium-term framework and other information shared during the November 2012 mission, which assumed a fiscal consolidation and to pre- crisis levels of expenditure. The medium-term macroeconomic framework used for the sce- nario analysis is consistent with the framework outlined by the BON using one of the scenar- ios in which the ratio of current expenditures to GDP is kept at the same level as was then expected for 2012. 18 17 NSA, “Preliminary National Accounts 2012,� Mach 2013. The macroeconomic framework released by MOF after the mission (in February 2013) assumes roughly similar growth rates: 4.3 percent in 2013–14 and 4.4 per- cent in 4.4 in 2015–16. BON’s economic outlook at the time of the mission was for growth of 4.4 percent in 2013. The IMF expects growth of 4.2 percent in 2013, 4.0 percent in 2014, and then 4.3 percent in 2015–17. 18 The BON’s analysis includes several alternative expenditure scenarios, one of which informed the baseline used in the MTDS analysis presented in this report. 9 30. There are two main reasons that a “higher fiscal deficit� scenario was used in the analysis. First, if one assumes the fiscal consolidation that was present in the government baseline, there will be no deficit, no new increase of public debt—on the contrary, there will be a reduction in the debt. This assumption will make the analysis of the debt strategies for the future almost indifferent. Second, in light of the government’s long-term development plans, especially those indicated in NDP4, it is reasonable to assume that some additional in- crease in the expenditures will take place in the coming five years. As events unfolded, the MTEF that was released after the mission projected an increased deficit for FY2013/14 rather than the consolidation projected in the previous framework. This underscores the message that the analysis of debt strategies should be updated and repeated annually, or more fre- quently if policies or conditions change significantly. 31. This report assumes that the Namibian dollar will continue to be pegged to the rand and that monetary policy will be conducted as in the past under the CMA—policies that have served Namibia well. 19 Inflation, which increased recently due to food and fuel prices, is ex- pected to decline below 5 percent a year towards the end of the medium-term. Due to the global uncertainty in international financial markets, exchange rates with major world curren- cies are likely to remain volatile. The rand (and therefore Namibian dollar) appreciated sharp- ly from 2009 to 2011 when South Africa faced capital inflows as a consequence of monetary easing in the U.S. and Europe. Since mid-2011, the rand has depreciated substantially against the U.S. dollar. B. Key Vulnerabilities 32. Namibia’s open and small economy faces downside risks that come mainly from the global economic performance of industrial and emerging economies. Trade and balance of payments performance are significant to explain domestic dynamics. Specifically, the follow- ing risks could be mentioned: • The current external environment remains unsettled. Difficulties in the Euro zone and slow recovery in the U.S. implies downside risks to commodity prices, espe- cially minerals, which play important roles in the primary sector and exports. • Rand exchange rate volatility implies higher volatility for the Namibian dollar and higher risk for debt management • The downgrade of South African debt by rating agencies may have some negative spillovers in the medium-term. 33. Domestic investment may be hit by external events as well. Exports of goods and ser- vices are estimated at 44.1 percent of GDP in 2012 and will remain for a long period of time one of the key channels through which external events affect the domestic economy. 20 A slowdown in the global economy would reduce private investment in export industries, with spillovers on the rest of the economy. The GRN’s ability to finance development projects may be hit severely as well, through reduced tax revenue (including royalties and export tax- 19 This is in line with the latest IMF country report “Namibia: Article IV Consultations,� February 2013. Moody’s cited the peg to the rand as one important factor behind its investment-grade rating of Namibian sover- eign debt (Moody’s Investors Service, “Namibia,� January 31, 2013). 20 IMF, “Namibia: Article IV Consultations,� February 2013. 10 es on minerals exports), tightening in credit conditions, and increased uncertainty about the economic recovery worldwide. 34. The continuation of the fiscal stimulus would contribute to economic growth but could lead to deterioration of the external balance. Additional fiscal expenditures may put upward pressure on non-tradables’ prices, imports, and the current account deficit. If the ex- ternal balance deteriorates, foreign reserves would likely fall below desired levels needed to protect the exchange rate peg and serve as a buffer against shocks. The issuance of the Euro- bond and JSE medium-term note are attracting increased external scrutiny of Namibia’s do- mestic economy, policies and macroeconomic indicators. 35. High unemployment and income inequality are sources of risk. The most recent data estimate the Gini index at 0.597 and the national unemployment rate at 27 percent, with the rate climbing to 35 percent in some regions. 21 High youth unemployment rates—56 percent for the 15–19 year old cohort and 49 percent for 20–24 years—are of special concern because they imply that the Namibian is forfeiting the demographic dividend of its large youth cohort, which if efficiently employed would contribute to economic expansion. Furthermore, labor unrest in South Africa may disrupt Namibia’s supply chains or spill over to similar unrest in Namibia. 36. Transfers from the SACU common revenue pool also present a risk for the country and for the budget. Transfers are large and vary greatly from year to year—accounting for 25–40 percent of total revenue in recent years. Their future is uncertain. Transfers would de- cline if the South African economy slows. SACU members are negotiating changes to the revenue-sharing formula, which likely would result in lower transfers to Namibia. Either a decline or increased volatility in transfers could hinder the government’s medium-term ex- penditure program and compromise economic growth. 37. Contingent liabilities and guarantees provided to SOEs and other entities remain a key source of fiscal risk. 22 Currently explicit guarantees are small—2.1 percent of GDP at the end of 2012—and well below the 2005 SDMS’s threshold of 10 percent. The stock of guarantees could grow rapidly if urgently needed infrastructure investments are made as envisioned in NDP4. The DSA recently carried out by the authorities and shared with the mission does not take into account any future contingent liabilities arising from these commitments, however. Future DSAs and revisions of the SDMS should treat a proportion of the guarantees or other contingent commitments of the government as potential public debt (or expenditure). One simple way of quantifying potential government contingent liabilities is to calculate how much guarantee was called in the past as proportion of the total and then apply the ratio for the future guarantees, with the broad assumption that the past is representative of the future. If there are also on-lending operations with SOEs of subnational government, the same meth- odology could be applied. 38. In conclusion, the macroeconomic risks suggest that targeting domestic borrowing could be a safer strategy, but these risks need to be weighed against the high domestic interest 21 Namibia Statistics Agency, Namibia Household Income and Expenditure Survey 2009/10 and Namibia Labor Force Survey 2012. Unemployment rates include discouraged workers. 22 Section VI, starting on page 21, discusses contingent liabilities stemming from government guarantees. 11 rates in a more rigorous manner. Namibia’s revised SDMS also needs to take into account developing more flexible domestic financing sources, and potential crowding out effects. C. Debt Sustainability 39. Historically low levels of public debt are due to prudent fiscal policies that have con- tributed to macroeconomic stability. During the decade and one half prior to the 2011 fiscal stimulus, overall fiscal deficits averaged less than 3 percent of GDP and debt usually fell be- low 20 percent of GDP, as shown in Figure 4. The budget tabled in 2011 pushed spending up to an unprecedented 40 percent of GDP at a time when revenues had barely recovered from the global economic crisis. 40. The increase in the fiscal deficit—and consequently on public debt—was due to the fiscal stimulus and an important decline in government mineral revenues that followed the global financial crisis. Figure 4. Fiscal Trends, 1996–2016 45% 35% Share of GDP 25% 15% 5% -5% -15% Domestic debt External debt Revenue Spending Overall Balance Source: Ministry of Finance, Bank of Namibia Notes: Future projections from MTEF tabled in February 2013 41. The government baseline scenario assumes a fiscal balance near zero by FY2015/16. All fiscal stimulus, including the Targeted Intervention Program for Employment and Eco- nomic Growth (TIPEEG), is unwound, therefore, gross debt will fall to 21 percent by 2016/17 from around 26 percent of GDP in 2011. 23 It shows that following the general fiscal strategy indicated in the current MTEF, the government will be generating surplus, reducing public debt as a proportion of GDP, but will be systematically, spending less, especially in investments. 23 Namibia’s medium-term (gross) public debt sustainability used the quantitative framework developed by the IMF for middle-income countries. The IMF methodology uses simple algebra on debt sustainability of a country based on various key macroeconomic indicators: government revenue, expenditure, deficit, debt-to-GDP ratio, nominal interest, GDP Deflator, real GDP growth rate, inflation rate, and interest rate. 12 42. In a scenario where current fiscal policies are assumed unchanged in the medium- term, gross public debt would rise to 44.2 percent of GDP by 2016/17, indicating an unsus- tainable trend. Unchanged fiscal policies scenario means an average fiscal deficit to GDP of about 6.2 percent in the medium-term, which would push debt above the GRN’s indicative limit of 35 percent of GDP. 43. A more realist scenario of deficits averaging 4 percent of GDP would likely generate a public debt trend within historical limits. At the same time, it would create a necessary fis- cal space for the government to continue to stimulate the economy and complete basic public infrastructure projects. V. Cost-Risk Analysis of Alternative Debt Management Strategies 44. This section applies the IMF-World Bank MTDS Analytical Tool to evaluate conse- quences of following a particular debt management strategy under various scenarios for mac- roeconomic and market variables. The Analytical Tool uses data on Namibia’s current debt portfolio and macroeconomic policies discussed in the previous sections. The MEFMI-World Bank team worked with BON and MOF debt management specialists to develop illustrative debt strategies and plausible shock scenarios. The base year for the analysis is 2012, and sce- narios have a five-year horizon. The output generated by the MTDS Analytical Tool is a number of cost and risk indicators. The analysis provides insight into the key vulnerabilities embedded in the specific strategy under consideration. A. Baseline Interest and Exchange Rate Assumptions 45. As described earlier in Section II.B, the GRN’s debt portfolio contains a range of in- struments with different maturities, interest rates, and currencies. To facilitate the cost-risk analysis of different debt management strategies, the team aggregated 63 external and domes- tic instruments recorded in the debt database as at end of October 2012 into ten instruments according to their contractual terms before applying the IMF-World Bank MTDS Analytical Tool. These instruments are: multilateral loans, bilateral loans denominated in U.S. dollars, bilateral loans denominated in euros, floating-rate external loans, Eurobonds, JSE bonds, treasury bills, and 3, 5, 10 and 15-year treasury bonds. 24 This section describes these different categories and presents assumptions the report makes regarding pricing and exchange rates. External sources 46. The GRN’s external debt is grouped into the following categories: bilateral loans in U.S. dollars and euros; floating-rate loans; the 10-year Eurobond; and the 10-year JSE bond (in South African rand). • Bilateral loans in U.S. dollars and euros: We assume that the current rates are the same of the last loans of this kind contracted by Namibia. These loans typical- ly have a maturity of 15 years with a grace period of 5 years, which result in an 24 The instrument referred to as bilateral debt denominated in U.S. dollars includes all bilateral debt in curren- cies other than euros. The existing government bond debt includes bonds of maturities ranging from 3 to 20 years that are grouped into the five representative maturities. 13 average repayment of 10 years. We assume that rates will follow the same path as the 10-year forward U.S. Treasury. • Floating-rate foreign loans: These are linked to the Johannesburg interbank agreed rate (JIBAR) plus a constant spread of 0.5 percent. We assume that the JIBAR will follow the same movements of the London interbank offered rate (Li- bor) over the next five years, and take the U.S. dollar Libor 6-month forward rates as reference. • Eurobond: The 10-year Eurobond is priced using the 10-year forward U.S. Treas- ury plus a spread of 216 basis points (bps). This spread corresponds to the spread over U.S. Treasuries of Namibia’s outstanding 10-year Eurobond, as of November 3, 2012. • JSE bond: The 10-year, rand-denominated bond issued in November 2012 on the JSE is priced identically to the 10-year bond issued in the domestic market. (The pricing of domestic bonds is explained below.) Figure 5. Namibia: Pricing Assumptions for external instruments 12% 11% 10% Interest Rate 9% 8% 7% 6% 5% 2012 2013 2013 2014 2014 2015 1y 3y 5y 10y 15y Source: Team calculations 47. For baseline currency depreciation scenarios, we took GRN projections of annual de- preciation against the U.S. dollar for the next 5 years and applied it to the current NAD/USD exchange rate.25 We assumed that the NAD/EUR exchange rate will follow a similar path of the NAD/USD exchange rate, but with a lower rate of depreciation due to the economic and financial problems in Europe. 25 MOF, FY2012/13–FY2014/15 Medium-term Expenditure Framework. 14 Figure 6. Namibia: Baseline Scenario for Exchange Rates 12.0 per foreign currency Namibian dollars 11.0 10.0 9.0 8.0 2012 2013 2014 2015 2016 2017 NAD/USD NAD/EUR Source: Team calculations Domestic sources 48. As described above, Namibia issues T-bills of 3, 6, 9 and 12 months maturity, and currently has seven series of bonds outstanding with maturing from 2014 to 2030. 26 In the analysis of alternative borrowing strategies, we assume one representative T-bill of 1-year maturity and four benchmark bonds of 3, 5, 10 and 15-years maturity. The following are the pricing assumptions for these domestic instruments: 27 • BON’s published daily yield curve is used for domestic instruments current yields. The team then assumed that domestic yields will follow the same path as the yields of South African bonds of similar maturities. • We took the current spreads of Namibia domestic bonds over South African bonds of similar maturities and assume that these spreads will remain constant over the period of the analysis. • Then, we use the South African forward curve and add the spread of the Namibian bonds to price the instruments in the following five years. Figure 7. Namibia: Pricing Assumptions for the domestic instruments 8% 7% 6% Interest Rate 5% 4% 3% 2% 1% 0% 2012 2013 2013 2014 2014 2015 Bilateral USD and EUR Floating Loan Eurobond Source: Team calculations 26 In July 2013 Namibia will begin issuing the GC35 bond, which matures in 2035. 27 The steepness of the yield curve depends on more than market expectations, i.e., also the liquidity premium for longer-maturity instruments. This generates a forward curve that is steeper than would otherwise be the case. 15 B. Description of Shock Scenarios 49. The cost-risk implications of alternative debt management strategies are assessed un- der four scenarios that combine interest and exchange rate shocks. Interest rate shocks are applied in year 2014 and exchange rate shock is applied in year 2016. In all the cases, shocks are deemed permanent. • Scenario 1—Exchange rate shock: a 30 percent depreciation of the Namibian dollar against the euro and U.S. dollar. • Scenario 2—Interest rate shock 1: an almost parallel shift of around 200 bps of interest rate (domestic and external), reflecting either tighter monetary conditions or risk aversion in international markets. As a reference, we modeled a downgrade scenario by the difference between the spreads over U.S. Treasuries for different tenors of Baa3 and Ba3-rated entities (Moody’s). Although spreads are different for different maturities, they were all around 200 bps. • Scenario 3—Interest rate shock 2: a shock with similar magnitude (200 bps) as above for the medium- and long-term maturities (equal or above 5 years), and a stress shock of 400 bps for the short term maturities (up to 3 years). • Scenario 4—Combined shock: a 15 percent depreciation shock in combination with interest rate shock 1 (domestic and external). C. Description of Alternative Debt Management Strategies 50. Working closely with debt management specialists at MOF and BON, the team de- veloped the following four borrowing strategies to be used in the analysis, also summarized in Table 5 on page 17. • S1–Current Strategy is the GRN’s current strategy (or implicit strategy) extended for the next five years. The 20/80 external/domestic debt benchmark (see Box 2 on page 21) serves as a proxy for the flow. The team assumed that 20 percent of the funding needs will be met from foreign sources (including the JSE bond) while the rest will be borrowed from the domestic market. The external funding will be mostly through U.S. dollar-denominated project finance loans and the rand-denominated JSE bond. We assume that rand-denominated bonds do not car- ry foreign exchange risk because of Namibia’s membership in the CMA and cred- ible currency peg. Therefore, the only foreign exchange exposure arises from the project finance loans. • S2–Reduced refinancing risk in the domestic market: In order to reduce the high refinancing and interest rate risk due to T-bills, strategy S2 aims to analyze the impact of extending the maturities in the local market by shifting half of the T-bill issuances into the rest of the bonds, all else same as current strategy. This is a big shift and at the moment an extreme strategy. • S3–Increased FX Risk: This strategy assumes that there is a new Eurobond issu- ance in 2013. Project finance loans in U.S. dollars, new rand issuances, and Euro- bonds fund the increased external borrowing. • S4–Reduced FX risk: Since the share of external borrowing already exceeds the 20 percent indicative limit in the 2005 SDMS, strategy S4 assumes domestic issu- 16 ances as in S1 but no new external borrowing at all. 28 This would bring the do- mestic/external debt ratio in line with the benchmarks in the 2005 SDMS. 51. In order to clearly analyze the impact of the alternative borrowing strategies and in- terpret the output, it is recommended that one characteristic is changed at a time as can be seen from the four strategies. Table 5. Strategies and Objectives Strategy Key objective Domestic vs Domestic External Feasibility external S1 Current Strate- 80%–20% Majority in Mostly project Most likely to be im- gy T-bills (87% finance loans in plemented, reflects of total do- USD (68% of market constraints in mestic) total external) terms of appetite for the short term S2 Reduced refi- 80%–20% Shift half of Same as current Extreme strategy nancing risk in the T-bill strategy which can be achieved the domestic issuances to in the medium term as market bonds. the market deepens and steps in order to stimulate the primary and secondary market are taken S3 Increased FX 2013: Same as in - Issuance of There are no market risk 45%–55% Current USD500 million constraints to imple- 2014–17: Strategy Eurobond in ment this strategy as 65%–35% 2013 investor appetite for - Project finance the Eurobond is most loans in USD likely to continue and euros - Higher rand issuance S4 Reduced FX 100%–0% Same as in No external is- The only constraint risk Current suance, just could be the size of Strategy some residual the domestic market. disbursement Given only 20% of the from existing funding is shifted to loans the NAD funding with big part of it in T-bills, this strategy seems achievable. Source: MEFMI-World Bank team 52. Table Table 6 presents the implied net borrowing—internally and externally—as a share of GDP across the four strategies. 28 Even the November 2012 JSE bond issue is excluded from this strategy for illustrative purposes. The simula- tions were conducted before Namibia issued the JSE bond. 17 Table 6. Implied Net Borrowing as Share of GDP External Net Borrowing 2013 2014 2015 2016 2017 Average S1 1.7% 1.9% 2.1% 2.2% 2.9% 2.2% S2 1.7% 1.3% 1.3% 1.4% 2.0% 1.5% S3 5.1% 2.6% 2.9% 3.0% 4.1% 3.5% S4 -0.3% -0.3% -0.2% -0.2% -0.2% -0.2% Internal Net Borrowing 2013 2014 2015 2016 2017 Average S1 0.8% 1.4% 1.5% 2.1% 4.2% 2.0% S2 0.8% 2.1% 2.4% 3.0% 5.2% 2.7% S3 -2.7% 0.7% 0.6% 1.2% 2.8% 0.5% S4 2.7% 3.6% 4.0% 4.7% 7.5% 4.5% Source: Team calculations 53. Figure 8 shows the projected composition of the debt portfolio on average under each strategy, compared to the current portfolio composition. Figure 8. Gross Borrowing by Instrument S4 S3 S2 S1 Current Stock 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% bilateral_USD FX bilateral_EUR FX SA_10y FX Floating loans FX Eurobond FX T-bills DX 3 Year Bond DX 5 Year Bond DX 10 Year Bond DX 15 Year Bond DX Source: Team calculations Notes: Shares are averages across 2013–17; “DX� refers to internal debt and “FX� to external (including rand-denominatd debt). D. Cost-Risk Analysis of Alternative Debt Management Strategies 54. The cost indicators selected in the analysis are the debt stock to GDP and interest payments to GDP ratios. 29 The output represents the debt portfolio at the end of 2017 and contains information on the composition of debt. The evaluation of the cost and risk indica- tors (see Table 7) under the baseline scenario show the following: • Even without shocks, the “Current Strategy (S1)� does not meet any of the current strategic benchmarks and even increases the foreign exchange risk. • Conversely, the “Reduced refinancing risk in the domestic market (S2)� strategy helps meet the refinancing and interest rate risk benchmarks. If the JSE bond is 29 It is possible to choose other indicators such as the nominal amounts or as ratio of revenues, for example. This depends on the circumstances of the country and the authorities’ choice. 18 removed from external debt (as it does not carry foreign exchange risk), the share of external debt converges to the benchmark level (22.9 percent). Therefore, under the baseline this strategy performs well according to this criterion. • On the other hand, the “Reduced FX Risk (S4)� strategy generates the riskiest debt portfolio in terms of refinancing and interest rate risk. This strategy allows reaching the 2005 SDMS external debt benchmark in three years, with external debt falling to 20.1 percent of GDP.30 • The “Increased FX Risk (S3)� strategy performs relatively well by reducing the refinancing and interest rate exposures of the debt portfolio, although increasing the foreign exchange risk. This strategy has the lowest implied interest rate (6 per- cent) and total debt to GDP (32.2 percent) compared to the other strategies, owing to lower Eurobond and loan rates. 55. The next step in the analysis is to assess the behavior of the strategies under the shocks mentioned above. Figure 9 illustrates the cost indicator (debt-to-GDP or interest-to- GDP) on the vertical axis and the maximum deviation in this cost from the baseline cost on the horizontal axis as the measure of risk. From these charts and the numeric results from the tool (see Table 7 and Table 8), it is possible to compare the strategies and observe the trade- offs between costs and risks as well as between the different types of risk. These two cost in- dicators react differently to the shocks: • Debt-to-GDP responds the most to foreign exchange shocks since depreciation of the Namibian dollar affects payments of both principal and interest. • The impact of the interest rate shocks will appear more clearly in the interest-to- GDP indicator because they directly and immediately influence the entire interest amount. Table 7. Risk Indicators of the strategies under the baseline scenario: 2012 vs 2017 Outcome in 2017 2012 under different strategies Risk Indicators S1 S2 S3 S4 Nominal debt (share of GDP) 24.1 32.4 32.7 32.2 32.7 Present value of debt (share of GDP) 23.6 31.6 32.0 31.2 32.5 Implied Interest rate 6.5 6.7 7.2 6.0 7.8 Refinancing risk ATM, external portfolio (years) 7.8 7.6 7.2 7.6 4.6 ATM, domestic portfolio (years) 3.4 2.8 5.4 3.0 3.8 ATM, total portfolio (years) 4.9 4.9 6.0 5.8 3.9 Interest rate risk ATR (years) 4.8 4.9 6.0 5.8 3.9 Debt refixing in 1 year (share) 36.0 38.9 20.1 26.5 48.6 Fixed-rate debt (share of total) 98.1 99.5 99.5 99.5 99.5 Foreign exchange risk FX debt (share of total) 32.7 44.3 35.2 61.4 12.6 Source: Team calculations using MTDS Analytical Tool 56. The “Increased FX Risk (S3)� is the least costly strategy, measured by both debt-to- GDP and interest-to-GDP indicators.31 It is the least risky in terms of interest-to-GDP but the most risky in terms of debt-to-GDP. 30 The analytical tool produces results for 3, 4, 5 and 8 years. Therefore, it is possible to consult all the output sheets up to the horizon of the analysis, in this case the 3, 4 and 5 years. 31 The cost is low due to lower expected interest rate as this strategy assumes the issuance of a Eurobond which would be cheaper than the T-bills. The strategy basically shifts the T-bill issuance to the Eurobond. 19 57. Conversely, the “Reduced FX Risk (S4)� strategy performs in the other direction, i.e., it is the riskiest under interest-to-GDP indicator and less risky under the debt-to-GDP, while its cost is the highest, due to more expensive domestic debt compared to external debt. 32 58. The “Current (S1)� and “Reduced refinancing risk in the domestic market (S2)� strat- egies show comparable performance for both cost indicators. Extending the maturities in- creases costs due to higher interest rates for longer maturities while reducing risk thanks to lower interest rate and refinancing risk. Figure 9. Cost-Risk Trade-offs 32.8% 2.2% Cost: Interest Payments to GDP Cost: Debt Stock to GDP 32.7% S4: All S2: Ext. 2.1% S4: All NAD Maturies 2.0% NAD S2: Ext. 32.6% 1.9% Maturies 32.5% 1.8% S1: 1.7% S1: 32.4% Current Current 1.6% 1.5% S3: More 32.3% FX S3: More 1.4% 32.2% FX 1.3% 32.1% 1.2% 1.0 1.5 2.0 2.5 3.0 3.5 0.2 0.4 0.6 0.8 Risk: Deviation between Baseline Risk: Deviation between Baseline and Minimum Cost and Minimum Cost Source: Team calculations using MTDS Analytical Tool Table 8. Debt stock to GDP as at end 2017, Baseline and Shock Scenarios Current Extend Ma- More FX All NAD Shock Scenario (S1) turities (S2) (S3) (S4) Baseline scenario 32.4 32.7 32.2 32.7 Exchange rate shock (30%) 34.8 34.9 35.5 34.1 Interest rate shock 1 (200 bps) 33.2 33.3 32.8 33.5 Interest rate shock 2 (400 bps stress shock) 33.7 33.6 33.1 34.2 Combined shock 34.4 34.4 34.4 34.2 Maximum risk 2.5 2.2 3.3 1.5 Source: Team calculations using MTDS Analytical Tool Table 9. Interest payments to GDP as at end 2017, Baseline and Shock Scenarios Current Extend Ma- More FX All NAD Shock Scenario (S1) turities (S2) (S3) (S4) Baseline scenario 1.7 1.8 1.5 2.0 Exchange rate shock (30%) 1.8 1.9 1.6 2.1 Interest rate shock 1 (200 bps) 2.0 2.1 1.8 2.3 Interest rate shock 2 (400 bps stress shock) 2.1 2.2 1.9 2.6 Combined shock 2.1 2.2 1.9 2.4 Maximum risk 0.6 0.4 0.4 0.7 Source: Team Calculations using MTDS Analytical Tool 32 It is worth noting that the debt/GDP indicator typically better captures foreign curency risk while inter- est/GDP better captures interest rate risk. 20 Box 2. Strategic Benchmarks Namibia has a set of strategies to address the risks. The original benchmarks from the 2005 SDMS have evolved over time, see the table below. Benchmark 2011–12 2012–13 2013–14 2014–15 Actual FX Risk External Debt 20% 30% 31% 29% 29% Stock (including rand) / Total Refinancing Debt falling due 30% 39% 32% 30% 30% Risk within 12 months Long term debt 70% 61% 68% 70% 70% 2005 SDMS FX Risk External Debt 5 years 3.9 years Stock / Total (***) (*) Refinancing Average time to 5 years 3.9 years Risk maturity (*) Interest rate Fixed interest share 90% 98.1% risk (**) Source: 2005 SDMS, CDM Presentation (*) MTDS analytical tool based on central government debt data as of October 2012 (**) T-bills are considered as fixed rate debt although they carry the characteristics of floating rate debt. If defined as floating, the rate would have been 65.1% (***) Implied by the external debt to GDP and domestic debt to GDP limits. The existing strategic benchmarks do not capture the different risks that the government intends to manage. For instance, the fixed/floating share benchmark can conceal the exposure to interest rate risk (depending on whether the T-bills are defined as fixed and on the maturity of the debt stock). Indeed, Namibia’s debt stock is vulnerable to interest rate fluctuations even though the debt stock is almost completely in fixed rates. A better benchmark would have been the share of debt to be refixed within 1 year complemented by the average time to refixing. The 2005 SDMS’s foreign currency risk benchmark is to keep external debt—including the rand- denominated debt—below 20 percent of total debt. As a member of the CMA, Namibia’s rand debt does not carry any foreign exchange risk and could be excluded from this benchmark to provide a cleaner risk benchmark. It should be noted that the strategic benchmarks have been exceeded over time, which raises questions about their utility. More realistic but binding benchmarks should be defined and monitored. Although they cannot be achieved in the short term, as it takes time to obtain the desired debt composition, progress towards the strategic benchmarks is possible. Any deviation should be explained in detail in an annual debt management report. As a concrete example, the analysis shows that the ATM of 5 years can only be achieved by shifting the T- bills to longer dated bonds, which is hard to accomplish, or increasing external debt above the 20/80 benchmark, which is already exceeded. 59. This analysis of shock scenarios and alternative strategies suggests that the “Extend- ing maturities in the local market (S2)� strategy best corresponds to the GRN’s priorities of reducing interest rate and refinancing risks. It also helps to deepen the local market, which would in the medium to long term help bring the yields down and contain foreign exchange risk. 21 VI. Contingent Liabilities Stemming from Government Guarantees 60. Contingent liabilities, especially those arising from guarantees and on-lending activi- ties, have important implications for government debt management and sustainability. Guar- antees issued by the GRN have been declining over the past four years, particularly guarantees for foreign loans (see Figure 10). Namibia’s loan guarantees for both foreign and domestic coverage are currently relatively small at N$2.12 billion (equal to 2.1 percent of GDP).33 This falls well below the 2005 SDMS threshold of 10 percent of GDP. Nevertheless, the ambitious investment in public infrastructure implied by NDP4—estimated by NPC at N$187 billion over five years—part of which seems likely to be undertaken by SOEs, and growing interest in structuring development projects as PPPs suggest a need for greater scru- tiny by MOF in contingent liabilities. 34 Figure 10. Publicly Guaranteed Debt Stock, 2008–2012 4.0 6.0% Billions of Namibian dollars 3.5 5.0% Share of GDP 3.0 4.0% 2.5 2.0 3.0% 1.5 2.0% 1.0 1.0% 0.5 0.0 0.0% 2008q1 2008q3 2009q1 2009q3 2010q1 2010q3 2011q1 2011q3 2012q1 2012q3 Domestic Guarantees Foreign Guarantees Total Guarantees/GDP (right-hand axis) Source: Bank of Namibia 61. The State Finance Act of 1991 makes MOF responsible for issuing and managing the government guarantees in Namibia, and the 2005 Sovereign Debt Management Strategy doc- ument describes a governance framework for the issuance and management of the govern- ment guarantees. These guidelines reported are not always followed in practice, they omit provisions that are important for managing risks of contingent liabilities, and they lack the force of law. This section of the report provides general principles in line with sound practice to ensure standardized approach could be of some guidance to CDM and the MOF. 35 33 Bank of Namibia Annual Report 2012. Data are as of December 2012. 34 It should be noted that managing contingent liabilities arising from loan guarantees is only one small dimen- sion of how the central government interact with state-owned enterprises (SOEs). The suggestions for enhanced risk management of loan guarantees presented in this section are no substitute for a reform of the broader SOE governance framework. 35 For more guidance and support, there is need to assess the situation more in depth and develop a road map specifying the different steps for developing the guarantees framework as well as the timeline for the implemen- tation. 22 A. Overall Governance Framework 62. The 2005 SDMS established rules governing when the Minister of Finance may pro- vide guarantees to or enter into on-lending arrangements with Namibian companies. 36 The BON-MOF-NPC Inter-Agency Committee reviews all proposals. Criteria for evaluating pro- posals are the following: • There is a clear intervention rationale. • Existing development finance institutions (e.g., the Development Bank of Namib- ia) are unable to intervene. • The project is economically viable. • The project is in the national interest. • The project does not pose liquidity and credit risk exposure to the government. 63. MOF charges borrowers an annual levy on all non-statutory guaranteed and on-lent loans, set at 2 percent of the outstanding loan value, as an insurance payment against possible default. 64. In practice, decisions are taken on a case-by-case basis by CDM and the Minister of Finance, typically under pressing time constraints. Some evaluation criteria fall outside MOF’s area of expertise (e.g., project viability and national interest). Equally important, the 2005 SDMS does not contain criteria for quantifying credit risk, principles on disclosing credit risk, or guidelines for managing contingent liabilities—all of which are important for a debt management office to consider when evaluating requests for guarantees. 65. An alternative to the present governance framework would be for the executive level of the Inter-Agency Committee to define the broad principles of issuing guarantees, but dis- tribute responsibility for implementation to the respective institutions and to line ministries, according to their areas of expertise. For example, NPC would be responsible for evaluating whether a project is in the national interest, the relevant line ministry would assess technical viability, and MOF would concentrate on issuing the guarantees and managing the credit risk of the guaranteed debt portfolio. B. Issuance rules and procedures by the MOF 66. The guarantee sub-unit of the CDM Division consists of two staff, who take responsi- bility for carrying out all the procedures for the approval and issuance of the guarantees, re- cording all financial transactions related to guarantees, monitoring the progress of the project through lenders and borrowers, and collecting annual audited financial reports to help moni- tor project’s performance and minimize default risks. The BON provides assistance to the CDM team in conducting assessments of guarantee operations. A draft internal Ministry of Finance document contains clear and detailed guidelines for issuing new guarantees and managing the contingent liabilities arising from them. According to this document, applica- tions must include audited financial statements for at least five years, information about se- cured and unsecured debts, a medium-term forecast of the financial position, and information about how a government guarantee would assist the project. The extent to which these guide- 36 These procedures are presented in section 5.1 of the SDMS. Guarantees may be provided to private compa- nies owned by Namibian citizens as well as parastatals. Joint ventures are eligible, but only for exposure in pro- portion to Namibians’ equity participation. 23 lines are followed or how the collected documents are evaluated to assess the eligibility of the project for a guarantee was not made fully clear to the team. 67. CDM could revise the rules and procedures for clarity, simplicity and applicability. The procedures could define: • who can apply for the guarantee, e.g., parastatal or line ministry, which level company representative, etc., • the timing of the application, service standards for processing by MOF, and • which documents and approvals must to be enclosed with the application (e.g., let- ter from NPC specifying the project is in the national interest, letter from the line ministry stating the project is technically and environmentally viable, letter of un- dertaking etc.). 68. MOF could also develop a template for the financial statement requested at the time of application. The template would help standardize the information provided by the appli- cants and facilitate the analysis CDM would conduct to assess the credit risk arising from the transaction. In order to fulfill the tasks assigned to MOF, CDM’s staff capacity to assess and manage credit risk of loan guarantees should be increased. C. Risk Management 69. The framework for managing risks from contingent liabilities can be significantly im- proved. Currently the GRN manages its exposure from government guarantees through a lim- it set for the stock of guaranteed debt (10 percent of GDP) and a guarantee fee (2 percent of the value of outstanding loans). The guarantee levy is paid annually by the beneficiary on the amount of principal outstanding. This fee is sometimes waived on a case-by case basis. Par- tial loan guarantees are among the measures to contain the credit risk. There are currently no guidelines on how to assess the credit risk of guarantees. Thus no information on the risk ex- posure from an individual guarantee request is provided to the finance minister at the time of issuance. The overall exposure of the contingent liabilities is not monitored. 70. An efficient risk management framework would have two stages. 37 In the first stage— at the time of issuing the guarantees—risks are mitigated by avoiding risky counterparties to limit total exposure when selecting beneficiaries. Financial feasibility of the project, financial strength of the beneficiary, debt level of the beneficiary and re-payment performance of the beneficiary in the previous projects can be considered as the examples for plausible criteria. More specifically, MOF could choose not to issue guarantees to the beneficiaries that have arrears to the MOF. 71. In the second stage, i.e., once the guarantee are provided, credit risk management mechanisms such as guarantee fees, redemption accounts and budget appropriations are set for potential defaults. These measures require risk managers to make comprehensive risk as- sessments. 72. As noted earlier, the 2005 SDMS contains no guidelines for assessing credit risk. Credit risk is defined by the probability of default of the guarantee beneficiary and the amount that might be undertaken in case of default. The probability of default could be fore- 37 Colombia, Poland and Turkey are examples of countries that follow this two-step procedure. 24 casted through various methods and a good starting point is to explore the default history. The forecasting could be improved through the analysis of the financial statements of the beneficiaries, which could give an indication of financial difficulty. There is no one method, and the essential point is to use a consistent approach across the board. Some countries de- velop advanced credit risk management models and allocate internal ratings as a way to summarize the riskiness of the beneficiaries. As a start, Namibia could adopt a simple rating framework to differentiate among entities based on their riskiness. The ratings could be a use- ful source of information when deciding on a guarantee issuance and to assess the overall risks to the guarantee portfolio. D. Collection of Receivables 38 73. MOF signs a one-page letter of undertaking with the beneficiaries at the time of the guarantee issuance to ensure collection of the called debt in case of default. The letter is gen- eral and does not describe the details of how the recovery of the undertaken debt will be ac- complished. In practice, the collection process does not work efficiently, and some parastatals accumulate arrears towards the MOF. 74. Secondary legislation detailing the principles for the creation and collection of the re- ceivables could be developed. This would systematize the ways of collecting the called amounts such as restructuring the debt over a given period of time so that the beneficiary is able to pay. Restructured debt could carry an interest rate comparable to the government’s borrowing costs to reduce the impact on the government budget. E. Transparency and Accountability 75. Transparency and accountability are of utmost importance. This is especially relevant to contingent liabilities, given that in some countries guarantees represent concealed subsidies that are provided with no trace in the budget. It is critical that the principles, rules and proce- dures are publicly available, and that information on guaranteed debt and contingent liabili- ties is published. 76. Namibia publishes good statistics on new guarantee issuances and guaranteed debt stock (by beneficiaries and lenders) in the BON bulletins. This information could be com- plemented with the realized repayments (made by institution or undertaken by MOF) and re- payment projections, followed by reports that discuss the developments on the guarantees and risk assessment of contingent liabilities. In addition, reflection of these numbers in the budg- et—through an appropriation or as a memo item, for example—would promote transparency and accountability. 77. A second way to promote accountability is to enhance accountability of the guaran- tees framework is to embody it into secondary legislation. Publishing the criteria and rules for issuing guarantees in the SDMS is important, but this document is not binding and does not have the force of law. It would be very powerful to incorporate the new principles and rules for issuing guarantees into secondary legislation to make them enforceable. This would also transmit a clear signal that the government is committed to building and respecting a gov- 38 A restructuring of the accumulated debt of the guarantee beneficiaries towards the MOF might be needed be- fore developing rules on the collection of the undertaken debt. 25 ernment guarantee framework. Given the time it might take to prepare and enact the second- ary legislation, MOF could start implementing the new framework until it formally enters in- to action. This would allow seeing how the implementation works to give an opportunity for improvement in the final text of the secondary legislation. F. Public Private Partnerships 78. PPPs represent another source of contingent liability. The Namibian government con- templates to support the NDP4 goals amounting to N$187 billion with the participation of the PPPs. The amount to be financed from the budget and traditional financing is unknown yet as well as how the PPPs projects will be structured. 79. It is often the case that government guarantees are requested under PPP schemes. In addition to a sound and robust governance framework, MOF should carefully assess the cred- it risk from these new contingent liabilities which can amount to significantly high amounts given the scale of the PPP projects in general. Therefore, the credit risk evaluation and man- agement methods developed by the MOF for the loan guarantees would set the basis for deal- ing with the contingent liabilities due to PPP projects. VII. Domestic Debt Market Development 80. The feasibility of alternative debt management strategies is closely linked to the de- gree of development of the domestic debt market. A deep and liquid bond market facilitates the implementation of a debt management strategy and broadens the set of potential funding strategies. This section describes constraints posed by the size and structure of the govern- ment debt market in Namibia, and it offers suggestions for increasing the liquidity and depth of the market. Figure 11. Relative Sizes of Bond Markets, 2011 Outstanding Bonds as share of GDP by Issuer Thailand South Africa Peru Namibia Mexico Malaysia Hungary Croatia Chile 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% Government Financial Corporations Non-Financial Corporations Sources: Namibian data from IJG and BON; for other countries, values of bond issues are from Bank for Inter- national Settlements and GDP values from World Development Indicators Notes: Government bonds exclude bills with maturities of one year or shorter 26 81. The debt market in Namibia is small and dominated by GRN bonds and bills. Even including corporate bonds, total domestic bond market capitalization in Namibia as a share of GDP is roughly one-quarter to one-third the size of bond markets in Chile, Hungary, Mexico, or South Africa, as shown in Figure 11. 82. Total debt issued equals N$20.4 billion, equivalent to 20 percent of GDP (Table 10). Government securities have been described earlier. There are currently 19 bonds issued by the corporate sector: 11 by three of the four commercial banks, and eight by state-owned en- terprises. 39 Table 10. Domestic Debt Securities Traded in Namibia, 2001 Bonds Value Share of GDP Government (Treasury bonds) 9,346 9% Corporate Banks 1,633 2% Non-financial 1,428 1% Total corporate 3,061 3% Total bonds 12,407 12% Treasury bills 8,042 8% Total debt securities 20,449 20% Source: Bank of Namibia, IJG Securities, January 2013 Notes: Values in millions of Namibian dollars 83. Not only are there few issuers of debt, there are also are few buyers—primarily the commercial banks and the GIPF, which appears to hold the majority of government debt. 40 This exposes the government qua borrower to demand side risk. 41 Combined with the limited supply of financial assets, the legal requirement that pension funds and long-term insurers invest a minimum of 35 percent of their assets in Namibia means that there are more buyers than sellers in the market for debt. The combination of these factors results in a hold-to- maturity-behavior from all relevant participants and, because of that, very few bond trades take place, as shown in Figure 12. Figure 12. Secondary Bond Market Trading, 2005–2012 250 2,000 (millions of Namibian Number of Trades 200 Value Traded 1,500 dollars) 150 1,000 100 500 50 0 - 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Government Corporate Total Trades (number) Source: IJG Securities, January 2013 39 Banks placing bonds in the Namibian market include Bank Windhoek, FNB, and Standard Bank. The Road Fund Administration, Nampower and Telecom Namibia represent the non-financial corporate bond issuers. Is- suances range from N$10 million to N$500 million in size. Original maturities range from 2.2 to 12.3 years. SOE bonds average slightly larger (N$179 vs N$148 million) and longer (8.0 vs 5.4 years). 40 From evidence the mission collected, GIPF’s share of outstanding bonds could be as high as 90 percent. 41 A more diversified investor base is the key to reducing this source of risk. 27 84. Development of the market for government debt is complicated by several other fac- tors, including the number of different series of bonds offered in the same auction, poor price discovery, the limited dialogue between the MOF and the market, the still not dematerialized bonds, the lack of information on the investor base for government debt. 85. The GRN has introduced measures to improve the primary market. The government recently started using Bloomberg platform to conduct the auctions and has published a bond page in Bloomberg with indicative pricing. These measures have been well received by the market and help fostering price discovery. 86. There is room to improve primary issuance, however. The government currently of- fers up to eight different series of bonds at most auctions. This does not help to create liquidi- ty for them, especially given the small size of the total market. Best practices suggest reducing the number of current on-the-run benchmark bonds. Additionally, the government could try to smooth the total borrowing over the year in order to have a more homogeneous volume of bonds offered in the different auctions. 87. New regulations could improve the poor price discovery in the Namibian market. Alt- hough the central bank and some market participants regularly publish yield curves, this does not necessary reflect actual trades in the market. Market participants, except for the brokers, are not obliged to report their trades. Together with the structural factor of the shallowness of the market, this results in poor and scarce information about actual trades. Requiring every market participant to report its trades and publishing these at the end of day would improve (ex-post) price discovery and transparency in the market. 88. The market would also benefit from increased dialogue with MOF. Market partici- pants seem to be hungry for information from and regular communication with the issuer. Good communication with the market helps to reduce uncertainty, and through that the cost of borrowing, and helps the government to better understand the demand side, which is essen- tial to the design of a debt management strategy as well as to understanding the necessary measures to further develop the local market. MOF could set up an agenda of regular meet- ings with market participants and publish more debt management documents, such as the an- nual borrowing plan and semi-annual auction calendars. The DMO’s front office should also tighten its communication with the market, as this would provide important feedback from the market regarding risk appetite and demand for the auctions. 89. In broader terms, the establishment of a formal forum to discuss the development of the capital markets could bring benefits to debt management and spillover effects for the whole economy. MOF and BON could formally create working group with representatives from the private and public sectors to discuss development of the private bond market and to plan concrete short- and medium-term measures to develop the government debt market. 90. Generating better information about the current investor base would improve under- standing of the dynamics of the debt market. Information about the actual holders of the gov- ernment securities is not available. Understanding of the actual investor base is a necessary step to the design of a medium term debt management strategy and measures to develop the domestic market. 28 91. The size and nature of the biggest player in the market, the GIPF, is challenging to deepen the domestic debt market. The GIPF holds the vast majority of government bonds. As a pension fund, the GIPF is a natural buy-and-hold investor. Together with its large size, this limits deepening of the market. 92. Meanwhile, some measures such as the promotion of a securities lending market could have positive effects. 93. Enhancing the infrastructure of the market would help broaden the investor base. The 2011 issuance of a Eurobond and the recent issuance in the South African market clearly show that there is foreign demand for Namibian government bonds. Attracting these investors to the local market could be a way to develop the domestic market while broadening the in- vestor base. In fact, the government could benefit more from the proximity to the South Afri- can market, which is a large market with relevant presence of international investors. But, in order to do that, it is necessary to put in place an adequate infrastructure. • The fact that bonds are not dematerialized (i.e., they are issued and traded in paper form) prevents non-residents from investing directly in the Namibian market for government debt. • Creating a central securities depository (CSD) is an important step to develop the bond market. It would provide an appropriate infrastructure to the bond market, providing more confidence to investors (especially South African and other for- eigners). Having a single CSD could also facilitate the monitoring of the trades (price discovery) and the holders of the bonds • There is no derivatives market, which typically adds liquidity to the bond market. 94. The above mentioned fact that bonds are not dematerialized is not only important for broadening the investor base, but might be one of the main bottlenecks for market develop- ment in the short term. Other measures will not have positive effects if bonds are still materi- alized and the overall market infrastructure is poor. A manual and certificated process is very unlikely to promote market liquidity. 95. Additionally, the current regulation does not provide a sound legal environment to the development of the repo market and the launch of a primary dealers program, one of the re- cent attempts of the government, requires changes in the current legal framework, namely the Insolvency Act.42 On its turn, an active repo market can play a significant role in the devel- opment of the bond market, encouraging commercial banks to invest more in government se- curities. 96. The government faces trade-offs when issuing new instruments. The government has been analyzing the possibility of issuing inflation-linked bonds and launching a retail debt program. Although these instruments might reach new investors and consequently broaden the investor base, because the size of the market and other restrictions mentioned before, add- ing new instruments might mean draining liquidity instead of adding it to the market. 97. Finally, active liability management operations could help the government to cope with refinancing risk and exchange old securities for standardized ones. While the concentra- 42 Because of the size of the market and the reduced number of players, a careful analysis should be done about the costs and benefits of putting a primary dealer program in place. 29 tion of issuances in fewer benchmark bonds helps to improve liquidity in the secondary mar- kets, it also represents an increase in the refinancing risk because of the higher volumes in single maturity dates. A good way to deal with this potential negative impact is to implement active liability management operations, such as switches or buybacks. Switches and buybacks could also be used by the government to exchange old bonds that are still in the market for the new standardized bonds, what would help to foster the process of providing more liquidi- ty to these bonds. VIII. Legal, Institutional and Implementation Issues A. Legal Framework 98. Namibia’s overall long-term debt management objectives and strategy, along with policy level decisions, are defined by MOF and approved by the cabinet. The implementation of the strategy is achieved by the technical-level entities at MOF and BON. 99. Namibian legislation fulfills minimum requirements related to the existence, coverage and content of the legal framework for debt management. In the future, when the legislation is revised, it is recommended that clear debt management objectives and requirements for au- dits and reporting to parliament be included. B. Institutional Arrangements 100. The managerial structure for government related borrowings and debt-related transac- tions, as well as to issue guarantees and on-lending are well established. MOF has the overall responsibility for the preparation and implementation of debt management strategy and coor- dination with other entities, especially the BON. 101. The clear separation of front-, middle- and back-office functions within the debt man- agement office as outlined in the 2005 SDMS have not yet been implemented. The back of- fice is need revamping including training more staff, as there is only one staff conversant with the Commonwealth Secretariat Debt Recording and Management System (CS-DRMS) used to manage external debt. Likewise, there is a need to procuring a secure server to house information on all foreign and domestic loans. Currently, the external debt is recorded on a single workstation. C. Annual Borrowing Plan 102. The desired debt composition in a medium-term debt strategy provides the basis for the annual borrowing plan, through which the government seeks to find an appropriate bal- ance between meeting debt management objectives, increasing spending on new priorities (including development projects), and developing the domestic financial markets. In Namib- ia, MOF and BON jointly determine the annual borrowing plan, taking into account medium- term expenditure framework, the capacity of the local market, interest and exchange rate forecasts, and monetary and foreign reserve developments. This collaboration is expected to continue when developing the annual borrowing plans for the coming (2013/14) fiscal year and under the GRN’s revised SDMS. The domestic borrowing plan is translated into the auc- tion calendar for government securities, which BON posts on its website. 30 103. The IMF-World Bank MTDS framework provides a number of guidelines that could be considered when implementing the new SDMS. • The annual borrowing plan should be informed by updated analysis of risks and costs, as described above. • In terms of external debt, the plan set out the possible timeframe to access the in- ternational capital market. • Include purposes of borrowing in the annual borrowing plan • Publish the annual borrowing plan • Include proposed amounts in the auction calendar • The calendar should be well structured and (a) avoid crowded issuance dates so issuances won’t compete with each other; (b) take into account government cash position and budget outturns in the fiscal year. D. Implementing, Disseminating, Monitoring and Reviewing the Debt Strategy 104. Namibia’s revised SDMS should incorporate a comprehensive framework for regular monitoring and review. Furthermore, although debt strategies are developed for a medium- term horizon, they should be updated annually. If there are significant deviations in the out- turns relative to targets or in the assumptions underlying the strategy, it should be revised ac- cordingly. There is also the need for active investor-relations and market consultation to get up to date information on the market (see Section VII). This will help determine a prior the investor appetite for the various instruments before it is done 105. Statistical reporting: A first step is to release regular debt portfolio reports that in- clude key cost and risk indicators. BON reports public debt stocks and flows in its quarterly bulletins and annual reports. MOF’s annual budget documents report levels of central gov- ernment debt, spending on debt, and sources of financing that will be used to meet the gov- ernment’s borrowing needs. The FY2012/13–2014/15 MTEF document announces an intention to begin publishing a debt statistical bulletin on an annual/quarterly basis in the fu- ture. This would make an important contribution to increasing transparency in public finan- cial management, increase accountability, and contribute to strengthening the development of Namibian capital markets. In addition information in existing BON and MOF reports, a good debt statistical bulletin would report cost and risk indicators such as those used in the analysis above and presented in Table 3 on page 6. It is also important to report data on contingent liabilities, especially arriving from guarantees and on-lending activities, as these have impli- cations on debt sustainability levels and debt management. 106. Annual review: Besides expanding statistical reporting, the government should regu- larly review performance of the SDMS, comparing outcomes against the debt management objectives and assessing whether underlying assumptions remain valid. MOF’s annual budget documents should include a specific section on debt management for this purpose. Reporting annually on debt to Parliament would enhance the overall transparency of debt management operations and ensure accountability. 31 IX. Conclusions 107. The 2005 SDMS was good. Combined with prudent fiscal policies, the SDMS has helped Namibia manage public debt effectively. How could it be improved? 108. Crunch the numbers. The fundamental message that emerges from this review is that an explicit analysis of costs and risks of alternative debt strategies in the face of plausible economic shocks should inform the revised SDMS. Applying the MTDS Analytical Tool to several possible debt strategies in the face of plausible economic shocks suggests that a strat- egy that emphasizes longer maturities in the domestic market performs better against the GRN’s risk priorities than does the current strategy. The MTDS Analytical Tool also illus- trates the trade-offs between relying more on external versus domestic markets for financing, which are important to consider when preparing the new SDMS: A strategy of borrowing more aggressively from international sources reduces borrowing costs but increases exposure to the risk that these costs could rise unexpectedly, while the domestic bond market provides financing with lower risks but higher costs.43 109. Repeat analysis frequently. A related message is that it is essential to repeat this for- mal analysis regularly to ensure that the SDMS remains consistent with economic conditions and the GRN’s policy priorities. As a small and relatively open economy, Namibia is contin- ually exposed both to adverse shocks but also to new opportunities that influence the ability to meet debt management objectives in the SDMS. The analysis undertaken during the No- vember 2012 capacity building mission should be repeated by the working group revising the SDMS, using the latest macroeconomic and fiscal policy framework. In the future, it would be appropriate to update the debt management strategy annually. 110. Rationalize guarantee issuance. Although the stock of debt guaranteed by the GRN is relatively small, it seems likely to grow in the future, and reforming the framework for is- suing government guarantees would help the GRN reduce its exposure to contingent liabili- ties. One priority is for MOF to introduce a system for analyzing the credit risk of all new requests. A second priority is to reallocate responsibilities for reviewing requests, assigning to the relevant line ministry the tasks of evaluating project feasibility and to NPC the decision on whether a project is in the national interest. 111. Improve bond market institutional infrastructure. Dematerializing government bonds, establishing a central securities depository, and requiring that all bond trades be re- ported and published would enhance debt market efficiency through improved price discov- ery and reduced transactions costs. The 2005 SDMS identified several legal and regulatory reforms that are prerequisites to undertaking these measures. 112. Increase transparency and accountability. Publishing the borrowing plan, increasing information contained in the auction calendar, circulating a regular debt bulletin containing comprehensive information on public and publicly guaranteed debt, and releasing an annual evaluation of the SDMS would contribute to greater transparency and accountability. Incor- 43 For example, the MTDS Analytical Tool provides one means for the authorities to assess how well the 80/20 benchmark and other rules correspond to their risk preferences and cost constraints in the face of plausible shock scenarios. 32 porating reporting provisions into secondary legislation and submitting the debt strategy to parliamentary review would also strengthen accountability. 33 X. Annexes Annex 1: Description of the MDTS Analytical Tool 44 The MTDS Analytical Tool (AT) is an integral part of the MTDS Toolkit, developed by staff of the International Monetary Fund and the World Bank to provide a quantitative analysis as input to the MTDS decision-making process. The AT is a spreadsheet-based application that allows projecting cash flows as a function of the following data: • existing debt • macroeconomic assumptions, i.e. the primary balance • new borrowing strategies • financial variables, including interest rates and exchange rates. The tool then simulates different cash-flows under various scenarios. The output of the tool is a quantification of the costs and risks associated with a particular debt management strategy. The AT facilitates the quantification of costs and risks for each strategy under consideration. By illustrating the consequences of following a particular strategy under various scenarios for macroeconomic and market variables, it gives insight into the key vulnerabilities embedded in the specific strategy under consideration. The output, generated by the AT is a number of cost and risk indicators, for example annual interest payment-to- GDP and the nominal stock of debt-to-GDP. Risk is measured in terms of the increase in cost, given a particular macro and market scenario, relative to the baseline. The AT different cost and risk indicators, allow countries to focus on those measures most relevant for their needs. The AT has been designed to show the details of all cash flow calculations at every step of the process. Intermediate cash flows as well as spreadsheet functions are explicitly shown at every stage, allowing the user to track the assumptions underlying the analysis. Thus, the AT is not only useful for the quantitative analysis underlying a debt strategy, but is also a useful device for building capacity in the debt office. Finally, once the desired debt management strategy is implemented, the AT can be used to measure adherence to the strategy, and reevaluate the cost-risk alternatives should there be a change in market conditions or the au- thority’s risk preference. 44 The following is taken from World Bank and IMF, “Developing a Medium-Term Debt Management Strategy: The Analytical Tool,� May 2009. 34 Annex 2: External Debt Table 11. Foreign Loans Outstanding, 2012 Interest Year Original Remaining Creditor Original Amount Rate Issued Maturity Principal People’s Republic of China RMB100,000,000 Free 1991 15 RMB 65,595,879 African Development Fund JPY93,221,240 0.75% 1992 50 JPY 82,500,685 African Development Fund USD1,689,708 0.75% 1992 50 USD 1,529,186 African Development Fund CHF327,857 0.75% 1992 50 CHF 295,072 African Development Fund EUR1,736,949 0.75% 1992 50 EUR 1,563,254 African Development Fund USD2,718,217 0.75% 1993 50 USD 2,473,578 African Development Fund CHF3,293,126 0.75% 1993 50 CHF 2,996,745 African Development Fund JPY254,938,990 0.75% 1993 50 JPY 231,994,487 African Development Fund EUR1,331,324 0.75% 1993 50 EUR 1,061,324 African Development Fund EUR3,119,185 0.75% 1993 50 EUR 1,953,951 African Development Fund EUR271,656 0.75% 1993 50 EUR 247,207 Kreditanstalt für Wiederaufbau EUR3,269,180 2.00% 1993 21 EUR 1,753,624 European Investment Bank EUR3,000,000 2.00% 1994 20 EUR 453,300 Kreditanstalt für Wiederaufbau EUR12,424,393 2.00% 1994 21 EUR 7,276,706 Nordic Investment Bank EUR1,971,765 0.75% 1994 30 EUR 1,695,718 Nigerian Trust Fund USD4,150,480 0.75% 1995 50 USD 1,686,229 Arab Bank for Econ.Dvlp.t USD1,150,000 3.00% 1996 15 USD 355,000 Kreditanstalt für Wiederaufbau EUR14,316,173 0.75% 1996 30 EUR 11,457,029 Kreditanstalt für Wiederaufbau EUR2,045,168 2.00% 1996 21 EUR 1,382,168 Kreditanstalt für Wiederaufbau EUR1,417,047 2.00% 1996 31 EUR 1,118,047 Kreditanstalt für Wiederaufbau EUR8,129,541 0.75% 1996 32 EUR 6,984,247 Kreditanstalt für Wiederaufbau EUR7,209,216 2.00% 1996 20 EUR 5,229,493 Arab Bank for Econ.Dvlp.t USD2,937,049 3.00% 1997 20 USD 1,300,653 Kreditanstalt für Wiederaufbau EUR5,777,598 2.00% 1997 21 EUR 4,230,940 Kreditanstalt für Wiederaufbau EUR9,561,158 2.00% 1997 21 EUR 6,769,474 European Investment Bank EUR2,500,000 3.00% 1999 15 EUR 1,304,750 European Investment Bank EUR10,500,000 3.00% 1999 15 EUR 5,211,150 Kreditanstalt für Wiederaufbau EUR8,896,479 2.00% 1999 20 EUR 7,339,595 Kreditanstalt für Wiederaufbau EUR6,135,503 2.00% 1999 20 EUR 5,675,340 Kreditanstalt für Wiederaufbau EUR7,672,466 2.00% 1999 21 EUR 6,736,802 European Investment Bank EUR4,000,000 3.00% 2000 9 EUR 529,954 Eksportfinans ASA EUR6,906,225 Free 2000 10 EUR 345,311 Kuwait Fund for Arab Ec. Dev KWD5,162,115 3.00% 2000 17 KWD 2,860,465 African Development Bank ZAR202,420,718 JIBAR 2001 15 ZAR 121,452,431 Arab Bank for Econ.Dvlp.t USD7,600,000 3.00% 2001 20 USD 6,102,000 Institituto de Credito Official EUR6,328,532 0.35% 2002 20 EUR 6,174,178 Kreditanstalt für Wiederaufbau EUR5,084,460 2.00% 2002 21 EUR 4,960,460 African Development Bank ZAR193,470,368 JIBAR 2003 15 ZAR 135,429,258 Arab Bank for Econ.Dvlp. USD6,525,129 3.00% 2003 20 USD 6,159,129 Banco Bilbao Vizcaya Argentaria EUR16,988,577 4.27% 2003 11 EUR 6,842,074 Institituto de Credito Official EUR16,988,578 0.40% 2003 20 EUR 16,563,864 African Development Bank ZAR246,760,000 JIBAR 2004 15 ZAR 189,182,667 Kreditanstalt für Wiederaufbau EUR883,767 0.75% 2005 30 EUR 883,767 BNP Paribas EUR13,427,212 Free 2006 12 EUR 12,309,194 Japan Int’l Cooperation Agency JPY10,091,000,000 0.90% 2006 11 JPY 8,649,414,000 Export-Import Bank of China RMB41,000,000 2.00% 2008 12 RMB 33,296,685 Export-Import Bank of China RMB300,000,000 2.00% 2008 15 RMB 238,206,981 Kreditanstalt für Wiederaufbau EUR7,045,000 2.00% 2008 20 EUR 4,045,000 BNP Paribas EUR31,861,702 Free 2011 12 EUR 31,861,702 Source: Ministry of Finance Notes: Remaining principal amounts as of October 2012. 35 Annex 3: Main Functions and Required Skills in a Debt Management Office The functions and required skills of the three main divisions of a debt management office (DMO)—namely, the front, middle and back offices—are described in this annex, as request- ed by MOF. In addition, there is a similar description for a legal unit of a DMO, which is sometimes included in one of the other offices. 1. FRONT OFFICE -PORTFOLIO MANAGEMENT UNIT The front office (FO) or portfolio management unit is responsible for the analysis and effi- cient execution of all portfolio transactions, consistent with the debt management policy and strategy. The degree of sophistication of its functions will largely depend on the types of funding available for that particular country. Typically there will be a learning curve as coun- tries advance from official sources (concessional and non-concessional multi-lateral and bi- lateral) to more market-based funding, both domestic and international. The learning curve continues as the DMO starts using debt transactions such as exchanges and buy-backs, as well as hedging transactions including derivatives such as currency and interest rate swaps. The FO may also be involved in transactions of on-lending to sub-national governments and extending guarantees of various types to other government entities and/or the private sector. In some countries it may also be involved in executing cash management operations. 1.1 Typical Front Office Functions • Designing and executing funding transactions, both from domestic and international sources of funding. This may involve an analysis of competing financial proposals across different currencies, markets, maturities and transaction structures. Steps can include the evaluation, negotiation, pricing, launch of bond issues/contracting of loans, and subsequent market monitoring • Designing and executing trading and hedging transactions, to move the actual debt portfolio closer to the strategic targets or benchmark (e.g. including debt buy-backs and exchanges, interest rate and currency swaps, etc.) • Continuous monitoring and reporting of market conditions, including analysis of its potential investor base, both domestic and international • Managing investor relations, possibly both domestic and international • Investing foreign currency liquidity and any excess cash balances associated with the government’s daily departmental cash management (in countries where the DMO is responsible for cash management) • Analyzing projections of funding needs (in partnership with the middle office and Treasury/fiscal unit) and giving input on implications for a funding strategy • Providing input on the design of a sovereign’s funding strategy (together with middle office) • Offering advice on government policy initiatives to foster the development of the primary and secondary government bond markets • Offering advice on possible market reaction to new fiscal information • Evaluating funding requests by SOEs and sub-nationals (relevant in some countries, and possibly in partnership with the middle office) which would result or not in an au- 36 thorization from the DMO for their contracting new debt; or possibly in on-lending by the central government • Evaluation of expected cost/pricing of contingent liabilities such as guarantees (e.g. by the central government to SOEs or sub-nationals) (in some countries and possibly in partnership with the middle office, sometimes the middle office takes the lead). 1.2 Front Office Skills Required The mix of skills will be important, and will differ according to the level of development of the country and the markets being sourced for funding and debt management transactions. The degree of training required will depend on whether the new staff already has the required type and level of skills, or whether it is necessary to build up some or most of these modules: • Advanced finance and financial markets know-how (money markets, international and domestic debt markets, primary and secondary markets, main players, etc.) • Knowledge of non-market sources of funding, e.g. IFI’s and their funding characteris- tics • General macroeconomic notions o In particular, links of debt management with monetary policy, cash manage- ment and fiscal policy • Knowledge of the role of a modern debt management office and within it, the func- tions and responsibilities of the three main divisions • Public policy skills (e.g. to carefully manage relations with the markets in a suitable context, and understanding risk-taking in a public sector context). • Strong communications skills (e.g. for managing investor relations, negotiations, etc.) More precisely, knowledge of: • Analysis, pricing and execution of : o funding transactions ( domestic and international; capital markets, non-capital markets) o debt management related transactions, such as derivatives including currency and interest rate swaps, debt exchanges (e.g. by maturity,) buy-backs of illiq- uid bonds, etc. o guarantees extended by central government • Conceptual framework for portfolio and risk management, the role of strategic targets and/or benchmarks in public debt management, and the role of the front office in complying with a benchmark • Monitoring of, and reporting on, different financial markets and market participants, both domestic and international; (including knowledge of Reuters, Bloomberg, etc.) • Relationship management with market intermediaries and investors • Public finance and cash management and how to determine the government’s funding needs; • Domestic government money market and bond markets, so as to be able to provide advice on policy initiatives to foster the development of the primary and secondary government markets 37 • Investment of foreign currency liquidity and excess cash balances associated with dai- ly cash management (in those countries where the DMO is also in charge of cash management). • Pricing of guarantees and other contingent liabilities, and methodologies for evaluat- ing expected cost (with middle office staff) • For some offices, evaluation of funding requests by public entities such as SOEs and sub-nationals, so as to provide them with the corresponding authorization (with mid- dle office staff) • English language training, if transacting in the international markets • Specialized IT training (related to pricing of transactions, market monitoring and re- porting, practical knowledge of Bloomberg, Reuters, etc.) 2. MIDDLE OFFICE – ANALYSIS AND COMPLIANCE UNIT The core competence of the middle office (MO) is the design of a public debt strategy, for final authorization of senior government authorities, which will involve risk/cost modeling and an analysis of macroeconomic and market constraints. Another important but more oper- ational function is monitoring and compliance. In those countries where the DMO is also in- volved in cash management the MO will participate in proposing a strategy for managing the portfolio of cash surpluses. 2.1 Typical Middle Office Functions Middle office (MO) functions typically include: • Risk modeling of the aggregate debt portfolio (initially deterministic scenario analysis and subsequently stochastic simulations) • Analysis of potential constraints on debt portfolio management (macroeconomic, fi- nancial market, etc.) and their influence on debt strategy • Debt strategy formulation and design of strategic targets and/or benchmarks • In some DMOs, acting as Secretariat to a Debt Management Committee which advis- es the Finance Minister on debt management strategy, both on design and monitoring of execution. Many MO’s in debt offices also handle one or more of the following areas: • Monitoring compliance with the established portfolio and risk management policies including regular reports monitoring market and credit risk • Performance assessment (if there is active trading vis-à-vis the strategic targets,) • Preparation of input to the state budget (reports on debt servicing forecasts, etc.) • Operational risk control (together with back office) • Producing reports on debt management for different parties (e.g. the finance minister, the debt management committee, legislature, multilaterals), etc. and providing input on debt management to the corresponding website (working closely with the Back Of- fice) • New product development (together with front office) • Debt sustainability analysis, in conjunction with other government units and the cen- tral bank (ideally this would be carried out by the Fiscal Policy Unit, not the DMO) • For some DMOs, the MOF may be involved in analyzing cost risk trade-offs and de- signing a strategy for the management of cash surpluses. 38 2.2 Middle Office skills required The skills requirements listed below are needed to create a strong Middle Office function. The mix of skills will be very important for the success of a middle office, and will differ ac- cording to the level of development of the country, the ambition of the Ministry of Finance for the development of this area and the level of risk exposure of the government to different types of financial risk. In general terms there is a need for the following: • Advanced technical skills in finance and risk analysis, in particular, risk quantification and portfolio management • Financial market skills; given the impact from debt management on debt market de- velopment, an understanding of the workings of the local markets is also important • Experience in managing the types of market, credit and operational risks associated with the middle office’s responsibilities. • Public policy skills – an understanding of the role of debt management within the con- text of the overall macroeconomic policies. Preferably some experience in public sec- tor financial management and some macroeconomic knowledge for analyzing integration with the rest of the economy • Strong mathematical and modeling skills • IT skills are necessary for quantification of cost and risk of the government’s debt; at a minimum, advanced knowledge in the use of spreadsheets, but ideally strong skills in different analytical software packages. For many countries, a practical knowledge of Bloomberg/Reuters is required. • Strong communication skills – ability to translate analysis into material than can be the basis for the Minister’s decisions on debt management More specifically, good working knowledge of: • Debt risk indicators and reporting on debt portfolio profile • Risk modeling of debt servicing cash flows; scenario analysis, stochastic analysis • Debt strategy design, and the influence of other considerations apart from cost/risk tradeoffs, on the debt management strategy (e.g. the need to develop the domestic money and debt markets) • ALM framework for public debt management (and possibly cash management) • How to monitor compliance with the established portfolio and risk management poli- cies including regular reports monitoring market and credit risk • Performance assessment (if there is active trading vis-à-vis the strategic targets) • Preparation of input for the Government budget (reports on debt servicing forecasts, etc.) • Operational risk control (together with back office) • Producing reports on debt management for different parties (e.g. the Finance Minister, the Debt Management Committee, Congress, multilaterals), etc. and providing input on debt management to the corresponding website (working closely with the Back Of- fice) • New product development • Debt Sustainability Analysis, in conjunction with other government units and the cen- tral bank (ideally this would be carried out by the Fiscal Policy Unit, not the DMO) 39 • For some DMOs, the MOF may be involved in analyzing cost risk trade-offs and de- signing a strategy for the management of cash surpluses. 3. THE BACK OFFICE The core competence of the back office is operational, involving transaction confirmation, settlements, reconciliation and payments, as well as maintaining records of new contracts, disbursements, payments, debt restructuring and on-lending. In some countries the DMO may be requested to have a register of debt of sub-national entities. 3.1 Typical Back Office Functions The typical responsibilities of the back office (BO) are: • Confirmation of the transactions undertaken by the front office, i.e. independently verifying with the counterparty’s back office that the terms of the transaction are as the front office stated; • Settlement of the transactions once they have been confirmed, i.e. issuing/receiving payment instructions to/from counterparties; • Reconciliation of bank and custody accounts to ensure that they agree with the organ- ization’s own records. In addition, there are usually a number of administrative functions that are undertaken by the BO, including: • debt registration and management of the debt data base (which may include not only the central government debt, but also debt of sub-nationals and SOEs, as well as guar- antees of various sorts) • administering loan documentation • external reporting requirements (together with the middle office, but at a minimum providing basic statistics) • operational risk management, including business continuity and disaster recovery ar- rangements, as well as documented guidelines for overall operational risk manage- ment • managing the relationships with fiscal agents (e.g. the central bank) 3.2 Back Office Skills required The BO handles very significant transactions and processing errors can be extremely expen- sive. Improved technology can help reduce the incidence of these, but few systems are fail- safe. Therefore the BO requires professional staff with strong operational skills, including strong numeric skills, attention to detail, an ability to follow procedures and identify quickly when exceptions have occurred. More senior settlements staff will need negotiation skills, to manage exceptions with counterparties, and be able to stand up to front office staff, who typi- cally enjoy a higher status in the organization. The degree of training that will be required will depend on the extent to which the DMO can “buy� as opposed to “build� the skills it needs. To obtain the specialized knowledge listed 40 above, the vendors of IT systems (e.g. DMFAS and COMSEC) and operators of payment and custody systems and exchanges usually provide training. External training may also be possible in areas where the local private sector has similar re- quirements, e.g. the management of operational risk, training in market instruments for set- tlements staff. In most DMOs, there is significant on-the-job training of back office staff in addition to ex- ternal sources, because of the unique mix of instruments and IT systems that a sovereign bor- rower uses. Also, all organizations have developed procedures to manage operational risk which, while similar in principle, may differ in detail. For this reason, it is important that suf- ficient senior staff is retained to train new employees and ensure continuity of knowledge in the organization. This can be supplemented by cross-training and rotation through different positions. In order to perform their role, BO staff requires specialized knowledge that will reflect the environment in which they work, including: • Basic finance and financial markets • Basic accounting • Administration of loan documentation • Strong IT skills • the rules and conventions of the payment systems, custody systems and exchanges that they use (both in the local and international markets), as well as how to use any electronic interfaces with these entities; • an understanding of the financial instruments in which their organization transacts; • management of the IT systems the organization uses for transaction processing and debt recording (e.g. COMSEC, DMFAS etc). • administrative procedures of the lenders and/or donors from which they borrow; • other procedures and requirements of external organizations, e.g. for reporting pur- poses. More specific public debt management skills required would include: • Debt registration and database management in the particular IT system being used by that DMO • Reporting and statistics required by third parties (e.g. IFI’s) and by law • Settlement and payment systems in use and for potential use • Operational risk management for DMOs 4. LEGAL UNIT Debt managers must ensure that they receive appropriate legal advice and that the transac- tions they undertake incorporate sound legal features. As such the legal unit, or the lawyers incorporated in one of the main units e.g. MO, have an important role to play. 4.1 Typical Debt Office Legal Functions Typically, the legal staff/unit of the debt management office should: 41 • Actively contribute towards ensuring that the DMO’s operations are conducted ac- cording to the law. • Provide assistance in the form of legal research and advice, especially on new opera- tions such as derivatives (e.g. swaps) • Participate in the drafting of the regulations governing the activities of the DMO, and the primary and secondary markets of Government securities. • Take an active part in negotiations on contracts, and ensure that the “legal’ clauses do not excessively bind the borrower, that the borrower can fulfil its obligations without undue hardship, and that the contract at least is not stricter than the contracts entered into by other sovereign borrowers with similar credit status. 4.2 Debt Office Legal Skills required The degree of sophistication of its skills requirements will largely depend on the types of funding available for that particular country. Typically there will be a specific set of legal skills required for countries obtaining funding from official sources (concessional and non- concessional multi-lateral and bi-lateral), and then a learning curve for legal skills related to more market-based funding, both domestic and international. The legal skills learning curve continues as the DMO starts using debt transactions such as exchanges and buy-backs, as well as hedging transactions including derivatives such as cur- rency and interest rate swaps. The FO may also be involved in transactions of on-lending to sub-national governments and extending guarantees of various types to other government en- tities and/or the private sector. In some countries it may also be involved in executing cash management operations. As long as the DMO only borrows and enters into transactions in the domestic market, it is enough that the lawyers have solid knowledge of the domestic legal system, particularly fi- nancial and public law. Once the DMO transacts in foreign markets, broader skills and knowledge are required, as the foreign legal system will determine what the borrower legally can do. As most of the foreign capital markets have some form of consumer protection, the lawyer should be aware and check that the requirements for soliciting and offering securities to retail investors have been met. Also, loan documentation becomes more complex once the borrower enters foreign markets. Here the sovereign must be prepared to follow the rules of the market. The banks/investors normally want the same protection as they have when lending to main corporate borrowers, i.e., representations and warranties, default clauses (sometimes even cross-defaults and cross- acceleration clauses), negative pledge, and a waiver of sovereign immunity. Apart from being capable in reading and understanding these loan contracts, the lawyer also must be a skilled negotiator. The legal unit normally is the only function within the DMO which keeps record on the contractual restrictions on the sovereign. Once the borrower enters the international derivatives market, the legal issues become even more complicated. The standard agreement used, the ISDA agreement, is complex. The law- yers also must check if the contract is enforceable against the counterparty, which is gov- 42 erned by the counterparty’s home jurisdiction. If some kind of credit support for the exposure is called for, foreign rules on pledges also need to be reviewed. More specific needed skills include, but are not limited to, knowledge of: • Public sector law, in particular, law for public sector indebtedness, on-lending, guar- antees, etc. • Securities law, both domestic and international • Basic finance for lawyers, including financial markets and IFIs and official sources of funding • Financial instruments, both domestic and international, and hedging instruments, and debt-related transactions and their contracts(e.g. debt exchanges and buy-backs, guar- antees and on-lending contracts) • Legal negotiation of funding and bond issuance contracts • Drafting of regulations governing the DMO • Basic international securities legislation • Important clauses in international financial agreements • Structure of the ISDA agreement and related legal issues in derivatives transactions 43