An Assessment of the Use and Effectiveness of CS-DRMS for Managing Domestic Debt in the MEFMI Region

Transcription

1 Macroecon omic an d F in an cial Man agemen t In stitu te of E astern an d Sou th ern Africa An Assessment of the Use and Effectiveness of CS-DRMS for Managing Domestic Debt in the MEFMI Region A Discussion Paper Prepared as Partial Fulfilment of the Requirement for Accreditation Under the MEFMI Fellowship Programme By Lekinyi N. Mollel April,

2 Table of Contents TABLE OF CONTENTS... 2 LIST OF TABLES... 3 LIST OF FIGURES... 3 ACKNOWLEDGMENT... 4 ABSTRACT... 5 ABBREVIATIONS INTRODUCTION OBJECTIVES OF THE ASSESSMENT RELEVANCE OF THE STUDY ORGANIZATION OF THE PAPER BACKGROUND AND LITERATURE REVIEW BACKGROUND INFORMATION Economic Reforms Increased public Investments The Global Financial Crisis DOMESTIC DEBT DEVELOPMENTS IN THE REGION Trend of Domestic Debt Cost and Risks Analysis of Domestic Debt LITERATURE REVIEW SOURCE: ASSESSMENT METHODOLOGY FINDINGS AND ANALYSIS THE EXTENT TO WHICH CS-DRMS IS USED Management of Public Domestic Debt Functions of Domestic Debt Performed Using CS-DRMS Instruments Managed Using CS-DRMS CHALLENGES IN USING CS-DRMS FOR MANAGING DOMESTIC DEBT USERS EXPECTATIONS ON CS-DRMS VERSION CONCLUSION AND RECOMMENDATIONS CONCLUSION RECOMMENDATIONS REFERENCES APPENDICES APPENDIX I: QUESTIONNAIRE REPONSES APPENDIX II: QUESTIONNAIRE

3 List of Tables TABLE 1: DOMESTIC DEBT COSTS AND RISKS IN SELECTED MEFMI COUNTRIES TABLE 2: MEFMI COUNTRIES RESPONSES ON THE FUNCTIONS AND CATEGORIES OF DOMESTIC DEBT MANAGED USING CS-DRMS (DEC 2014) TABLE 3: NON-MEFMI COUNTRIES RESPONSES ON THE FUNCTIONS AND CATEGORIES OF DOMESTIC DEBT MANAGED USING CS-DRMS (DEC 2014) TABLE 4: MEFMI COUNTRIES RESPONSES ON THE CHALLENGES, AND UNDERSTANDING AND EXPECTATIONS ON CS- DRMS VERSION TABLE 5: NON-MEFMI COUNTRIES RESPONSES ON THE CHALLENGES, AND UNDERSTANDING AND EXPECTATIONS ON CS-DRMS VERSION List of Figures CHART 1: TREND OF GOVERNMENT STOCK OF DOMESTIC DEBT (BILLIONS OF TZS) CHART 2: TREND OF DOMESTIC DEBT IN KENYA (BILLIONS OF KES) CHART 3: DOMESTIC DEBT IN MEFMI POST HIPCS (PER CENT OF GDP) CHART 4: KEY FUNCTIONS OF DOMESTIC DEBT MANAGEMENT (DECEMBER 2014) CHART 5: INSTRUMENTS MANAGED USING CS-DRMS (END OF DECEMBER 2014) CHART 6: CHALLENGES IN USING CS-DRMS FOR MANAGING DOMESTIC DEBT CHART 7: EXPECTATIONS ON CS-DRMS VERSION

4 Acknowledgment First and foremost, I would like to give thanks to Almighty God for sustaining and leading me throughout the entire period of the apprenticeship and preparation of this discussion paper. My bona fide thanks to my family for their moral and material support as well as their prayers and acceptance of my long hours of absence during the entire period of the apprenticeship. I am indebted to the MEFMI Secretariat for giving this opportunity to undergo the apprenticeship process and considering me for accreditation among many eligible Graduate Fellows. Special thanks to Mr. Raphael Otieno, the Director of Debt Management Programme at MEFMI for his guidance during the entire period. I am also deeply grateful to my colleagues at MEFMI, particularly those in the debt management programme for their moral support during the process. It is my sincere appreciation of the cooperation received from officials in the debt management offices of the CS-DRMS user countries in the MEFMI region for completing the questionnaires that made this assessment successful. I would not be doing justice if I do not recognize the contribution of the Commonwealth Secretariat for inviting me to the CS-DRMS Training of Trainers in London, where I met CS-DRMS users and experts from outside the MEFMI region. Appreciation is extended to Mr. Vikas Pandey who gave his experience, on behalf of COMSEC, on the use of CS-DRMS in managing domestic debt. Thanks to the participants of the training for responding to my questionnaire. Sincerely your contribution enriched the findings of this study. The list is inexhaustible, I just say to them AHSANTENI SANA May God Bless You All! The support from individuals and institutions mentioned above notwithstanding, the responsibility for the views, and any errors, omissions or misinterpretations in the study remain solely mine. 4

5 Abstract Public domestic debt has gained significant importance globally since mid-2000s as a result of both external and external factors. In the MEFMI region, domestic debt has been increasing in absolute amounts, as a proportion of total public debt, and as a ratio of GDP. For instance, domestic debt increased from an average of 12 per cent of GDP in 2005 to 14 per cent by The domestic debt is already in the critical range of percent of GDP in Malawi, Uganda and Zambia while Tanzania is also approaching this range. Among the internal reasons behind the build-up of domestic debt in developing countries, including the MEFMI region, are: outcome of the reforms geared towards development of the domestic financial markets as a means of countering the perceived risks associated with external financing; implementation of monetary policies; and financing of infrastructure projects; and the relaxed fiscal policies to utilize the breathing space created by external debt reduction initiatives in 2000s. External factors, included the impacts of the global financial and economic crisis of late 2000s and the recent European debt crisis and to a lesser extent the aid inflows that necessitated government borrowing to sterilize them. The general increase in public domestic debt and the associated cost and risks created a need for effective management of this category of debt using a sophisticated computer based debt management system. Against this, the Commonwealth Secretariat continued to enhance the Commonwealth Secretariat Debt Recording and Management System (CS-DRMS) that was initially developed to address external debt crisis in mid 1980s. This study, therefore, assessed the use and effectiveness of CS-DRMS (Version 1.3) in managing public domestic debt. A total of 15 CS-DRMS user countries including eight from MEFMI region were assessed through a structured questionnaire. Views of the Commonwealth Secretariat were incorporated in the assessment. The assessment found that most of the debt offices are only using the system at minimum level and in fact parallel databases and records are maintained using other systems. Most of the countries are only using the system for recording and producing limited reports on debt service forecasts for treasury bills and bonds. The limited use of the system emanates from inadequate user technical 5

6 know-how, system capabilities and fragmented institutional framework for managing public debt in some countries. Deficiencies in managing public domestic debt, particularly, the reporting features were identified as a major challenge. CS-DRMS also has no features necessary to be used as a central depository system (CDS), a key function of domestic debt management. It was also noted that domestic debt issuance (auctions) in most countries is undertaken by central banks using their tailor made systems. This, is partly attributable to fragmented institutional arrangement in debt management. The consequence of all these is that domestic debt is not managed optimally and is characterized by duplication of resources including parallel records in spreadsheets. Nevertheless, the responses depicted users higher expectations on CS-DRMS version 2.0 released towards end of 2014 which was yet to be adopted in most of countries at the time of this assessment. Thus to optimize the use of CS-DRMS for managing public domestic there is a need to address both technical skills of the debt managers and deficiencies of the system, as well as consolidation of domestic debt management. Regarding technical skills, COMSEC, MEFMI and countries need to organize training events (regional and in-country) targeting only domestic debt management using CS-DRMS rather that combining with external as experience shows that whenever they are mixed less emphasis is given to domestic debt. Given the increasing trend of domestic debt there is a need for deliberate acceleration of research and development towards further enhancement of the domestic debt module of CS-DRMS, by COMSEC, to respond to emerging issues in the financial markets. Country reforms toward consolidation of debt management functions along the sound functional units, including consolidation of debt databases, would help to enhance the use of CS-DRMS in managing domestic debt as an integral of total public debt. Consolidation will facilitate bringing together domestic and external debt back office functions and adoption of single system (CS-DRMS) for recording and reporting public debt in totality. Findings of the study, if adopted, will enhance use of CS-DRMS in managing public domestic debt in the MEFMI region and improve management of public debt in totality with the ultimate goal of achieving the public debt management objectives of minimizing government financing costs and risks. 6

7 Abbreviations AFRODAD ATM ATR BOT BPM CBDMS CBK CDS CMSA COMSEC CS-DRMS DMFAS DSE FX GDP HIPCs HRT IMF JCTR JSE KES LIC MDRI MEFMI MTDS NDF TZS WAIR African Forum and Network on Debt and Development Average Time to Maturity Average Time to Re-fixing Bank of Tanzania Balance of Payments Manual Computer based debt management systems Central Bank of Kenya Central Depository System Capital Markets and Securities Authority (of Tanzania) Commonwealth Secretariat Commonwealth Secretariat Debt Recording and Management System Debt Management and Financial Analysis System Dar es Salaam Stock Exchange Foreign Currency Gross Domestic Product Heavily Indebted Poor Countries Horizontal Repos Transactions International Monetary Fund Jesuit Centre for Theological Reflection Johannesburg Stock Exchange Kenyan Shillings Low Income Countries Multilateral Debt Relief Initiative Macroeconomic and Financial Management Institute of Eastern and Southern Africa Medium Term Debt Management Strategy Net domestic financing Tanzanian Shillings Weighted Average Interest Rate 7

8 1. Introduction Public domestic debt 1 is increasingly gaining importance globally among the emerging markets and less developed countries since mid-2000s. In the MEFMI region, domestic debt has been increasing in absolute amounts, as a proportion of total public debt, and as a ratio of GDP. For instance, domestic debt increased from an average of 12 per cent of GDP in 2005 to 14 per cent by The domestic debt is already in the critical range of percent of GDP in Malawi, Uganda and Zambia while Tanzania is also approaching this range. The trend is not different globally. Ugo Panizza (2008) estimates that over the period , domestic public debt in the developing countries has increased from 19 per cent of GDP to 23 per cent. Giovanna et el (2014) also estimated that in Low Income Countries (LICs) domestic debt has been on the rise, increasing from 12.3 percent of GDP in 1996 to 16.2 percent in 2011, and in both HIPCs and non-hipcs, the public domestic debt represented 40 percent of the total public debt in 2011, almost three times the share observed in This happened while average total debt levels, including external, were decreasing from 75 per cent to 64 per cent of GDP in these developing countries. As a consequence, the share of domestic debt over total public debt went up from 30 to 40 per cent. In Mexico, for instance, the share of domestic debt went from 60 per cent of total public debt in 2004 to 73 per cent in The increase in domestic debt was mainly due to new borrowing contrary to that of external debt that was to large extent due to accumulation of arrears. Both external and domestic factors contributed to the increase in the domestic borrowing by governments. According to Ugo Panizza (2008), external factors are the main drivers of the accumulation of public domestic debt in both emerging market and low income countries. From external factors, the increase in domestic debt is attributable to either too little foreign aid or too much foreign aid. This is because, developing 1 For the purpose of this assessment, the definition of public domestic debt is adapted from IMF BPM6 (2013) that defines gross domestic debt, at any given time, as the outstanding amount of those actual current, and not contingent, liabilities that require payment(s) of principal and/or interest by the debtor at some point(s) in the future and that are owed to residents of an economy. 8

9 countries used the domestic debt market only when they did not have access to external resources or to sterilize aid flows. Other external factors are the impacts of the global financial and economic crisis of late 2000s and the recent European debt crisis. Among the internal reasons behind the build-up of domestic debt in developing countries, including the MEFMI region, are reforms geared towards development of the domestic financial markets. Countries resorted to develop domestic financial markets as a means to counter the perceived risks of external financing, implement monetary policies, and finance infrastructure projects. Other internal factors include the high and volatile domestic interest rates that translated into higher budget deficits, necessitating rolling over interest payments, and relaxed fiscal policies especially as a result of the breathing space created by external debt reduction initiatives in 2000s. Despite the positive impact associated with expansion of domestic debt on economic growth as found by Isaya et el (2008), this category of debt entails higher costs and vulnerability of government debt portfolio to refinancing risks, in particular. This is more pronounced in less developed markets like those in the MEFMI region. According to some countries medium term debt management strategies 2, on average, domestic debt has relatively higher weighted average interest rate (11.6 per cent), and shorter average time to maturity of 2.8 years and average time to re-fixing of 2.5 years. About 47.6 per cent of domestic debt matures within the next 12 months (Table 1). These risk indicators, emanating mainly from the short tenors of the domestic debt instruments and the fact that they are issued at markets conditions, are unfavorable when compared to those of external debt. This is mainly due to the fact that domestic borrowing is usually on prevailing market terms which among other things, the investors factor in inflation expectations and risk premiums as compared to concessional financing from some of multilateral and bilateral creditors. 2 MTDS for Tanzania, Lesotho, Namibia, Zambia, Malawi, Uganda, Mozambique and Angola 9

10 Given this general increase in public domestic debt and the associated cost and risks, there has been a growing need to effectively manage 3 this category of debt. This requires among others, maintenance of a detailed database on domestic debt within the principal debt management entity, using a sophisticated computer based debt management system (CBDMS) that meets sound operational, statistical and analytical needs of debt managers and other agencies involved in elaborating debt strategies. The CBDMS should have abilities to carry out among many functions the: storage of individual and aggregate data to reasonable detail; analysis of debt statistics; production of standard and customized reports for analysis and policy formulation; and conducting sensitivity analysis. This need facilitated development of diverse computer programmes by various entities. Although there are a number of customized computer software used to manage public, the Commonwealth Secretariat Debt Recording and Management System (CS-DRMS) and the Debt Management and Financial Analysis System (DMFAS) have gained popularity globally. The CS-DRMS has been in use since 1980s in many countries for recording and analyzing public external debt. However, over the last few years, COMSEC has been emphasizing, as well as assisting, countries to record domestic debt in CS-DRMS so as to have a common repository of the Government s debt portfolio. This translated to continuous revisions and upgrading of the system in line with new developments in the global financial sector, which has allowed it to withstand challenging needs and user demands. The latest development saw the release of CS- DRMS Version 2.0 in September 2014, which embeds enhanced tools for managing risks of the portfolio, improved accuracy of debt service projections by better alignment of the software to creditor practices and terms of the debt instruments, as well as rich reporting facilitates (The Commonwealth, 2014). As at end 2014, two countries in the MEFMI region (Namibia and Mozambique) had adopted the new version, becoming among the pioneers globally. Out of 14 Macroeconomic and Financial Management Institute of Eastern and Southern Africa (MEFMI) member states, eight countries 4 are using CS-DRMS in managing their (mainly external) 3 For the sake of this assessment effective debt management refers to the process of establishing and executing a strategy in order to meet government funding requirements with a view to ensuring that both the level and the growth of debt are fiscally sustainable. 4 MEFMI member countries using CS-DRMS as at end of 2014 are Kenya, Malawi, Swaziland, Lesotho, Botswana, Tanzania, Mozambique and Namibia. 10

11 debts. Nevertheless, fewer countries are using the system to manage domestic debt and to a limited extent. Some countries in the region are managing this category of debt using software developed in-house, mostly in Microsoft-excel based spreadsheets. Most of these alternative forms of storing debt records, were not developed purposely to manage debt and hence have inadequate functional capabilities to allow sound practices in domestic debt management. As a consequence, countries have been maintaining duplicate records of domestic debt and in some cases spread across a number of institutions. In turn, these do not only compromise quality of data, but also resources duplication, slow processing of data, and inadequate analysis of debt data. These happen despite the presence of relatively better systems such as CS-DRMS and DMFAS. The duplication of the databases is also attributable to fragmented debt management functions whereas in most countries, the central banks manage domestic debt while ministries of finance manage external debts, It is against this background that this assessment, prompted mainly by observations that in most countries, the over the shelve CBDMS, particularly CS-DRMS and DMFAS, are not used to manage domestic debt, was conducted. The findings are meant to shed light on the use and effectiveness of CS-DRMS 5 in managing domestic debt. The assessment found that countries are using the system at limited level because of its inadequate functional features as well as due to inadequate skills among the debt managers. Recommendations from this study are expected to inform further capacity building activities and enhancement of the system with a view to promoting use of CS-DRMS in managing all categories of domestic debt. The ultimate goal is to enhance efficiency in domestic debt management. 1.2 Objectives of the Assessment Since CS-DRMS and its application has evolved overtime in line with technological advancements and user requirements, there was need to investigate how best the system can be used to effectively to manage public domestic debt. This study assessed the extent of system usage in managing domestic debt and evaluated its adequacy in managing domestic debt in line with user 5 This assessment is based on CS-DRMS Version 1.3. This is because by time of data collection only a limited number of countries globally had already adopted the latest version. For this case some of the recommendations to be drawn may have already been addressed in CS-DRMS Version

12 requirements. The assessment is informed by the findings from the region based on the business side responses, answered the following specific questions that pertain the objectives of the study: (i) The extent to which CS-DRMS v1.3 is used to manage public domestic debt. (ii) The effectiveness of CS-DRMS v1.3 in managing public domestic debt. The specific questions were meant to ascertain the hypotheses that CS-DRMS is effectively used by countries and has the necessary features needed for effective management of public domestic debt. 1.2 Relevance of the Study The recommendation drawn from the findings are expected to enhance use of CS-DRMS in managing public domestic debt with a view, among others, to enhancing: provision of accurate, timely and reliable statistics on public domestic debt; elimination of duplication of resources (multiple databases and systems); consolidation and analysis of country s total debt (external and domestic); and formulation of sound/credible macroeconomic policies. The findings will inform further improvements of CS-DRMS and designing of customized training programmes. The individual country s findings are also expected to inform capacity building interventions by MEFMI and COMSEC in the respective countries. 1.3 Organization of the Paper This discussion paper is organized into five main parts: apart from the introduction, the next section reviews the background information and literature on managing public domestic debt in region. Part three discusses the methodology used in collecting and analyzing data. Findings are discussed in Part Four, while the last part concludes the assessment, including summarizing the findings and drawing recommendations for effective use of CS-DRMS in managing public debt in the MEFMI region. The questionnaire and its responses are annexed to this study. 12

13 2. Background and Literature Review 2.1 Background Information The growth of domestic debt in the developing countries, including those in the MEFMI region, since mid-2000s, is attributable to both internal and external factors. These factors have their origin in internal economic reforms that improved domestic financial markets; increased demand for public investments; and global financial crisis and economic recession of late 2000s Economic Reforms Economic liberalization including those related to domestic interest and exchange rate policies opened doors for developments in the financial sector, a key player in the financial and domestic debt markets. Particularly, starting early 1990s, most countries in the region embraced regulatory reforms aimed at developing domestic financial and debt markets. The reforms included liberalizing financial markets, modernizing the market infrastructure and diversification of domestic debt market instruments and investor base, as well as modernizing the legal frameworks governing the financial markets. Thus, governments started issuing market based financing securities, mainly Treasury bills and bonds. In most of the countries, the investors responded positively. In addition to financing primary deficits, domestic debt has also been used to implement monetary policies. While Botswana and Uganda, for instance, have been issuing domestic debt for monetary policies only for quite a long time in 1990s and 2000s, Lesotho, in early 2010s, issued some instruments of domestic debt mainly to develop financial and debt market. In Tanzania, the remarkable reforms followed enactment of the Banking and Financial Institutions Act of 1991 that among others aimed at developing the financial markets. In the aftermath, the Bank of Tanzania launched 91-day Treasury bills auctions to finance short-term government deficit, implement monetary policy, and as a reference point of the determination of market interest rate (AFRODAD, 2013). More developments in the markets followed including launch of 35-days 13

14 Treasury bills in 1993, 182-day and 364-days Treasury bills in 1994, and 2-year Treasury bonds in These were followed by establishment of Capital Markets and Securities Authority (CMSA) in 1997 to promote and regulate capital markets operations, and Dar es Salaam Stock Exchange (DSE) in 1998 to provide a platform for secondary trading. Milestone developments took place in 2002 following introduction of medium-term Treasury bonds (5, 7 and 10-years) and the 15-year Treasury bonds introduced in The Government listed the medium to long-term bonds in DSE, though secondary trading has not been active due to buy and hold strategy of investors. These developments manifested to increase in domestic borrowing. The domestic debt therefore grew from TZS billion in 2003/4 to TZS 9,535.5 billion in 2013/14 (Chart 1). On average about 99 percent of Government domestic debt in Tanzania was issued through the market, which is partly attributable to the reforms implemented since early 1990s. Chart 1: Trend of Government Stock of Domestic Debt (Billions of TZS) Source: Bank of Tanzania (MER June 2014) In Kenya, Horizontal Repos Transactions (HRT) platforms were adopted in 2008 to enhance liquidity distribution across the financial system. This was followed by the Automated Trading Systems that was adopted in 2009 to improve efficiency and safety of secondary trading for bonds. Action based offers through issuance method and pricing has promoted price discovery which is crucial for market development. Pension and Insurance reforms created demand for long term 14

15 bonds. The reforms implemented during period starting in 2008 coincided with increased domestic borrowing by Government to mitigate the impacts of the global financial crisis and the thrust to finance infrastructure projects. Consequently, government domestic debt grew from KES billion by end of 2008 to KES 1,284.3 billion as at end of June 2014, of which the securitized debt accounted for an average of about 97 percent (Chart 2). Chart 2: Trend of Domestic Debt in Kenya (Billions of KES) Gov't Securities Others , Source: Compiled from CBK Publications These reforms unlocked private sector resources to finance governments budgetary operations that were initially financed largely externally. These therefore translated to the build-up of domestic debt in the region Increased public Investments The period since 2000 witnessed greater appetite for public investments, targeting poverty reduction and in developing countries most of them linked to attainment of millennium development goals by To accelerate progress in attaining the millennium goals, the G8 6 The Millennium Summit of the United Nations in 2000 identified eight (8) goals to be achieved by The goals were: to eradicate extreme poverty and hunger; to achieve universal primary education; to promote gender equality and empower women; to reduce child mortality; to improve maternal health; to combat HIV/AIDS, malaria and other diseases; to ensure environmental sustainability; and to develop a global partnership for development. 15

16 countries agreed in June 2005 to provide resources through cancellation of debts worth between US$ 40 and US$ 55 billion owed to the World Bank, International Monetary Fund and African Development Bank. Nevertheless, the freed up resources were not adequate in financing attainment of the millennium goals and governments resorted to borrowing to supplement these resources. External sources have volatile, procyclical, and subject to sudden stops (Calvo, 2005), compelling governments to increase dependence on domestic borrowing as source of development financing. The significant decline of external sources, mainly from bilateral, multilateral and concessional sources, heightened after the global financial and economic crisis (OECD, 2014; and AFRODAD, 2011). This happened during the era when most of the post HIPCs in the region were implementing IMF supported programs that limit external non-concessional borrowing and monetary financing of government operations. AFRODAD (2011) alludes that there are cases where governments have been borrowing domestically to service external obligations The Global Financial Crisis The economic and financial crisis that materialized in 2008 affected significantly the developing countries that depend primarily on exports of commodities, not sparing those in the MEFMI region. The negative impact of the crisis manifested through reduced revenue earnings from primary exports, loss of jobs as businesses closed or scaled down production, and declining foreign direct investments. Governments experienced decline in both domestic revenue and aid inflows. In Zambia, JCTR (2009) estimated job losses of up to 12,000 in the mining sector alone following the closure or placement under care and maintenance of a number of mines. This was caused by a fall in copper export proceeds as its world price declined from US$ 8,980 to US$2,812 per tonne during the period July - December This led to reduced earnings for the government from mineral royalties and corporate taxes as well as falling income taxes from employees in the mining and related sectors. Consequently, the Government resorted to domestic borrowing to finance its 16

17 operations. As a result, domestic debt increased from ZMK8.3 billion in 2007 to ZMK8.5 billion in 2008 and further to ZMK 20.5 billion by September The situation was not different from the rest of the countries. Kenya experienced a sharp increase in domestic debt interest payments from around KES 40 billion in 2006/07 to KES billion in 2012/13. This is largely due to an increase in Government domestic borrowing to compensate for the shortfalls in the budget-anticipation of external financing (AFRODAD 2011). Although, Tanzania was also not spared with the impact of the Financial Crisis, it maintained the NDF target of one per cent of GDP as agreed under the IMF-PSI. Although nominal domestic borrowing were significant, Tanzania managed to maintain the NDF target due to the strong economic performance. In most countries globally, the export linked entities were adversely affected. To mitigate these impacts the governments, advanced and developing, opted to bail out the key productive entities, through stimuli packages that varied in magnitude and mechanism from one country to another. These included recapitalization, loss compensations, cleaning balance sheets particularly on nonperforming loans of the financial institutions originating from export losses, and issuance of guarantees. Irrespective of the magnitude and mechanism, governments incurred domestic liabilities in the process. The IMF (2010) estimated that the 2007 crisis contributed to large extent the deterioration of fiscal balances in emerging and developing economies from 0.51 in 2007 to in This was due to increased public spending and lower revenues. 2.2 Domestic Debt Developments in the Region The trend of domestic debt in the MEFMI region evolved, both in magnitude and cost and risk characteristics, in line with developments elucidated in the preceding section Trend of Domestic Debt Public domestic debt increased from an average of 12 per cent of GDP in 2005 to 14 per cent in 2011 in the region, with significant increases in Malawi, Tanzania and Uganda (Chart 3). 17

18 % of GDP Chart 3: Domestic Debt in MEFMI Post HIPCs (Per cent of GDP) critical level Malawi Mozambique Rwanda Tanzania Uganda Zambia Countries Source: Compiled from Countries DSA and MTDS Documents Cost and Risks Analysis of Domestic Debt This increase in domestic debt is happening amid limited securities that are concentrated on the shorter end of the maturity spectrum; narrow investor base dominated mainly by commercial banks and central banks; underdeveloped infrastructure; and inactive secondary markets. Consequently, the public domestic carries relatively higher interest rates and expose government debt portfolios to significant refinancing and interest rate risks. Table 1 gives a summary of cost and risks of domestic debt in some MEFMI countries. 18

19 Table 1: Domestic Debt Costs and Risks in Selected MEFMI Countries Cost-Risk Indicator Lesotho (2013) Angola (2012) Malawi (2013) Mozambique (2014) Namibia (2012) Tanzania (2013) Uganda (2013) Zambia (2014) Cost WAIR 1.2 na ATM (yrs) Refinancing Debt maturing in Risk yr (% of total) ATR(yrs) Debt refixing in Interest yr (% of total) Rate Risks Fixed rate debt (% of total) Source: Compiled from Countries MTDS documents Average 2.3 Literature Review Although literature on the impact of technological advancement on public debt management is scanty, few studies for example Mehmet and John (2013) found that after the incorporation of SETS 7, the Johannesburg Stock Exchange (JSE) has become more independent and it now offers better diversification opportunities for international investors. Thus, deliberate reforms including adoption of state of the art technology in managing domestic debt are critical to avert a vicious cycle of recurring debt burden and its associated macroeconomic distortions particularly in elevating domestic interest rates and consequent crowding out effects. Adoption of electronic trading in most African countries reduced costs of executing trades, and consequently increasing the trading volumes as well as facilitating cross border portfolio investments (OECD, 2014). In assisting the developing countries to improve their capacity to record and disseminate information on the structure of total public debt, UNCTAD and COMSEC developed the DMFAS and CS-DRMS, respectively. Since the release of CS-DRMS in 1985, countries have been using 7 SETS is electronic order book (system) used by the London Stock Exchange. 19

20 it to record and manage debt by providing a comprehensive repository for both external and domestic debt data. The system continued to be enhanced to respond to changes in instruments, creditor practices, debt reporting standards, and technology. The system has the necessary features to perform the key functions of sound management including those related to recording, analysis and reporting. The material used in the UNCTAD s e-learning training course on Debt Sustainability Analysis (DSA) 8 indicates that CS-DRMS has a comprehensive domestic debt module that allows for the recording of the full issuance cycle of domestic debt instruments such as treasury bills, bonds, and notes, and for the planning of issues, auctions, and analysis of bids. The system also captures actual and forecast transactions data as well as that on arrears are in a manner that meets the international external debt data guidelines (Box 1). Box 1: Key Debt Management Functions of CS-DRMS Debt Recording Debt Analysis Debt Reporting Other functions Maintain an inventory of all external and domestic debt instruments including: o Public debt and grants o Short-term and private sector debt o Restructuring agreements including rescheduling Record basic details and terms of an instrument Record other relevant debt related information such as exchange rate, interest rates, and macroeconomic data Record actual transactions of debt service and disbursements on a transaction-by-transaction basis Monitor loans and grant utilization and disbursements Monitor government lending including on-lending Perform sensitivity analysis on interest and exchange rate variation under various scenarios Test the implication of new borrowings, based on different assumptions of currencies, interest, and repayments terms Evaluate different loan offers Evaluate different proposals for refinancing and rescheduling of loans and compute debt relief Combine CS-DRMS debt data with exogenous economic data to project critical debt indicators, both on nominal and present value basis Evaluate exposure to exchange and interest rate risk Identify loans in arrears and calculate penalty payments Provide information and reports on any group or class of instruments Produce standard reports for various data requirements including government finance, balance of payments and IIP Provide easy generation custom reports using a purpose-build report generator Respond to specific user enquiries into the database Forecast debt service payments, both by instrument and in aggregate terms, with and without future disbursements QEDS, Horizon, etc Use Transfer debt data electronically to the World Bank s DRS, MTDS-AT, validation utilities to ensure database integrity and accuracy Integrate front, middle and back office functions via the database and security management options Perform housekeeping functions such as backup and restore and setting up modern access Source: External Debt Statistics, 2003, ch. 18 as quoted by UNCTAD s e-learning training course on Debt Sustainability Analysis 8 Available at 20

21 3. Assessment Methodology The assessment of the usefulness and effectiveness of CS-DRMS in managing public domestic debt is informed by business side users perceptions. The major source of data for analysis is based on primary data collected using structured questionnaire (see Appendix 1) and author s personal experience during the apprenticeship. The questionnaires were sent to CS-DRMS sites within MEFMI region, that is, in the eight countries using the system namely; Botswana, Lesotho, Kenya, Malawi, Mozambique, Namibia, Swaziland and Tanzania. All targeted offices responded by completing the questionnaires without undue delays. Information was collected on the situation as at end of December 2014 and in most cases based on CS-DRMS Version 1.3. For comparison purposes, data was also collected from seven non-mefmi member states namely; Antigua, Guyana, India, Jamaica, Nigeria, Sri Lanka and Tonga. Responses from these non-mefmi countries were collected through their officials who participated in the Training of Trainers Workshop organized by COMSEC in London from 12 th to 23 rd January The perception of COMSEC, the system developers, was also solicited through Mr. Vikas Pandey, a system development specialist. The questionnaire had three major parts, structured to answer the questions that pertain to the objectives of the assessment. The first section covered by questions one through four collected information on the extent to which CS-DRMS is used to manage public domestic debt. This identified debt management offices using the system to manage public domestic debt, and the functions and categories of domestic debt that are managed using CS-DRMS. The second section, covered mainly by question five, collected information on the challenges faced by domestic debt managers using CS-DRMS. The challenges are mainly those related to system features and/or skills of debt managers in using the system. Information collected in this section was meant to assist future revisions of CS-DRMS and in designing informed interventions to the specific countries. 21

22 Question six that forms part three was mainly meant to collect information on users understanding of the latest version of CS-DRMS (version 2.0), and expectations of the countries in addressing the challenges identified in the preceding part of the questionnaire. The subsequent section of this paper analyses the findings based on the three major sections of the questionnaire. Both quantitative and qualitative analysis of the collected will be employed to facilitate full extraction and use of information for sound interpretation and analysis. Thus, the findings is in form of narration, charts, figures and tables. 4. Findings and Analysis The findings discussed in this section are based on the responses from a total of 13 CS-DRMS user countries, including eight from the MEFMI region. All the targeted offices responded by completing the questionnaires. The findings are structured first on the extent to which CS-DRMS is used to manage domestic debt highlighting the functions and categories of debt managed. The second section discusses the challenges faced by CS-DRMS users and the last part exposes the user expectations on the latest version. 4.1 The Extent to Which CS-DRMS is used In foremost, the assessment was whether the responding office is managing domestic debt using CS-DRMS and then the functions and categories of public domestic debt managed in the office using the system Management of Public Domestic Debt The findings showed that all eight CS-DRMS user countries 9 in the MEFMI region are involved in managing domestic debt as an integral part of the total sovereign debt. All countries use CS- DRMS for managing certain functions and categories of domestic debt. However, out of the seven 9 The assessment involved debt management offices/units in the Ministries of Finance, except in Tanzania where the response was received from the Bank of Tanzania. 22

23 non-mefmi member states included in the survey, two countries are not managing domestic debt 10. Alongside using CS-DRMS for managing domestic debt, all offices reported maintenance of parallel records, mainly in Microsoft-excel, due to reasons to be discussed in the next sections of this assessment Functions of Domestic Debt Performed Using CS-DRMS In analyzing the extent to which CS-DRMS is used to manage public domestic debt, ten key functions of domestic debt management were listed each assigned equal weight. Each respondent had to indicate the functions applicable to his or her office. The assessed functions are: (1) Auctions, (2) Secondary trading, (3) recording of debt instruments, (4) analysis of costs and risks of new domestic debt borrowing, (5) analysis of costs and risks of the existing domestic debt portfolio, (6) analysis of domestic debt in relation to other categories of debt mainly external and total, (7) analysis of domestic debt in relation to macroeconomic variables, (8) forecasting domestic debt payments, (9) generation of statistical and analytical domestic debt reports, and (10) any others. In general, the findings indicate that CS-DRMS is adequately used in recording domestic debt and forecasting debt service payments as well generating statistical and analytical reports. The system is used in recording domestic debt in all countries in the MEFMI region and in five out of the seven non-mefmi countries. The system is also used to forecast debt service payments and in generation of statistical and analytical domestic debt reports. About half of the MEFMI countries use CS- DRMS to forecast debt service payments as compared to about 71 percent of non-mefmi countries. The findings further indicate that about 63 percent of CS-DRMS users in the region generate domestic debt statistical reports using the system. The findings are not different from those of Mr Vikas 11 that CSDRMS is capable of keeping track of the stock of domestic debt of the country as well as projecting on the debt service payments. 10 Respondents from Sri Lanka and India indicated, domestic debt is solely managed by their central banks using inhouse developed systems that are based on Microsoft-excel. 11 Mr Pandey (respondent to the questionnaire) is system developer at COMSEC, with his area of speciality being on domestic 23

24 However, limited use of the system in conducting auctions of domestic debt instruments was reported, whereas only one and two MEFMI and non-mefmi countries, respectively responded as using the system for auctions. This support the fact the information on the holding patterns of the securities is not up to date as CS-DRMS does not incorporate Depository functions and therefore cannot record the details of the movements of securities among investors in a secondary market trade. The findings further correspond to the OECD (2014) work which showed that, 16 out of 29 reviewed countries in Africa, have the auctions undertaken by the central banks. The explanation is that central banks have necessary staff with capital markets expertize to conduct auctions, managing tap sales, and interacting with financial market investors. The use of government securities for both monetary and financing purposes gives central banks the upper hand. The findings correspond to the COMSEC views that in most countries, the domestic debt is managed in their respective central banks. The Central Banks have their own systems for recording domestic debt and these systems have more functionalities not available in CS-DRMS, mainly the depository functions and links to payment system for debt servicing. In efforts to address this, COMSEC also provides assistance to CS-DRMS user in upload the data in respect of outstanding and historical domestic debt instruments, and setting up an electronic link with Central Bank s systems for upload of data in respect of new domestic debt issuances. Reponses also showed that most of the analytical features of CS-DRMS are not used. As expected none of the assessed countries is using the system to conduct secondary market trading. This is mainly due to the fact that secondary market trading ideally takes place outside of debt management offices. Chart 4 shows the countries and the key functions of domestic debt management managed using CS-DRMS. 24

25 Chart 4: Key Functions of Domestic Debt Management (December 2014) MEFMI Non-MEFMI Number of countrries a b c d e f g h i j 0 Key Functions* Note: * a = Auctions, b = Secondary trading, c = recording of debt instruments, d = analysis of costs and risks of new domestic debt borrowing, e = analysis of costs and risks of the existing domestic debt portfolio, f = analysis of domestic debt in relation to other categories of debt, g = analysis of domestic debt in relation to macroeconomic variables, h = forecasting domestic debt payments, i = generation of statistical and analytical domestic debt reports, and j = Others. Source: Author compilation Among the explanations for the limited use of the CS-DRMS to manage domestic debt is the fragmented debt management offices. Even in countries where debt is managed in one unit, the segregation of duties is not according to Front, Middle and Back office, but according to domestic and external debt. Experience shows in most debt management offices, CS-DRMS database is installed in the external debt units. Thus, the back office functions of domestic debt are separated from those of external debt. As such, domestic debt managers resort to excel spreadsheets as an easy way to record domestic debt. This is the fact even in cases where CS-DRMS has adequate features to perform such functions Instruments Managed Using CS-DRMS For comparison purposes, a list of instruments were listed for the respondents to indicate those managed in their countries using CS-DRMS. The instruments assessed were (1) Treasury bonds notes, (2) Treasury bills, (3) Central Bank advances, (4) debt loans/instruments, (5) undated 25

26 loans/instruments, (6) suppliers credits and/or arrears, (7) future or arrears of pension obligations, (8) embedded options and/or derivatives, and (9) others to capture any other instruments not listed from 1 to 8. The findings show that all CS-DRMS user countries in the MEFMI region use the system to record Treasury bills and bonds (Chart 5). The situation is, however, different in the non-mefmi countries where only four (about 57 per cent) use CS-DRSMS to record the instruments. This corresponds to views of COMSEC that the Domestic Debt module in CS-DRMS allows for recording of most types of securities commonly issued by the Government (e.g. T-Bills and various types of Bonds). The system allows the users to record these instruments in 3-layer architecture: Instrument properties of the instrument; Tranche Issuance details; and Bids Successful bidders. The system developers are, however, undertaking continuous enhancement of the functionalities to cater to various different types of instruments which the countries issue in their domestic market. Some of the major enhancements made recently are: handling amortised bonds; bonds with capitalised interest; annuity bonds with varying interest rates; and re-opening of bonds. It was also found that half of MEFMI countries record the dated loans/loans in the system as compared to three outside the region. However, only one country in each group is has reported recording undated instruments in the system and no country is capturing suppliers credits, pension obligations or embedded options. Two countries from the region namely Lesotho and Mozambique, reported to be using the system to record domestic guarantees. Lesotho records loans from commercial banks to members of the Parliament although repayments are done through direct deductions from their wages, Mozambique records building leases. 26

27 Chart 5: Instruments Managed Using CS-DRMS (end of December 2014) Number of countries MEFMI non-mefmi Source: Author compilation 4.2 Challenges in Using CS-DRMS for Managing Domestic Debt From the preceding section, the findings show that the MEFMI and Non-MEFMI CS-DRMS user countries use only 42.5 and 30 per cent of the system, respectively as per the areas covered by the assessment, that is, the instruments and key functions of domestic debt management. The questionnaire also collected information on the challenges faced by users in using the system to manage public domestic debt. The responses indicated both skills and inadequate features of the system as the main challenges (Chart 6). Three-quarters of the MEFMI s CS-DRMS user countries reported the two challenges. However, only two (29 per cent) of non-mefmi countries attributed these as challenges. The findings further shows that one country, outside the MEFMI region, reported absence of the domestic debt module in their database as a challenge. Inability to interface CS-DRMS with other accounting systems was also reported by one country in the region. Nevertheless, all respondents highlighted inadequate reporting features of the system as challenges in utilizing the system effectively. Consequently, all the assessed countries are maintaining parallel domestic debt databases and/or records in different formats, particularly for reporting purposes. 27