Public Debt Middle Office Wg Report

FIRST DRAFT OF THE WORKING GROUP REPORT ON THE NEED FOR A MIDDLE OFFICE FOR PUBLIC DEBT MANAGEMENT (as circulated in the Fifth Meeting of the Working Group on January 19, 2001)

1

FOREWORD The need for a comprehensive strategy for public debt management has been felt for some time because of high fiscal deficits, the consequent increase in debt burden of the Central Government and the absence of an integrated approach towards domestic and external public debt management. As a major step towards putting in place an effective system for public debt management, a Working Group has been constituted to examine the need for an integrated “middle office” for public debt management. The role of the “middle office” envisaged is to make analysis, provide advisory support and management information system inputs for debt management decisions. The issue was examined in the context of effective public debt management of the country keeping in view the possible long-term requirements. Given the wide array of issues involved, the Group has restricted its focus on the institutional structure, risk management requirements and best international practices. The recommendations made in this regard, for setting up a centralized middle office, should be viewed as the starting point towards creation of a strong institutional system for prudent public debt management. The recently introduced Fiscal Responsibility and Budget Management Bill 2000, which encompass stipulating limits on public debt as a proportion of GDP, besides other ceilings on fiscal indicators, have created conducive conditions for capacity building for public debt management. The initiative of the Group, as outlined in its recommendations, is also in tune with recent efforts by international financial institutions like the World Bank and IMF to identify as to what constitutes sound practices for sovereign debt management. The setting up of the Working Group and its Report was made possible by funding from the Institutional Development Fund Grant of the World Bank, aimed at strengthening the debt management capacity of the country. The Group is indebted to the valuable insights provided by Mr. Graeme Wheeler and Mr. Fred Jenson from the World Bank, Mr. Nihal Kappagoda for his consultancy report on best international practices, and Dr. Raj Kumar, from the Commonwealth Secretariat, London for complementing Mr. Kappagoda’s efforts. The Group also had the benefit to be enriched from different officials of select debt offices visited. The Working Group would like to acknowledge contributions made by Mrs. Usha Thorat from the Reserve Bank and Mr. Alok Chaturvedi from the Ministry of Finance. On behalf of the Group, I would also take this opportunity to place on record special appreciation for the technical inputs provided on the subject by Mr. Arindam Roy from the Ministry of Finance. The Group is indebted to officials from External Debt Management Unit in the Ministry of Finance for rendering valuable assistance and providing secretarial support for the Group. (Arvind Virmani) Tuesday, January 30, 2001

2

INDEX Report of the Working Group 1. 2. 3. 4. 5.

Introduction Debt Position of the Central Government Central Government Debt Management International Experience Recommendations

1 1 3 4 5

Annexures I.

Constitution of the Working Group

II.

Central Government Debt 2.1 Definition of Public Liability 2.2 Debt Position of the Central Government 2.3 Debt Burden of the Central Government 2.4 Net Liability of the Central Government 2.5 Contingent Liabilities of the Central Government 2.6 Legal Ceilings on Central Government Debt

III.

Existing Structure for Public Debt Management III.1 Internal Public Debt Management III.2 Other Domestic Liability Management III.3 External Public Debt Management III.4 Missing Links in Middle Office Functions for Public Debt Management III.5 Co-ordination among Debt Management Activities

IV.

International Experience and Best Practices in Public Debt Management 4.1 Approach in Study of Best International Practices 4.2 Risk Management Framework for Government Debt 4.3 Institutional Structure for Public Debt Management 4.4 Scope of Role and Functions of Debt Offices 4.5 Organisational Structure 4.6 Governance - Legal Framework and Accountability 4.7 Coordination with Fiscal Policy Authority and Monetary Policy Authority 4.8 Capacity Building in Sovereign Debt Management

V.

Summary Recommendations by the Commonwealth Secretariat on Best International Practices in Sovereign Debt Management

VI.

List of Select Debt Offices Visited by the Working Group 3

Report of the Working Group 1.

Introduction

1.1 The importance of public debt management has grown immensely over the years. In India, the need for a comprehensive strategy for public debt management has been felt for some time. The concern stemmed from the accumulation in the stock of Central Government debt, high interest burden on public debt, absence of an integrated risk-management approach towards domestic and external public debt management, and the need to manage contingent liabilities of the Government in a centralised manner. 1.2 As a step towards putting in place an effective system for public debt management, a Working Group was constituted on July 18, 2000 to examine the need for an integrated “middle office” for public debt management. The middle office is usually the entity located within a debt management office, which serves as the risk manager, formulates and advises on the debt management strategy and also develops benchmarks for assessing the risk-cost trade off of the portfolio. The role of the “middle office” envisaged is to make analysis, provide advisory support and management information system inputs for debt management decisions in the overall framework of the Central Government debt portfolio, keeping into perspective the possible long-term requirements. The Memorandum for setting up the Working Group, including the composition and terms of reference is at Annex 1. 1.3 The Working Group was given a timeframe of four months i.e. up to November 17, 2000 for submitting its Report. However, given the limited time and wide scope of the terms of reference, the term of the Working Group was subsequently extended up to January 31, 2001. 1.4 To study the need for a middle office, the Group analysed the indebtedness position of the Government for understanding the scale of risk and cost associated with it. Simultaneously, a review of the present debt management structure in India was attempted to identify missing links, if any, in the area of middle office functions for public debt management. In its endeavor to make recommendations, the Group also undertook an in-depth analysis of the international best practices for middle office role. 2.

Debt Position of the Central Government

2.1 The mounting debt of the central government due to a long legacy of high fiscal deficits and the increasing use of financing such debt for current expenditure has led to a continual deterioration of the indebtedness position of the central government. The total debt of the central government increased 4

annually at a compound growth rate of 13.4 per cent during end-March 1991 to end-March 2000. While, the ratio of Central Government debt-GDP ratio stood at 60.9 per cent during 1999-2000, the debt dynamics was characterised by a “cross-over” in the growth of nominal debt relative to the growth of nominal GDP, thus pointing towards further deterioration in the debt-GDP ratio. The composition of outstanding debt also, underwent a change, with the share of domestic debt increasing steadily since 1991-92. The growth in the domestic component of Central Government debt was spurred by a heavy accumulation of internal debt. The re-emergence of an increasing trend in revenue deficit as a proportion of GDP since 1996-97 could be attributed to the sharp increase in the internal debt of the Central Government in recent years. As a result, the ratio of internal debt to GDP deteriorated significantly from 25.3 per cent in 1996-97 to 37.7 per cent in 1999-2000. A detailed study of the indebtedness position of the Central Government is at Annex 2. 2.2 The steady accumulation in Central Government debt, particularly internal debt, resulted in increased interest payments burden. The total outgo on interest payments on Government debt, which accounted for one-fifth of the total expenditure by the Government in 1990-91 increased to one-third of the total expenditure in 1999-2000. Thus, interest payments, which pre-empted onehalf of the tax revenue in 1990-91, pre-empted nearly three-fourth of the tax revenue of the Central Government in 1999-2000. Similarly, the proportion of interest payments in total revenue receipts increased from 39.2 per cent in 199091 to 51.9 per cent in 1999-2000. The increase in interest burden, inter alia, stemmed from larger recourse to market borrowings at market-related rates. Interest payments on external debt, on the other hand, remained fairly stable, reflecting the concessional nature of such debt flows. 2.3 The over-hang in internal debt also led to potential for significant rollover risks in the medium term. The objective of the internal debt manager to raise debt at minimum cost coupled with a shift in investor preference towards short-term paper led to a policy of placing borrowings at the shorter-end of the market. Given the burgeoning market borrowings, the share in outstanding market borrowings in shorter maturity securities (with a maturity of less than 5 years) accordingly increased from 7.4 per cent of the total market loans in endMarch 1992 to 45.2 per cent in end-March 1998. Although, since 1998-99, the maturity structure was lengthened with the introduction of long-term securities and a conscious policy to avoid placing of short-term securities, the weighted average maturity creeped up steadily from 5.5 years in 1996-97 to 12.6 years in 1999-2000. This entailed large future redemptions during the next ten years. The external debt component, on the other hand, is characterised by smooth redemption profile during the next ten years, reflecting the long-tenor of official debt flows.

5

2.4 Thus, while the debt of the Central Government has reached unsustainable limits, there is a potential of significant interest rate and roll-over risks of the domestic debt component, at least in the medium term. For the external debt component, there remains significant foreign exchange risk, in the absence of any hedging or portfolio management strategy. In addition, contingent liabilities of the Central Government, in the form of guarantees on both domestic and external debt, which, amounted to 5.9 per cent of GDP in 1998-99, poses further risk for the Central Government’s debt position. 2.5 To redress the growing indebtedness position of the Central Government, in the past, several public agencies recommended enacting legislation for controlling the Central Government borrowings. The Constitution of India has provision for placing a limit on the public debt of the Central Government. Very recently, in an effort to promote overall fiscal prudence, the government has introduced the Fiscal Responsibility and Budget Management Bill, 2000 in the Parliament. If the proposed Bill finds passage in the Parliament, it could clearly serve as the starting point for improving the indebtedness position of the government. 3.

Central Government Debt Management

3.1. The Indian case of public debt management could, at best, be characterised by a clear dichotomy – management of internal debt by its central bank, and the other component of domestic debt and external debt, by the Government. The focus of internal debt management by the Reserve Bank has been to manage the refinancing and liquidity risk, while ensuring that borrowings are raised at minimum cost. Efforts aimed at managing other risk components in terms of portfolio management policies by the Reserve Bank have been constrained by the overriding concern of meeting the large scale of borrowing needs and the government debt market which is still in the process of transformation towards maturity. Portfolio management exercises for other domestic liabilities, which has largely been characterised by autonomous flows, is completely missing and the policy framework has been largely dominated by budgetary needs. On the other hand, for public external debt management, since the bulk of the debt came from official sources in the form of aid flows, the Ministry of Finance focussed solely on retiring expensive debt by prepayments or refinancing, without any portfolio management exercises. Only recently, a modelling exercise for risk management of the sovereign external debt portfolio was initiated, after an initiative by the IBRD to allow borrowers flexibility for deciding on currency and interest and also to allow hedging instruments on existing IBRD debt for portfolio management purpose. Annex 3 outlines the institutional structure for public debt management of Government of India. 3.2 The dispersed nature of public debt functions both across and within different agencies could be responsible for the lack of an integrated debt 6

management structure, which in turn has led to many a missing links in Government’s debt management strategy. The pivotal role in this regard, usually played by a middle office; which formulates the borrowing strategy, establishes a framework for risk management and periodically advises on debt management operations in an integrated framework has been largely conspicuous by its absence. Thus public debt management fell short of developing public debt sustainability benchmarks; medium and long-term public debt management strategy; making a choice between domestic and external debt; developing benchmarks for currency, interest and maturity mix as part of reducing costs and risks. Another area, where risk management practices are yet to be introduced in a comprehensive manner is contingent liabilities of the Government. These missing links in risk management for internal, other domestic debt and external debt portfolios, as well as for the total portfolio in an integrated framework, point towards the pressing need for a middle office. 3.3 The issue of conflict in objectives between debt management and monetary policy has also been an area of concern while reviewing the institutional structure for debt management. 4.

International Experience

4.1 International evidence of debt management practice by leading debt office bears many valuable lessons for countries in the process of strengthening their debt management capacity. Many countries - mainly advanced and some emerging market economies - have set up public debt offices and are successfully managing their sovereign debts. In some instances, public debt has been brought down from what were clearly unsustainable levels. There has also been a significant change since late 1980s in the institutional structure, the role and style of functions of public debt management towards risk management. This has been enabled by institutionalisation of the debt office with an in-house risk management culture, as a specialised institution, staffed with professionals and market specialists. The role of such debt offices, in many instances, gradually transformed into treasury operations on the lines of operations performed by investment banks, corporates and foreign exchange management by central banks. Within the debt office, middle office emerges as the risk manager, which formulates and advises on the debt management strategy and also develops benchmarks for assessing the risk-cost trade off of the portfolio. Annex 4 summarises the key sound practices in sovereign debt management. 4.2 The primary requirement in any comprehensive debt management exercise is to bring the size of public debt at sustainable levels. Without sustainability of debt, risk management would not have much impact towards insulating the debt portfolio from systemic risks. Thus the primary task of the middle office is to determine sustainability benchmarks and accordingly advice the Government to ensure that the debt level is brought to sustainable levels. 7

The main risks that needs to be managed for the sovereign debt portfolio are foreign currency risk, interest rate risk, credit risk, liquidity risk, refinancing risk, operational risk and payments and settlement risk. Many debt offices have addressed management of market risks like currency and interest rate risk by establishing a risk management framework for the sovereign debt in an assetliability management framework. 4.3 A prudential risk management framework is essential for reducing uncertainty among sovereign debt managers as to the government’s tolerance for risk, its willingness to trade off cost and risk objectives. Once the risks are identified, risks and costs for alternative debt strategies are measured in a scenario-based model under a base case scenario and different market rate scenarios; or in a simulation-based model under value-at-risk, cost-at-risk or budget-at-risk approach. The government then chooses the strategy that best represents the government’s preferences for managing the risk/cost trade-offs and generally tend to choose it along an efficient frontier, which entails minimum risk. The process of deciding a debt strategy by debt offices has been facilitated, by using a “strategic benchmark” portfolio, which represent the approved strategy. The actual debt portfolio could be then moved closer to approved designed benchmark by debt managers while deciding on key terms for new debt issues, by buyback operations, and also by using currency and interest swaps and other hedging activities. 4.4 The institutional structure for public debt management, world wide, could be broadly characterised into two categories – setting up of a full-fledged public debt office and scattered debt management responsibilities. Within the former category of a full-fledged debt office, which is the case for most of the advanced countries and some emerging market economies, there has been a preference to locate the debt office as a separate entity under the Ministry of Finance or within the mainstream Ministry. There are also instances of locating the debt office outside the Ministry as an autonomous agency. On the other hand, Denmark is the only country where the debt office has been located within the central bank. The tendency to locate the debt office outside the central bank has been largely dictated by the overriding concern of the conflict in objectives that arise between debt management and monetary management. 4.5 Governance issues promoting sound and professional approach towards debt management, required debt offices to clearly define and disclose its objectives for debt management, establishing an organisational structure that ensures clear accountability and transparency of responsibilities with appropriate internal controls, and establishment of a legal framework wherever possible. For enabling sound risk management practices, most debt offices established prudent risk management strategy and policy, strengthening middle office analytical capability, and defining a framework for risk management ensuring consistency with other macroeconomic policies and objectives. Debt 8

offices also accorded priority to recruitment of trained staff, and selection and implementation of effective management information systems. 4.6 The second category of institutional structure reflects dispersed debt management responsibility, with the Ministry of Finance responsible for external public debt management and the central bank responsible for the internal public debt management. Some of the emerging market economies with dispersed debt management responsibilities, as a first step towards strengthening their debt management capacity has started to set up a middle office under the Ministry of Finance for evolving an over-all risk management framework for the total public debt. The overall aim, for such countries, is to establish a full-fledged debt office for effective debt management within a comprehensive framework of risk management. 5.

Recommendations

5.1 The Group recognised the need for a centralised middle office that could focus on debt management advisory activities in a comprehensive risk management framework. The middle office should operate on the lines of a modern “treasury” so that it may impart a professional approach by adopting modern management techniques for public debt management. Setting up such an office, however, cannot be done in isolation and is to be seen in the context of total structure comprising back, middle and front office functions to ensure that they are in harmony with each other. The middle office role to be successful, therefore, requires active interaction between concerned agencies responsible for debt management. 5.2 The scope of the middle office for managing the Government debt should include domestic debt components like internal debt and other domestic liability; as well as external debt components like debt on Government Account from external assistance, defence debt, and other sources of Government debt like IMF debt and foreign institutional investment in Government securities. In addition, it was felt that the scope of the middle office could gradually be expanded to include management of Government contingent liabilities, both explicit and implicit. 5.3 Thus, as a first step towards building a strong institutional mechanism for public debt management in India, the Working Group felt it necessary to establish a centralised Middle Office in the Department of Economic Affairs, Ministry of Finance. The main reason for locating it in the Ministry of Finance is that there are strong links between the budgeting and debt management functions. At the same time, it would help to mitigate conflicts in objectives between debt management and monetary policy.

9

5.4 Operations of the Middle Office could be supervised by a Public Debt Co-ordination Committee comprising of senior executives from the Ministry of Finance, the Reserve Bank of India and the functional head of the budget making entity. This would ensure that advisory role of the Middle Office would be respected by different entities involved for government debt management. 5.5 The middle office should be staffed with officials from the Ministry of Finance and drawing professionals on a deputation basis from the Reserve Bank of India and financial institutions. Investment in infrastructure and human resource development should be an area of priority for the Government to promote professional approach towards debt management. 5.6 The Group observed that ceilings for Government debt as determined in the recently introduced Fiscal Responsibility and Budget Management Bill 2000, would serve as a useful starting point for the Middle Office to develop debt sustainability benchmarks for bringing the level of Government debt to sustainable levels. 5.7 At the same time, a comprehensive risk management framework for portfolio management exercise for the Government debt portfolio should also be established by the middle office, based on cost-risk tradeoffs and the risk tolerance limits of the government. 5.8 Based on the risk-management framework and cost-risk trade-offs, the Middle Office would be expected to determine strategic benchmarks for the debt portfolio. The strategic benchmarks could be the proportion of domestic and foreign currency debt; the currency composition, duration, mix of floatingfixed interest rate debt and maturity structure of foreign currency debt portfolio; and maturity structure and duration for the domestic debt portfolio. 5.9 Strategic benchmarks, designed by the Middle Office, should have the approval of the Finance Minister. For this purpose, the Public Debt Coordination Committee should advice the Finance Minister periodically on the appropriateness of the framework and the strategic benchmarks. 5.10 Once the strategic benchmarks are approved, the Middle Office should regularly disseminate the relevant benchmarks to the concerned agencies involved for contracting or issuing government debt. This would enable them to determine their debt management strategy so as to be consistent with the strategic benchmarks. 5.11 In essence, the character of the centralised Middle Office would vary in terms of its responsibilities when viewed against different components of debt for which debt management responsibilities rest with different agencies. For the internal debt component managed by RBI, since there is already a middle office 10

in place, the role of centralised middle office would be to provide MIS inputs in terms of broad strategic benchmarks. Thus, while RBI would continue to have the flexibility for internal debt management, it could incorporate the strategic benchmarks while deciding the debt management strategy. For RBI to undertake active portfolio management consistent with the strategic benchmarks, would also require middle office capacity to be strengthened at RBI. Active coordination between the centralised middle office in the Ministry of Finance and middle office in RBI, through sharing of data and necessary information could also facilitate portfolio management. 5.12 For other debt components like other domestic liability and external debt, for which debt management responsibility lies with the Ministry of Finance, the centralised middle office would serve as the sole Middle Office for debt management. Strategic benchmarks designed for other domestic liability and external debt should be provided as part of detailed MIS inputs to different front offices for such debt components like the budget making entity and different credit divisions respectively. The front offices, in turn could fine-tune its debt management policy so as to make the debt portfolio consistent with the strategic benchmarks. 5.13 The Middle Office should be responsible for undertaking a consolidated review of the portfolio to assess the risk characteristics and to ensure that actual debt portfolio moves closer to the strategic benchmarks. For this purpose, it is essential for monitoring, compliance and control functions of the middle office are respected from the time the Middle Office is established. 5.14 The Middle Office should also act as the central unit for managing the contingent liabilities. While the budget making entity would continue to take the policy decisions regarding issuing guarantees and accumulating other contingent liabilities like pension funds, recapitalisation cost towards public sector enterprises, the Middle Office would try to shape such policies by incorporating them gradually into the risk management framework. This would also require active coordination with the budget making entity. 5.15 Other specific roles of the middle office would involve bringing out an annual report on public debt, which should enhance the transparency of the debt position of the central government. The report should clearly define and disclose the main objectives of debt management, the riskiness of the portfolio and performance of portfolio management by the relevant agencies as measured by the cost of the actual debt portfolio relative to the portfolio based on the benchmark. 5.16 For fair assessment of performance in portfolio management of the Government debt, by different agencies could be made by annually by an 11

independent auditor. Auditor’s assessment should be based on the cost of the actual debt portfolio relative to the benchmark portfolio. 5.17 Portfolio management exercises might require hedging activities for the government debt portfolio. Hedging activities, if undertaken, should be restricted purely to the need to achieve the strategic benchmarks and not involve tactical trading. 5.18 The middle office should also act as the apex monitoring unit for central government debt. Thus, data on government debt should be regularly transferred by different agencies to the middle office. The middle office should ensure that it develops a debt data recording system on a centralised platform, which is amenable for portfolio analysis. The debt data recording system should be preferably open-ended, so that, data requirements for applying new portfolio management techniques are easily available. 5.19 Setting up a centralized middle office, should be viewed as the starting point towards creation of a strong institutional system for prudent public debt management. For resolving the issue of conflict in objectives between internal debt management and monetary policy, and to redress the need for striking a synergy in debt management activities of different functional units may require setting up of a full-fledged debt office eventually. For this purpose, the Group felt that the institutional structure of leading debt offices, as outlined in the international experience (Annex 4) of the Report, could serve as an useful guide to the Government, as and when, it decides to review the need for setting up a full-fledged debt office.

12

Annex 1.

Constitution of the Working Group O.O.No.F. No.1(3)2000-EDMU Ministry of Finance Department of Economic Affairs External Debt Management Unit ---------------------------------------New Delhi, July 18, 2000.

Subject: Working Group to examine the need for setting up an integrated debt management “middle office” for public debt covering both external and domestic debt. A Working Group is hereby constituted to examine the need for setting up an integrated debt management “middle office” for public debt covering both domestic and external debt. The role of the “middle office” envisaged is to make analysis, provide advisory support and MIS inputs for debt management decisions. If the Working Group finds justification for such an Office, the Group would make recommendations regarding the scope, role and structure of the “middle office” and steps for setting up such office for debt management purposes. The proposal is to be examined in the context of overall effective debt management of the country keeping into perspective the possible long-term requirements. The Working Group would comprise the following: Chairman Member

-

Member Member Member

-

Member

-

Member Member-Convenor

-

-Dr. Arvind Virmani, Senior Economic Adviser, DEA, -Mr. D. Swarup, Joint Secretary (Budget), DEA, MOF or his nominee. -Dr. Tarun Das, Economic Adviser,DEA, MOF. -Dr. J. Bhagwati, Jt. Secretary(ECB),DEA, MOF -Mr. K. Shankar, Controller of Aid Accounts & Audit, DEA, MOF -Dr. T.C. Nair, General Manager, Internal Debt -Management Cell, RBI. -Mr. Deepak Mohanty, Director (DIF), DEAP, RBI -Mr. Anil Bisen, Director (EDMU), DEA, MOF

The terms of reference of the Working Group are as under: a) b) c) d) e) f) g)

to examine the need/justification for an integrated debt management “middle office” for public debt covering both domestic and external debt; If justification is found for setting up an integrated “middle office” for public debt, the Working Group would define the scope of the “middle office” taking into consideration Indian requirements and the best international practices; to specify role/functions/operations of the “middle office”; to review the functions of the existing institutional set-up for debt management; to propose structure of the debt office and link/interface with various concerned institutions/agencies like Ministry of Finance, Reserve Bank of India etc; to recommend steps and formalities for establishing the debt management “middle office”, and to make such other recommendations as the Working Group may deem appropriate on the subject; 13

The Working Group would hold necessary meetings and seek advisory support from multilateral / bilateral and other institutions, including investment banks, who have expertise in the area. The Working Group would submit its Report within 4 months from the day it is constituted. External Debt Management Unit shall act as the Secretariat for the Working Group. The Group may co-opt any other person as deemed appropriate. This issues with the approval of Secretary, Department of Economic Affairs. s/d (Tarun Das) Economic Adviser

14

Annex II. 2.1

Central Government Debt

Definition of Public Liability

2.1.1 As per the Government of India’s budgetary practice, there are three sets of liabilities, that constitute central government debt viz. (a) internal debt, (b) external debt, and (c) “other liabilities”. Internal debt and external debt constitute “public debt” of India and are secured under the Consolidated Fund of India. Internal debt includes market loans, special securities issued to the Reserve Bank of India (RBI), compensation and other bonds, treasury bills issued to RBI, state governments, commercial banks and other parties, as well as non-negotiable and non-interest bearing rupee securities issued to international financial institutions. External debt represents loans received from foreign governments and bodies under the external aid window and is usually classified as external debt on “Government Account”. Debt liabilities other than internal and external debt, termed as other domestic liabilities, include interest bearing obligations of the Government such as post savings deposits, deposits under small savings schemes, loans raised through post office cash certificates etc., provident funds, interest bearing reserve funds of departments like railways and telecommunications and certain other deposits. 2.1.2 The “other liabilities” of the Government arise in government’s accounts more in its capacity as a banker or trustee rather than as a borrower. Hence such borrowings, not secured under the Consolidated Fund of India, are shown as part of Public Account. Further more, some of the items of “other liabilities” like small savings are more in the nature of autonomous flows, which to a large extent are determined by public preference for relative attractiveness of these instruments. Nevertheless, it should be emphasised that such liabilities are contractual obligations of the Government and are economically indistinguishable from the public debt. Moreover, through a change in the accounting framework, it is possible to bring these liabilities under the Public Account within the ambit of public debt under the Consolidated Fund. In fact, under the new system of National Small Savings Fund (NSSF), with effect from the fiscal year 1999-2000, a substantial portion of “other liabilities” have been converted into Central Government securities. 2.1.3 The Report, therefore not only includes the above liabilities as components of debt of the Central Government, but also includes external debt incurred by borrowings from official agencies for defence purpose, including Rupee debt from the former Soviet Union, and also from IMF, which is outside the purview of the budgetary classification followed by the Government. External debt of the Government, referred to in this Report, would include external debt accrued in the “Government Account” as well as other sources of external debt. The term ‘public debt’ of the Central Government, referred to in 15

this Report would follow the usual budgetary practice, comprising of internal and external debt of the Central Government. The term ‘Government Debt’, referred to in this Report, would include internal debt, external debt and “other liabilities” of the Central Government. 2.2

Central Government Debt

2.2.1 Total stock of public liability of the Central Government has increased by more than three-fold, at an annual compound growth rate of 13.4 per cent, from Rs. 380,069 crore at the end of March 1991 to Rs. 1,176,174 crore at the end of March 2000 (Table 1). The growth in the stock of debt was steeper during the eighties, increasing at an annual compound growth rate of 19.9 per cent from end-March 1981 to end-March 1991, or by nearly six-fold. Table 1 : Total Debt of the Central Government (Rs. Crore) 1=2+3 Total Public Debt

2

3

4 5=2+4 6=1+4 Other Total Internal External Domestic Domestic Total Public Debt Debt* Liability Debt Liability (Rupees Crore) 1980-81 44343 30864 13479 17587 48451 61930 1990-91 251040 154004 97036 129029 283033 380069 1991-92 325270 172750 152520 144964 317714 470234 1992-93 369185 199100 170085 160555 359655 529740 1993-94 420866 245712 175154 184911 430623 605777 1994-95 453732 266467 187265 221215 487682 674947 1995-96 489768 307868 181900 247115 554983 736883 1996-97 520860 344476 176384 276961 621437 797821 1997-98 572997 388998 183999 333964 722962 906961 1998-99 655625 459696 195929 374856 834552 1030481 1999-2000 931660 728627 203033 244514 973141 1176174 (% of GDP)# 32.6 22.7 9.9 12.9 35.6 45.5 1980-81 44.1 27.1 17.1 22.7 49.8 66.8 1990-91 49.8 26.4 23.3 22.2 48.6 72.0 1991-92 49.4 26.6 22.8 21.5 48.1 70.9 1992-93 49.0 28.6 20.4 21.5 50.1 70.5 1993-94 44.9 26.4 18.5 21.9 48.3 66.8 1994-95 41.4 26.0 15.4 20.9 47.0 62.3 1995-96 38.2 25.3 13.0 20.3 45.6 58.6 1996-97 37.8 25.7 12.1 22.0 47.7 59.8 1997-98 37.2 26.1 11.1 21.3 47.3 58.5 1998-99 48.2 37.7 10.5 12.7 50.4 60.9 1999-2000 * : External Debt for end-March 1981 refers to outstanding external debt on “government account” and does not include other components of government external debt like IMF and Defence debt. # : Debt to GDP ratios are based on the New Series of GDP estimates at market prices published by the Central Statistical Organisation. Source : India’s External Debt – A Status Report, 2000; RBI Annual Report, 1999-2000; and Economic Survey; 1997-98.

16

2.2.2 The dynamics of public liability in terms of a “debt-GDP crossover”, as measured by the annual growth rate in public liability and the nominal GDP growth, could be broadly identified into three phases during the nineties, i.e., 1992-93 to 1993-94, 1994-95 to 1996-97, and 1997-98 to 1999-2000 (Figure 1). The annual growth in public liability by 23.7 per cent during 1991-92 stood higher than the nominal GDP growth. The first phase (1992-93 to 1993-94) showed a declining trend in the growth of public liability, growing at an average annual rate of 13.5 per cent and was marginally lower than the nominal GDP growth. The growth rate in public liability further decelerated during the second phase (1994-95 to 1996-97) by 9.6 per cent and was significantly lower than the GDP growth. There was a reversal in the declining trend in the growth of public liability, increasing again by 13.8 per cent during the third phase (1997-98 to 1999-2000), with the wedge between debt-GDP growth contracting sharply and crossed the GDP growth rate during 1997-98 and 1999-2000. As a proportion of GDP, the total public liability, which stood at 45.5 per cent at end-March 1981, accordingly declined from a peak of 72.0 per cent during 1991-92 to 58.6 per cent during 1996-97 before increasing to 60.9 per cent during 1999-2000. Figure 1 : Nominal Debt Growth - GDP Growth Crossover 25.0

20.0

(per cent)

15.0

10.0

5.0

0.0 1991-92

1992-93

1993-94

Total Govt. Debt

1994-95

1995-96 Nominal GDP

1996-97

1997-98

1998-99

1999-2000

Total Domestic Debt

2.2.3 The debt-GDP growth crossover witnessed during the third phase was propelled by growth in domestic debt of the central government. Increasing reliance on domestic sources of borrowing mainly in the form of market borrowings resulted from the changing nature of the fiscal position with reemergence of primary deficit since 1997-98 (Figure 2). While the share of market borrowings financing the gross fiscal deficit more than doubled from 30.0 per cent in 1996-97 to 70.8 per cent in 1999-2000, the share of internal 17

finance increased steadily from 85.1 per cent in 1990-91 to a high of 99.2 per cent in 1999-2000. Although the fiscal adjustment process initiated in 1991was successful in arresting the growth in primary deficits during 1991-92 and 199293, the resurgence in primary deficits since 1997-98 led to a steady accumulation of domestic debt and market borrowings thereby reversing the declining trend in the debt-GDP ratio. While the domestic component of the government liability increased marginally from 45.6 per cent as a proportion of GDP in 1996-97 to 50.4 per cent in 1999-2000, internal debt, which accounted nearly half of the market loans, increased its share from 25.3 per cent to 37.7 per cent during the corresponding period. This raises concern about the sustainability of the central government debt and in particular about the domestic component and internal debt position. F ig u re 2 : F in an cin g o f G ro ss F iscal D eficit 100

6

90

(Borrowings as % of Gross Fiscal Deficit)

70

4

60 50

3

40 2

30 20

(Primary Deficit as % of GDP)

5

80

1

10 0

0 1980-81

1985-86

1990-91

1991-92

Market B orrowin g s

1992-93

1993-94

1994-95

1995-96

Total In tern al F in an ce

1996-97

1997-98

1998-99

19992000

P rim ary Deficit - G DP R atio

2.2.4 The share of domestic component of total public liability increased steadily from 67.6 per cent at the end of March 1992 to 82.7 per cent at the end of March 2000, mainly due to a sharp increase in internal debt (Figure 3). Internal debt increased at an annual compound growth rate of 18.8 per cent during end-March 1991 to end-March 2000 or by nearly five-fold. Accordingly, the share of internal debt in the total public liability of the government increased from 40.5 per cent at end-March 1991 to 44.6 per cent at end-March 1999. The share jumped to 62.0 per cent at end-March 2000 because of the sharp increase in the growth of internal debt by 58.5 per cent during 1999-2000. The sharp increase in internal debt during 1999-2000 was however, due to conversion of other liabilities amounting to Rs. 1,80,273 crore into Central Government securities, under the new system of National Small Savings Fund (NSSF), with 18

effect from the fiscal year 1999-2000. Accordingly, the share of other liabilities in total public liability of the central government increased from 34.0 per cent in end-March 1991 to 36.4 per cent in end-March 1999 and declined thereafter to 20.8 per cent in end-March 2000. Figure 3 : Components of Total Government Debt Outstanding

100%

80%

60%

40%

20%

0% 1990-91

1991-92

1992-93

1993-94

Internal Debt

1994-95

1995-96

Other Domestic Liability

1996-97

1997-98

1998-99

1999-2000

External Debt

2.2.5 On the other hand, total external debt of the government witnessed an absolute decline from US $ 50.0 billion at end-March 1991 to US $ 46.5 billion at end-March 2000. However, in terms of the local currency, the external component of government debt increased at an annual compound growth rate of 8.5 per cent during end-March 1991 to end-March 2000 mainly due to the depreciation of the rupee. In spite of the modest increase in rupee terms, the share of the external debt in total public liability declined steadily from a peak of 32.4 per cent in end-March 1992 to 17.3 per cent in end-March 2000. The ratio of external debt to GDP also declined from a peak of 22.8 per cent in 1991-92 to 10.5 per cent in 1999-2000. 2.2.6 Public debt, which includes internal and external debt of the government, also grew in consonance with the phase in growth of total government debt as identified above (Figure 1). Increasing by 29.6 per cent during 1991-92, the average growth rate in stock of public debt moderated during 1992-93 to 199394 by 13.7 per cent, further decelerated by 7.4 per cent during 1994-95 to 199697, and finally reversed the trend by increasing during 1997-98 to 1998-99 by 12.2 per cent. The spurt in public debt during 1999-2000 by 42.1 per cent resulted from the conversion of other domestic liabilities into government securities under the new system of National Small Savings Fund (NSSF), with effect from the fiscal year 1999-2000. Accordingly, the ratio of public debt 19

declined from a peak of 49.8 per cent in 1991-92 to 37.2 per cent in 1998-99 before increasing to 48.2 per cent in 1999-2000. 2.3

Debt Burden of the Central Government

2.3.1 The overhang of government debt during the nineties, particularly domestic debt poses significant risk for medium-term macro-economic stability of the economy. For internal debt management, the large scale of market borrowings has constrained the leverage for minimising the borrowing cost due to increase in yields driven by interest rate premium and a shift in preference for short-term paper caused by increasing uncertainty about future interest rates. This has led to a policy of placing borrowings at the shorter end of the market, so as to minimise the cost of borrowing. Thus the share of market borrowings in shorter maturity securities (loans maturing within a period of 5 years) was as high as 50.0 per cent during 1996-97 and 36.9 per cent during 1997-98. As a consequence, the share of the shorter maturity market loans in total outstanding market loans increased sharply from 7.4 per cent in end-March 1992 to 45.2 per cent in end-March 1998. 2.3.2 Recognising the roll-over problems arising from the bunching of redemption in the medium term, the maturity structure was lengthened with the introduction of long-term securities (ranging from 11 year to 20 year maturity) during 1998-99. While borrowing in longer-term maturity accounted for 13.5 per cent of the total market borrowings during 1998-99, the share of the shorter maturity declined to 31.5 per cent. This was followed by placing all government borrowings above 5-year maturity and about 65.0 per cent of the borrowings through issuance of above 10-year maturity during 1999-2000. The weighted average maturity of market loans thereby increased from 5.5 years in 1996-97, to 6.6 years in 1997-98, 7.7 per cent in 1998-99 and 12.6 per cent in 1999-2000.

20

Figure 4 : Residual Maturity of less than 1-year for Market Loans & Govt. Account External Debt 9.0

8.0

7.8

7.0

6.8

(per cent)

6.0

5.9

5.7

4.8

5.0

4.1

4.3

4.5 4.2

4.0

4.5

4.4

5.0

4.9

4.9

3.0

2.0

1.4

1.0

0.9

0.0 1992-93

1993-94

1994-95

1995-96

1996-97

Market Loans

1997-98

Govt. Account External Debt

21

1998-99

1999-2000

2.3.3 Notwithstanding the success in lengthening the maturity structure of dated securities, the immediate redemption pressure (i.e. market loans maturing within 1-year) on market loans increased. The share of market loans with residual maturity of less than 1-year increasing from 0.9 per cent in end-March 1994 to 7.8 per cent in end-March 2000 (Figure 4). The magnitude of the rollover problem is also reflected from the future redemption profile of the stock of market loans as on end-March 2000 entailing large redemptions during the next ten years (Figure 5). Figure 5 : Redemption Profile of Domestic Market Loans & External Debt as on 31.03.2000

45000

40000 10807 35000

11247

11186

10329 9411

10951

8907

8254

9898 8024

(Rs. Crore)

30000

25000

20000

15000

31252

28321

28260

28263

2000-01

2001-02

2002-03

31159

29394

27473

30151

30223 26195

10000

5000

0 2003-04

2004-05

Domestic Market Loans

2005-06

2006-07

2007-08

2008-09

2009-10

External Debt

2.3.4 On the other hand, the maturity profile of external debt of the government at end-March 2000 is skewed towards long-end maturity with 51.2 per cent of the debt over 10-year residual maturity and only 26.9 per cent under 5-year residual maturity. The stock of total external debt with a residual maturity of less than 1-year amounted to only 5.5 per cent of the total debt stock in endMarch 2000. The stock of external debt on Government Account with a residual maturity of less than 1-year increased marginally from 4.1 per cent in endMarch 1993 to 4.5 per cent in end-March 2000 (Figure 4). This is mainly due to the long-tenor of external aid in the form of soft loans and is reflected by the smooth redemption profile of the portfolio as on end-March 2000 (Figure 5). 2.3.5 The adverse debt burden of the government is also reflected by a persistent increase in the interest burden on internal debt, the bulk of which, is attributable due to interest payments on market borrowings (Figure 6). Thus, while the share of interest payments in total expenditure increased from 19.9 per 22

cent in 1990-91 to 31.7 per cent in 1999-2000, the share in tax receipts increased from 50.1 per cent in 1990-91 to 76.2 per cent in 1999-2000. The deterioration was also due to stagnation in the share of tax-GDP ratio. The share of interest payments also increased from 39.2 per cent of revenue receipts in 1990-91 to 53.7 per cent in 1999-2000. The heavy outgo on interest payments had serious implications for the fiscal position of the government. Figure 6 : Interest Burden on Government Debt 90.0

80.0

76.2

74.4 70.0

65.3 61.1

(per cent)

60.0

50.0

50.1

40.0

39.2

30.0

21.2

10.0

11.3

45.4

47.1

53.7

52.1

49.0

51.9

40.4

28.1

20.0

63.5

53.2 48.4

72.3

68.6

27.4

23.9

28.1

29.6

28.3

31.7 27.9

31.2

19.9

0.0 1980-81

1990-91

1991-92

1994-95

(as % of Tax Revenue)

1995-96

1996-97

(as % of Revenue Receipts)

1997-98

1998-99

1999-2000

2000-2001

(as % of Total Expenditure)

2.3.6 While interest payments on internal debt increased by nearly eight-fold during 1990-91 to 1999-2000, more than the six-fold increase in the stock of debt during the same period; the total interest payments increased by six times during the same period. On the other hand, interest payments on external debt grew moderately, reflecting the concessional nature of such loans. Thus, the share of interest outgo on internal debt increased gradually from 42.5 per cent of total interest payments in 1990-91 to 51.1 per cent in 1998-99 (Figure 7). The share increased sharply to 75.4 per cent in 1999-2000. While the share of other domestic liabilities in total interest payments remained stable around 45.0 per cent during 1990-91 to 1998-99, it declined sharply to 20.0 per cent in 19992000. On the other hand, the share of external debt in total interest payments declined from 8.7 per cent in 1900-91 to 4.6 per cent in 1999-2000.

23

Figure 7 : Share in Total Interest Payments 100%

80%

60%

40%

20%

0% 1980-81

1990-91

1991-92

1994-95

External Govt. A/c

1995-96

Internal Debt

1996-97

1997-98

1998-99

1999-2000

Other Liabilities

2.3.7 Interest payments, inter alia, stemmed from larger recourse to market borrowings at market-related interest rates. Since the interest rates on government securities were somewhat aligned to the market during the latter half of the eighties, and market related interest rates were offered on government securities since 1992-93, the borrowings from the market were mobilised at higher interest rates. The weighted average interest rates on dated securities rose from 7.03 per cent in 1980-81 to 11.41 per cent in 1990-91 and further to 13.75 per cent in 1995-96 (Table 2). However, the internal debt management operations were successful in stabilising the interest rates at a relatively low level as reflected in the weighted average interest rate of the dated securities at 11.77 per cent in 1999-2000 and 11.86 per cent in 1998-99. The average implicit interest rates on other domestic liabilities, which mainly comprises of small savings and provident fund, also increased from 7.22 per cent in 1980-81 to 10.81 per cent in 1990-91 and further to 12.15 per cent in 1997-98. Thus, the implicit nominal interest on overall domestic debt of the central government increased steadily from 8.65 per cent in 1990-91 to 10.69 per cent in 1999-2000. Thus while the debt structure changed from low cost to high cost constituents, the rising interest rates on such borrowings also raises concern about the sustainability of debt. The sustainability of domestic debt could also be viewed from the perspective of real interest rate on such components, particularly when the real interest rate exceeds the real growth rate of GDP as witnessed during 1997-98.

24

Table 2 : Nominal Interest Rate on Domestic Debt of Central Government Fiscal Year Weighted Average Implicit Nominal Implicit Nominal Interest Rates on Interest Rates on Interest Rates on Marketable Securities Small Savings and Total Domestic Debt Provident Funds 1990-91 11.41 10.81 8.65 1991-92 11.78 11.28 9.09 1992-93 12.46 11.06 9.38 1993-94 12.63 12.38 9.72 1994-95 11.91 12.67 9.72 1995-96 13.75 11.72 9.82 1996-97 13.69 12.70 10.39 1997-98 12.01 12.15 10.22 1998-99 11.86 N.A. 10.45 1999-2000 11.77 N.A. 10.69 Source : RBI Annual Reports. 2.3.8 Internal debt management was till recently characterised by undertaking the bulk of market borrowing programme during the first six months, when there would be less pressure on the banks to lend to the commercial sector. In terms of the magnitude, this implied that the central bank had to manage total Government borrowing every month to roughly 1 per cent of GDP during the first half of the year. During the last three years, saddled with bulging repayment obligations and objective to raise debt at minimum cost, debt management operations had to maintain balance between changing the maturity mix of borrowings and deriving maximum benefit of liquidity conditions. The increasing coordination of debt management and monetary policy resulted in fairly active open market operations, with the central bank releasing the privately placed government stocks when interest rate expectations became favourable. 2.4

Net Liability of the Central Government

2.4.1 For assessing the indebtedness position of the government, net outstanding debt position is also a useful indicator. The net domestic liability of the Central Government, has been derived after deducting the book value of the domestic financial assets from the total domestic debt. Net domestic debt of the Central Government, increased by more than eight-fold during 1990-91 to 1999-2000 (Table 3). On the other hand, net external debt of the central government declined by nearly 50.0 per cent during the same period. Net external debt of the Central Government has been derived by deducting foreign exchange assets of the central bank from the total external debt of the central

25

government. However, the total net liability registered an increase of 3 fold during 1990-91 to 1999-2000. Table 3 : Net Liability of the Central Government (Rupees Crore) Net Total Liability Net Domestic Liability Net External Liability

1990-91 1995-96 1996-97 1997-98 1998-99 1999-2000 138941 281458 284323 325680 368010 445594 46293

158004

188307

244188

297493

395485

92648

123454

96016

81492

70517

50109

2.4.2 In terms of liability asset ratio also, while the external liability-asset ratio improved from 2211.4 per cent in 1990-91 to 132.8 per cent in 1999-2000, the domestic liability asset ratio deteriorated from 119.6 per cent to 168.5 per cent during the same period. Accordingly, the total liability-asset ratio increased from 157.6 per cent in 1990-91 to 161.0 percent in 1999-2000. Figure 8 : Liability-Asset Ratio for Central Government 350.0

311.2 300.0

250.0

(per cent)

219.5 200.0 161.8 150.0

139.8

155.4

179.5 156.0 151.0

143.5

168.5

156.2 155.4

161.0

155.6 132.8

100.0

50.0

0.0 1995-96

1996-97 Domestic Liability-Asset Ratio

2.5

1997-98 External Liability-Asset Ratio

1998-99

1999-2000

Total Liability-Asset Ratio

Contingent Liabilities of the Central Government

2.5.1 Contingent liabilities of the Central Government arise because of its role to promote private sector participation in infrastructure projects by issuing guarantees. Contingent liabilities of the Central Government could be both domestic or external contingent liabilities and could also be explicit or implicit in nature. Domestic contingent liability of the Central Government constitute 26

direct guarantees on domestic debt, recapitalisation costs for public sector enterprises, or unfunded pension liabilities. External contingent liability constitute direct guarantees on external debt, exchange rate guarantees on external debt like Resurgent India Bonds and Indian Millenium Deposits, and counter-guarantees provided to foreign investors participating in infrastructure projects. Although from the accounting point of view, the contingent liabilities do not form part of the Government debt, it could pose severe constraints on the fiscal position of the Government in the event of default. Table 4 : Contingent Liability of the Central Government (Rupees Crore) Year Domestic Guarantees on Total (end-March) Guarantees External Debt Guarantees 1994 62834 38159 100993 1995 62468 38830 101298 1996 65573 34922 100495 1997 69748 29576 99324 1998 73877 28810 102687 1999 74606 30050 104656 Source : RBI Annual Report 1999-2000 and Status Report on External Debt, 2000.

2.5.2 The total outstanding direct credit guarantees issued by the Central Government both domestic as well as on external debt remained stable around Rs. 100,000 crore during end-March 1994 to end-March 1999 (Table 4). While domestic guarantees increased modestly during the corresponding period, there was an absolute decline in the guarantees on external debt. As a proportion of GDP, however, both domestic guarantees and external guarantees registered a declined of 3 per centage points during end-March 1993-94 to 1998-99 (Figure 9). Thus, the total guaranteed debt of the central government declined steadily from 11.8 per cent of GDP in 1993-94 to 1998-99.

Figure 9 : Contingent Liability of the Central Government 14.0

12.0

11.8

(per cent of GDP)

10.0 10.0

4.4 8.5 3.8

8.0

7.3 3.0 2.2

6.0

6.8 5.9 1.9 1.7

4.0

7.3 6.2

5.5

5.1

4.9

1997

1998

2.0

4.2

0.0 1994

1995

Domestic Guarantees

1996

Guarantees on External Debt

27

Total Guarantees

1999

2.5.3 In addition, exchange rate guarantee on external debt, also have implications for finances of the Central Government. For example, for Resurgent Indian Bonds, as per the agreement, exchange rate loss in excess of 1 per cent on the total foreign currency raised equivalent to US $ 4.2 billion, would have to be borne by the Government of India. The extent of such loss, since August 1998, the time when RIB were raised, up to August 2000 amounted to Rs. 946 crore. The actual loss on such liability would depend upon the exchange rate prevailing at the time of redemption in 2003. Very recently, a similar exchange rate guarantee was provided on the amount of US $ 5.5 billion raised through India Millenium Deposits, during October-November 2000. For counter guarantees provided to foreign investors participating in infrastructure projects, similar risk arises for the Government exchequer. At the same time, there is a growing volume of implicit domestic contingent liabilities in the nature of pension funds. 2.6

Legal Ceilings on Government Debt

2.6.1 The Indian constitution under Article 292 provides for placing a limit on public debt secured under the Consolidated Fund of India but precludes “other liabilities” under Public Account. However, through a change in the accounting framework, it is possible to convert these liabilities under Public Account within the ambit of public debt under the Consolidated Fund and thereby within the ambit of Article 292. Given the legacy of huge public debt and interest burden due to a long history of high fiscal deficits, which has increasingly constrained the maneuverability in fiscal mangement, the Central Government has recently introduced a Fiscal Responsibility and Budget Management Bill, 2000 in the parliament. The proposed bill aims to ensure inter-generational equity in fiscal management and long-term macro-economic stability. This would be achieved by achieving sufficient revenue surplus, eliminating fiscal deficit, removing fiscal impediments in the effective conduct of monetary policy and prudential debt management consistent with fiscal sustainability through limits on central government borrowings, debt and deficits, and greater transparency in fiscal operations. The specific targets for debt management in this regard is to ensure that the total liabilities of the central government (including external debt at current exchange rate) is reduced during the next ten years and does not exceed 50 per cent of GDP. Simultaneously, the Central Government shall not borrow from the RBI since April 1, 2003 in the form of subscription to the primary issues by the latter. In the meantime, the Government may continue to borrow from the RBI by ways and means advances to meet temporary excess mismatch between disbursement and receipts in accordance with the agreements entered into between them. 2.6.2 The bill also addresses to check the contingent liability by restricting guarantees to 0.5 per cent of GDP during any financial year. In particular, 28

transparency in budget statements would involve disclosure of contingent liabilities created by way of guarantees including guarantees to finance exchange risk on any transactions, all claims and commitments made by the central government having potential budgetary implications.

29

Annex III.

Existing structure for public debt management

At present, public debt management in India is spread between different divisions/units in the Ministry of Finance (MOF) and the Reserve Bank of India (RBI). The chart below (Table 3.1) highlights the institutional structure and major responsibilities for public debt management. 3.1 Internal Public Debt Management: The Reserve Bank of India acts as the Government’s agent for internal debt management responsibilities as per the RBI Act, 1934 and the Public Debt Act, 1944. The main units associated with internal debt management in the RBI are the Internal Debt Management Cell (IDMC); Public Debt Offices (PDO); and Securities Department in the Department for Government and Bank Accounts (DGBA). Front office roles like debt issuance and debt service payments are performed by Public Debt Offices and by the IDMC for open market operations. Middle Office roles are mainly entrusted with the IDMC. Such responsibilities include evolving appropriate policies relating to internal debt management, like timing, amount of debt issuance or repurchase and fixation of interest rate according to state of liquidity and expectations of the market; cash and liquidity management; besides promoting an active and efficient government securities market. In doing so, the IDMC consults the Budget Division in the Ministry of Finance on key issues. In addition, periodic analysis on internal debt by the IDMC is complemented by the Department of Economic Analysis and Policy in the RBI by contribution in annual publications like the RBI Annual Report and Report on Currency and Finance and periodic research publications like Development Research Group and RBI Bulletin. 3.2 Other Domestic Liability Management: As outlines earlier, central government debt also includes other domestic liabilities like interest bearing obligations of the government such as post savings deposits, deposits under small savings schemes, loans raised through post office cash certificates etc.; provident funds; interest bearing reserve funds of departments like railways and telecommunications; and certain other deposits. The key agency involved for policy issues on such components is the Budget Division in the Ministry of Finance. Front and back office roles are however dispersed across different Departments/Ministries of the Government of India. 3.3 External Public Debt Management: As far as the external public debt is concerned, the Government of India has so far borrowed from external assistance sources only in the form of loans/credits from bilateral and multilateral sources. The responsibility for management is with the Department of Economic Affairs in the Ministry of Finance. For this purpose, front office 30

roles are mainly entrusted with various credit divisions within the Department of Economic Affairs (DEA), Ministry of Finance for negotiating and contracting loans from different bilateral and multilateral agencies/countries. The O/o Controller Aid Audit & Accounts in the Department of Economic Affairs, Ministry of Finance is also partially involved with front office roles in respect of ensuring disbursement of loans/credits. In addition, the Ministry of Defence also acts as the front office and back office for defence debt from external sources. External Debt Management Unit (EDMU) in the Department of Economic Affairs, Ministry of Finance acts as the middle office for external public debt. Major functions of EDMU include monitoring of public external debt and direct contingent liabilities; providing MIS inputs for debt management decisions e.g., choice of currency, interest and maturity mix; and bringing out an annual Status Report on External Debt, which also covers public external debt. Back office responsibility for data recording and accounting lies with the O/o CAA&A in the Ministry of Finance for external assistance and with the Ministry of Defence for defence debt. 3.4 The need for a Middle Office also stems from recent developments in the international financial markets. IBRD borrowers, including sovereigns, are now required to choose currency, interest and maturity mix of their borrowing on IBRD debt. IBRD is also going to offer derivative products like currency and interest swap, caps, collars etc to the borrowers to allow them to actively manage their IBRD portfolio. ADB is also contemplating similar market friendly loan products. Therefore, the management of Government external debt would now require continuous studies on benchmarks regarding currency, interest rate and maturity mix. For this purpose, a sovereign external debt modeling exercise has been initiated recently by EDMU, in the Ministry of Finance.

31

Table 3.1 : PRESENT STRUCTURE of PUBLIC DEBT MANAGEMENT IN INDIA

Office Levels (1) FRONT OFFICE: Carry out all financial transactions in the money and capital markets – 1) Borrow (issue securities); 2) Make debt service payments. MIDDLE OFFICE: Advice on debt management strategy to be adopted. This includes – 1) Develop debt sustainability benchmarks. 2) Determine benchmarks for currency, interest rate, and maturity mix as part of risk management strategy. 3) Decide on borrowing instruments, timing etc.

External Debt

Domestic Debt

(2)

(3) A. Primary MarketPublic Debt Office in RBI carries out functions like debt issuance and debt service payment.

1) Credit Divisions

In MOF (like FundBank Division, ADB, EEC, etc.) negotiate and contract new loans. 2) O/o CAA&A in

MOF is entrusted with debt servicing and disbursements. EDMU in MOF monitors public external debt and direct contingent liabilities; brings out an annual Status Report on External Debt which also covers public external debt. Provides MIS inputs for debt management decisions e.g., choice of currency, interest and maturity mix. Also working on a sovereign external debt modelling exercise.

Missing Links / Action Areas (4)

B. Secondary MarketIDMC, RBI entrusted with open market operations. IDMC in RBI in consultation with Budget Division in MOF evolves appropriate policies relating to internal debt management, like timing and amount of debt issuance or repurchase; cash and liquidity management; promoting an active and efficient government securities market. IDMC and DEAP in RBI undertake public debt management related studies (published in RBI Annual Report and Report on Currency & Finance). Economic Division in MOF also performs the same role (Economic Survey Budget).

1. Developing public debt sustainability benchmarks; medium and long-term public debt management strategy. 2. Integrate internal and external debt. This would mean the following : a). Choice between Domestic and External debt. b). Developing benchmarks for currency, interest and maturity mix as part of reducing costs and risks. 3. Debt data management – a) Comprehensive computerization b) Data analysis c) Scenario exercises and sensitivity testing. 4. Transparency – Reports on Public Debt. 5. Ready availability of Public Debt information for the Parliament and the top management in Ministry of Finance.

BACK OFFICE: Record keeping, accounting and systems support.

O/o CAA&A does record-keeping and accounting functions for GOI’s external debt.

Central Debt Division, DGBA in RBI maintains all figures, acts as a controlling office for PDOs and extends general supervision over their working.

32

3.5 Missing links in middle office functions for public debt management: An analysis of Existing Structure characterises the “mixed” structure and responsibilities for the public debt management with several units in the Ministry of Finance and RBI discharging the responsibility. As mentioned above, in its effort to raise resources at minimum cost, the focus of internal debt management by the RBI has been on liquidity management and refinancing risk. The scale of annual borrowings by the government and the conflicts in objectives arising between dent management and monetary management constrained the latter for evolving comprehensive benchmarks and undertaking necessary risk management exercise for the debt portfolio. This in turn could be attributed to the fact government debt market is still evolving and there has often been incidence of devolvement of government securities on RBI. The secondary market still lacks the depth and breadth for enabling RBI to undertake active open market operations necessitated by portfolio management objectives. At the same time, substantial initial subscription of Government securities by the RBI meant that such securities have to be offloaded subsequently when the market conditions and investor apetite are conducive. This often led to potential for wrong signals for monetary management. Although, through active coordination between debt management and monetary management, RBI often succeed in striking a fine balance for minimising the conflicts in objectives, sometimes one of the objective had to be sacrificed. Thus, while for internal debt management the emphasis has been on front and back office functions, the middle office role is somewhat lacking, for both domestic and external debt components. Therefore, what is missing is a debt strategy so as to maintain public debt at sustainable levels and its management by developing a comprehensive risk management framework. Also conspicuous by its absence is an integrated approach – covering both domestic and external debt - towards public debt management. Thus public debt management fell short of developing public debt sustainability benchmarks; medium and long-term public debt management strategy; making a choice between domestic and external debt; developing benchmarks for currency, interest and maturity mix as part of reducing costs and risks. 3.6 Co-ordination among debt management activities: Another aspect, which could have prevented an integrated approach for middle office might be appointment of different Committees/Groups on an ad hoc basis to look at different aspects of public debt management. A list of such Groups/Committees are at Table 3.2.

33

Table 3.2 : List of Working Groups/Committees etc. currently looking at different aspects of Public Debt Management in India. A.

B.

C.

External Public Debt 1. Steering Committee and Core Group on Sovereign External Debt Modelling Exercise. (Secretariat at EDMU, MOF). 2. Working Group on Reducing Debt Service Cost of all External Debt on Government Account. (Secretariat at O/o CAA&A, MOF). 3. Working Group on “Reducing debt service cost of External debt on Govt. Account (Secretariat at ECB Division, MOF). Domestic Public Debt 1. Standing Committee on Cash and Debt Management. (Secretariat at Budget Division, MOF). 2. Technical Advisory Committee on Money and Government Securities Markets. (Secretariat at IDMC, RBI). Other Related Committees 1. Steering Committee on External Debt. (Secretariat at EDMU, DEA, MOF). 2. Monitoring Group on External Debt. (Secretariat at DEAP, RBI). 3. Permanent Technical Group on Reconciliation of External Debt Statistics with International Agencies. (Secretariat at O/o CAA&A). 4. Permanent Group on External Commercial Borrowings data. (Secretariat at DESACS, RBI).

3.7 Conflict in objectives between debt management and monetary policy: The issue of conflict in objectives between debt management and monetary policy has also been an area of concern for macro-economic management. Accordingly, separation of debt management and monetary policy functions were recommended by an informal RBI Working Group Report on Separation of Debt Management from Monetary Management. This issue was vindicated by a recent RBI Report of the Advisory Group in Transparency in Monetary and Financial Policies. More recently, the Report of the Committee on Fiscal Responsibility Legislation, which formed the basis for the introduced Fiscal Responsibility and Budget Management Bill, 2000; recognised that participation of RBI in primary issues of Government securities has constrained the maneuverability of RBI in using the monetary policy instruments, particularly the open market operations and the bank rate, for pursuing the goal of macroeconomic stability. The Committee accordingly recommended that freeing RBI from debt management functions should be pursued to accord greater operational flexibility to RBI for conduct of monetary policy and as part of fiscal responsibility. The proposed stance of doing away with borrowings from the central bank three years hence and limiting such borrowings during the interim period, in the introduced Bill would also contribute towards reducing such conflicts. While the central bank, through active co-ordination between debt management, monetary management and exchange rate policy has succeeded in maintaining a balance between ensuring that borrowing requirements are met at a minimum cost and ensuring monetary stability, this was often at the cost of sacrificing indirect instruments of monetary policy. 3.8 Contingent Liability Management: Although there is no centralised unit for managing the guarantees provided by the Government of India, the Budget Division in the Ministry of Finance acts as the Middle Office by 34

monitoring such liabilities and formulating policies relating to such guarantees. The Budget Division also monitors the total guarantees of the Government of India on a consolidated basis. Front and back office roles, on the other hand, is dispersed across several Ministries/Departments of the Government of India.

35

Annex IV.

INTERNATIONAL EXPERIENCE AND BEST PRACTICES IN PUBLIC DEBT MANAGEMENT

4.1 Approach in Study of Best International Practices: International evidence of debt management practice by leading debt offices bears many valuable lessons for countries in the process of strengthening their debt management capacity. While there is no unique answer as to what constitute sound debt management practice, selective discretion should be used while ameliorating debt management practice based on international experience. More importantly, the country specific requirements should be carefully analysed so that the international best practices could be grafted effectively. For studying the best international practices, Commonwealth Secretariat, London was appointed consultant for the project, mainly for providing inputs on “international best practices” in the area of public debt management. The Commonwealth Secretariat, hired Mr. Nihal Kappagoda, an international expert, who submitted a Report on the subject. The summary of the recommendations of the Commonwealth Secretariat is at Annex 5. Chairman and Members also visited select public debt offices to get first hand experience of their operations. The list of debt offices and officials visited by the Indian delegation is at Annex 6. The specific issues examined by the Group on best international practices related to the risk-management framework, the institutional and organisational structure for public debt management, governance issues promoting professional approach, co-ordination with fiscal and monetary management agency with the debt management agency, and capacity building in sovereign debt management. 4.2

Risk Management Framework for Government Debt

4.2.1 Sovereign debt management primarily aims to ensure that government borrowing needs are met efficiently. A second objective is to ensure that the stock of debt portfolio and incremental flows arising from budgetary and offbudgetary sources are being managed in a manner consistent with the government’s preferences for cost and risk. An objective of minimising debt servicing cost, irrespective of risk, should not be an explicit objective. Risky debt structures, characterised by excessive exposure to short-term or floatingrate debt or debt denominated in or indexed to foreign currency can substantially deteriorate the fiscal position of the government, constrain access to capital and even propagate financial market instability. This is particularly important when the government’s debt portfolio is large relative to the economy’s output. Prudent debt management aimed at reducing risks by establishing a low risk currency composition, interest structure and maturity profile of the government’s debt portfolio could make countries less susceptible to financial risk and contagion. Several OECD governments have accordingly set government debt management objectives aimed at minimising debts servicing costs over the medium and longer-term subject to a prudent level of portfolio risk. On the other hand, for many emerging market economies, debt 36

management objectives appear to be to cover their borrowing needs, lengthen maturities and diversify funding sources wherever possible. Less attention is paid to managing market risks and refinancing risk. 4.2.2 Table 4.1 summarises several risks arising from the debt portfolio of the government, which sovereign debt managers endeavor to manage. Depending on the characteristic of the debt portfolio, the debt manager might accord different degree of priority for managing different risks. The portfolio choices would also depend on the macroeconomic policies given the strong interlinkages between debt management, fiscal, monetary and exchange rate policy. Thus, to the extent possible, the day to day implementation of sound debt management policies should seek to reinforce the objectives of macroeconomic policies and policy reforms aimed at improving the efficiency of the domestic financial market. Understanding the interplay of these public policies and considerations and the technical analysis and market judgement involved in managing what are often very large and complex portfolios, make sovereign debt management a highly specialist business within the government. 4.2.3 Usually, governments are risk-averse in their sovereign debt management, often because governments have strong political incentives to adopt the risk apetite reflected by the “median voter” decision making. Evidence suggests that taxpayers or representative voters tend to be risk averse in their decisionmaking and expect the government to have similar risk apetite in managing its financial interests. Thus, while governments generally have preference for more stable tax rates over time they tend to be risk averse for their financial assetliability management. Table 4.1 : Risks Encountered in Sovereign Debt Management Risk

Description

Market Risk

Risks associated with changes in market prices, such as interest rates, exchange rates, commodity prices, etc. For both domestic and foreign currency debt, changes in interest rates affect debt servicing costs on new issues when fixed-rate debt is refinanced, and on floating-rate debt at the rate reset dates. Hence, short- duration debt (short-term or floating-rate) is usually considered to be more risky than long-term, fixed rate debt. (Excessive concentration in very long-term, fixed rate debt also can be risky as future financing requirements are uncertain.) Debt denominated in or indexed to foreign currencies also adds volatility to debt servicing costs as measured in domestic currency owing to exchange rate movements. Bonds with embedded put options can exacerbate market risks.

Rollover Risk

The risk that debt will have to be rolled over at an unusually high cost or, in extreme cases, that it cannot be rolled over at all. To the extent that rollover risk is limited to the risk that debt might have to be rolled over at higher interest rates, including changes in credit spreads, it may be considered a type of market risk. However, because the inability to roll over debt and/or exceptionally large increases in government funding costs can lead to, or exacerbate, a debt crisis and thereby cause real economic losses in addition to the purely financial effects of higher interest rates, it is often treated separately. Managing this risk is particularly important for emerging market countries.

Liquidity Risk

There are two types of liquidity risk. One refers to the cost or penalty investors face in trying to exit a position when the number of transactors has markedly decreased or because of the lack of depth of a particular market. This risk is particularly relevant in cases where debt management includes the management of liquid assets or the use of derivatives contracts. The other form of liquidity risk, for a borrower, refers to a situation where the volume of liquid assets can diminish quickly in the face of

37

unanticipated cash flow obligations and/or a possible difficulty in raising cash through borrowing in a short period of time. Credit Risk

The risk of non-performance by borrowers on loans or other financial assets or by a counterparty on financial contracts. This risk is particularly relevant in cases where debt management includes the management of liquid assets. It may also be relevant in the acceptance of bids in auctions of securities issued by the government as well as in relation to contingent liabilities, and in derivative contracts entered into by the debt manager.

Settlement Risk

Refers to the potential loss that the government could suffer as a result of failure to settle, for whatever reason other than default, by the counterparty.

Operational Risk

This includes a range of different types of risks including transaction errors in the various stages of executing and recording transactions; inadequacies or failures in internal controls, or in systems and services; reputation risk; legal risk; security breaches; or natural disasters that affect business activity.

Source: Draft Guidelines for Public Debt Management produced by the World Bank and the IMF.

4.2.4 For evolving a risk management framework for the government’s debt portfolio, establishing the risk tolerance limits of the government can be a difficult process. The government’s implicit risk preferences may need to be interpreted by debt managers, by exploring the preferences implied by earlier government policy decisions and also by assessing the risk management culture in the Ministry of Finance, other leading Ministries and the Central Bank. Important indicators of risk preference could include whether the government is scaling down the size of its balance sheet, its approach in managing its public sector enterprises and contingent liabilities, and its attitude towards risk-sharing proposals emanating from the private sector. A prudential risk management is essential for reducing uncertainty among sovereign debt managers as to the government’s tolerance for risk, its willingness to trade off cost and risk objectives and, consequently, which transaction to accept and reject. Without an integrated debt management strategy in terms of strategic guidelines, portfolio management decisions can end up lacking coherence and being based on particular individual’s speculation about market trends or key relative prices. 4.2.5 In designing a debt management strategy, the sovereign debt manager is faced with several choices regarding the financial characteristic of the debt. The key decisions in this regard include: • the desired currency composition of the debt portfolio, including the mix between domestic currency debt and foreign currency debt; • the desired maturity structure and liquidity of the debt; • the appropriate duration or interest rate sensitivity of the debt; • whether domestic currency debt should be in nominal terms or indexed to inflation or a particular reference price; and • whether the portfolio composition should be transformed through swaps and other hedges or through new issuance.

38

Common examples of risk-cost tradeoff associated with such decisions involve: • Although ex ante, foreign currency debt may appear to be cheaper vis-à-vis domestic debt mainly due to inflation risk, on an ex post basis foreign currency debt may prove to expensive when domestic economic policy settings and market conditions have forced the government to devalue. • Excessive reliance on short-term paper to take advantage of lower short-term interest rates, in a positively-sloped yield curve environment, may leave a government vulnerable to volatile and possibly increasing debt service costs if interest rates increase, besides increasing refinancing risk and risk of default in the case that a government cannot rollover its debt at any cost. • Although hedging costs for foreign currency debt might be expensive, particularly where past exchange rate volatility have been very high or country risk is substantial, excessive unhedged foreign currency debt may leave governments vulnerable to volatile and possibly increasing debt service if exchange rate depreciate. • Spreading out borrowing in the capital market enables borrowers to sample market conditions and possibly develop greater recognition in the market where as concentrating borrowing in a smaller number of issues enables the borrower to meet their borrowing needs more quickly and creates larger, more liquid, benchmark issues. • In order to build up greater market knowledge, portfolio managers may be allowed to manage tactical positions and thereby entail the risk of not generating acceptable risk-adjusted returns, vis-à-vis the portfolio management policy of aiming to move the portfolio closer to the strategic benchmark. • In order to reduce its debt service cost, debt managers may build benchmarksized debt issues at key points along the yield curve. Reopening previously issued securities can also enhance market liquidity, thereby reducing the liquidity risk premia in the yields on government securities. However, concentrating the debt in benchmark issues may increase refinancing risk. 4.2.6 A key issue for developing a risk management framework for the debt manager is to decide whether the debt structure should be examined in isolation or whether risk management is better served by considering the nature of government’s assets and cash flows. Linking the risk management framework to a asset-liability framework (ALM) would involve identifying and managing market risks by examining the risk characteristics of the cash flows available to the government to service its borrowings, and choosing a portfolio of liabilities which matches these characteristics as much as possible. Although it may be difficult to produce a full balance sheet of the government, unlike a corporate entity, conceptually all governments have such a balance sheet. The objective of the ALM approach is not to produce a balance sheet, which quantify all its 39

assets. Instead, the main objective is to consider the various types of assets and obligations the government manages, besides its tax revenue and direct debt portfolio, and explore whether the risk characteristics associated with those assets can provide insights for reducing the cost and risk of government’s liabilities. This, in turn, could provide a benchmark for quantifying the costs and measuring the risks of the debt portfolio. This quantification of cots and risks provide a basis for developing a strategy for managing the debt portfolio. The ALM approach also provides a useful framework for considering governance arrangements for managing the government’s balance sheet. 4.2.7 Two variants of the ALM approach are possible. The first approach is to use a simplified balance sheet framework where the only assets are government revenues, measured as the present value (PV) of future revenues, and the only liabilities are government debt and the present value of future expenditures, excluding debt service. In this framework, risk is measured by the volatility of differences between the debt servicing costs and the future primary surpluses (i.e., the excess of asset over liability) as a result of market price movements. By comparing the costs and risks of different debt management strategies, the debt manager can then choose the strategy that reflects the government’s tolerance for risk. 4.2.8 The other approach of ALM framework is a balance sheet with multiple assets and liabilities. Assets are foreign exchange reserves and the PV of future tax revenues and in some cases, equity investment in state owned enterprises; while liability constitute foreign exchange debt, domestic currency debt and in some cases, explicit contingent liabilities like guarantees and deposit insurance or even implicit contingent liabilities. In this case, it is often simplest to manage the overall balance sheet on a sub portfolio basis, i.e., by “match funding” the various asset portfolios, which have different financial characteristics with sub-portfolios of debt which match those characteristics. Thus market risks on the foreign currency debt would be minimised by matching the currency and interest mix of the foreign exchange reserves with the stable portion or target level of foreign exchange debt, thereby creating a natural hedge. Such an approach should ensure that the central bank would be free to design and manage its intervention and investment portfolios of reserves according to its own objectives. To preserve the liquidity objectives of the foreign exchange reserves, the debt manager should issue debt with longer maturities than the liquid reserves, and match the interest rate characteristics by issuing floating rate debt or by using interest rate swaps to match the duration of the reserves. For reserves held for investment purposes, risk would be minimised by matching the currencies and interest rate sensitivities of the reserves and the foreign currency debt. The choice of currencies for the debt portfolio matched with investment portfolio should take into account the cost of funding since, the spread of investment returns over the cost of debt might differ across countries. Debt and assets are fully matched in foreign currency financial 40

characteristics when countries solely borrow in foreign currency only to the extent they hold foreign exchange reserves. 4.2.9 For the remainder of foreign exchange debt, for countries, which borrow in foreign currency in excess of the foreign exchange reserves they hold, the risk is minimised by matching it with the domestic currency assets, which has the highest correlation with the domestic currency. Another option is to create shadow foreign exchange assets for the remainder of foreign exchange debt. For a pegged exchange rate regime, this could be done by issuing debt in foreign currency in accordance with the trade-weighted basket of currencies for the peg, which is most closely integrated in terms of trade and capital flows. For a floating rate regime, the shadow asset can be created by undertaking a statistical analysis of historical variances and covariances of foreign currencies relative to the domestic currency to find the portfolio of debt which has the minimum variance to the domestic currency. In some countries, where data limitations or comprehensive policy reforms may make past statistical relationships irrelevant, the shadow asset could be constructed on the basis of the country’s trade pattern and sources of capital flows. Similarly, financial characteristics of domestic currency debt are matched with the structure of domestic currency assets. 4.2.10For sovereign debt management, risk is measured as the volatility of debt service and other expenditures relative to the volatility of tax and other revenues. In present value terms, it can also be measured as the volatility of the value of the government’s assets relative to the liabilities. The risk measurement process quantifies the risks and costs of alternative debt strategies. This is usually done with financial models, ranging from relatively simple, scenario based models, to more complex models using sophisticated statistical analysis and techniques, like simulation-based models. While use of sophisticated and complex models can improve the degree of precision to the cost and risk estimates, it should not change the cost/risk rankings of a well-specified simple model significantly. Most models in use by sovereign debt managers are based on simulations of future debt servicing flows for the debt portfolio under a variety of different assumptions regarding the strategy for managing the composition of debt and path of future market variables. 4.2.11 In practice, most debt managers simplify the scenario-based model by comparing the range of debt servicing costs against a notional benchmark. Thus, the first step is to project debt servicing costs under a specific strategy for managing the structure of the debt portfolio, and using a “base case” set of assumptions for future market variables. This gives the expected cost of the debt strategy being evaluated over the time horizon specified. The risk of the strategy is measured as the potential increase in debt servicing costs over the same time horizon under the worst case scenarios of market prices. The next step is to project debt servicing costs under different debt strategies by using different proportions of fixed-floating rate debt or currency mix, and also using different 41

assumptions for market risk i.e. a wide range of alternative market price assumptions. This would give estimates of costs and risks of different strategies under the same base case and risky scenarios. 4.2.12The time horizon used by most debt managers is a medium-term time horizon, for example five years, since budget projections become increasingly uncertain the further out they are projected. In this case, risk is measured as the potential increase in cumulative costs relative to the base case over this time horizon, and corresponds to a notional volatility in debt servicing costs relative to the revenue flows of the underlying assets. The choice of the numeraire currency is also crucial for analytical purpose. Costs are usually measured in the domestic currency where the portfolio of debt being modeled funds domestic currency assets. A sub portfolio of debt funding foreign currency assets can be modeled separately for which a foreign currency can be used as the numeraire. Similarly, the base case assumption is also critical for the output of the model. A poorly selected base case can bias the choice towards strategies which might appear to be optimal under such assumptions, but could actually be riskier or higher cost strategy than expected. One way of avoiding this type of bias is to introduce “market neutral” assumptions for the base case. 4.2.13Simulation-based models, used by most debt managers, involves Monte Carlo simulations to generate a wide variety of different market risk scenarios and then to project different debt service costs for a particular debt strategy under different market scenarios. This generates a distribution of different cost outcomes for that debt strategy. The expected cost of each strategy can then be represented by the mean of the distribution, and the risk of the strategy can be represented by a measure of the upper range of costs above the mean, such as the variance or standard deviation of the distribution. Most debt offices use “value-at-risk” (VAR) approach for measuring risk, a variant of this approach. VAR measures the confidence interval of the range of possible risks measured by the volatility of the marked-to-market, or present values of the liability portfolio. The Denmark Nationalbank, on the other hand, adopted a “cost-at risk” (CAR) approach to calculate a confidence interval of the upper range of costs using the standard deviation from the distribution. The risk of a particular strategy is then measured by the confidence interval. The specification of the currency and time frame to use for cost and risk measures should be made on the basis of the same type of considerations used for scenario-based models. 4.2.14Once the costs and risks of alternative debt have been measured, an efficient frontier, which is an envelope of all those strategies that are “efficient” in the sense that there are no other strategies that have a lower risk for a given expected cost, could be constructed. The government should choose the strategy that best represents the government’s preferences for managing the risk/cost tradeoffs. In general, debt managers select a debt strategy that lies along the efficient frontier, and tend to be risk averse by selecting a strategy on the 42

frontier that represent higher cost but lower risk. Sometimes, other objectives, such as desire to issue debt at uniform points on the yield curve in order to promote the development of a pricing benchmark for private sector issues could lead to the debt manager choosing a strategy that does not lie on the frontier. This could entail a cost relative to the most efficient strategies. 4.2.15The process of deciding a debt strategy by debt offices, has been facilitated, by using a “strategic benchmark” portfolio, which represent the approved strategy. Such a benchmark is typically comprised of targets for the key risk characteristics of the approved strategy, such as the mix of fixedfloating debt, currency composition, or duration. The actual debt portfolio could be then moved closer to designed benchmark by debt managers while deciding on key terms for new debt issues, by buyback operations, and also by using currency and interest swaps and other hedging activities. The benchmark also serves as a control mechanism to ensure that the debt manager implements the approved strategy. For debt managers that actively manage debt to reduce costs, the benchmark can serve as the basis for measuring the performance of the debt manager by comparing the actual cost of the debt portfolio to what the cost would if the exact benchmark was attained. For governments permitting debt managers to undertake tactical trading within well-defined position and loss limits, these operations can also be assessed as to whether they have contributed net cost savings relative to the benchmark. Benchmarks should also reviewed from periodically to access their appropriateness and revise it if the government’s objectives change or there are significant changes in economic relationships. 4.2.16Benchmarks are usually set by debt offices for portfolio management for the foreign currency debt, and in some case, for the entire debt portfolio. Table 4.7 shows the different benchmarks in use by selected debt offices. For foreign currency debt, benchmarks should preferably be derived for the government’s net foreign currency debt portfolio rather than the gross debt portfolio. This implies that, whenever possible, part of the foreign currency debt should be matched against the financial characteristics of the foreign exchange asset. Debt managers would then be free to concentrate on their risk management activities on the remaining net foreign currency exposure. The usual benchmarks for foreign currency debt are modified duration and/or interest rate mix for managing the acceptable interest rate risk; currency composition along with the proportion of domestic and foreign currency debt for managing exchange rate risk; and the maturity profile, usually indicated by a ceiling on the amount of debt maturing at any given point, for managing the refinancing risk. 4.2.17Benchmarks for domestic debt are used only when the debt manager has considerable scope and the domestic capital market is sufficiently developed to adjust the composition of its debt portfolio through new borrowings, buyback trades and/or through derivative transactions. The domestic currency benchmark 43

usually focuses on the acceptable ranges for interest rate risk and the maturity profile for the domestic debt portfolio. However, given their dominant role as an issuer in the domestic bond market, most governments prefer not to purchase or trade their domestic securities, other than what is needed for normal liquidity management. 4.3

Institutional structure for public debt management

4.3.1 The institutional setting for public debt management in different countries, could be broadly classified into five categories i.e., (i) a full-fledged debt office either in the mainstream Ministry of Finance; (ii) under the Ministry of Finance as a separate entity like the Treasury or a Debt Management Agency; (iii) autonomous institution outside the Ministry; (iv) in the Central bank; and (v) the debt office is dispersed between the Ministry of Finance and the central bank. 4.3.2 The first four categories refer to a full-fledged debt office and could be differentiated by the location and the degree of autonomy accorded to the debt office. Table 4.3, which lists the institutional setting for full-fledged debt offices in place shows that the bulk of the debt offices are located as a separate agency under the Ministry of Finance with sufficient degree of operational independence (i.e. second category). In terms of the degree of independence accorded to the debt office, the institutional setting for the above category resembles the third category, i.e., where the debt office is located as an autonomous entity. Although, in the first three cases, the debt office may be located under the Ministry of Finance or as an autonomous agency, the central bank still retains some agency services for debt management and in some cases undertake foreign exchange operations for its foreign currency debt and cash management. Denmark is the only country, where its central bank, Nationalbanken houses the debt office for the government. 4.3.3 International experience suggests that centralised debt offices in most countries are located under the Ministry of Finance. Within this institutional structure, in most of the advanced countries, the debt offices are set up as an autonomous or separate entity within a Treasury or as a statutory unit. This enables the debt office to assume sufficient degree of operational independence Thus, for thirty countries for which data is available, twenty-four have their debt offices under the Treasury/ Ministry of Finance. This includes all the emerging economies for which information is available. The main argument for entrusting sovereign debt management responsibility within the Ministry of Finance or Treasury is the proximity of location, which enables the senior management within the Ministry of Finance to review and assess the performance of the entity more easily. This issue of geographical proximity is particularly important when a full-fledged centralised debt office is being set up and management competencies are unproven. Another factor, which have prompted many 44

governments to locate the debt office within the Ministry of Finance is that since public debt has budgetary implications, co-ordination between budget making and the debt office is facilitated. This arrangement thereby minimises chances of any conflict arising from the budgetary process wherein the annual borrowing requirements are determined and the management of such liabilities. The downside risk of unsustainable borrowing has been obviated in most of the cases, by legal enactment of authorising annual borrowing with a preset limit (Table 4.4) and practising policies of fiscal prudence. 4.3.4 Few debt offices (e.g., Australia, Austria, Ireland, Portugal and Sweden) have also been set up as an autonomous debt agency or corporation outside the Ministry of Finance giving it a distinct institutional presence. This was established with legislation (e.g. Ireland and Portugal) or without legislation (e.g. Australia and Sweden). Very recently, Germany have decided to take out the debt management functions from the Ministry of Finance to an autonomous set up as a private corporation. The overriding reason for creating an autonomous institution emanated from the concern of conflicts in objective between fiscal policy and debt management. Thus, an autonomous debt office would be less likely to engage in risky strategies designed to maximise shortterm political gains. Another common concern has been the influence on interest rate. Thus, any perception of insider trading or market manipulation would undermine the credibility of the government, the debt office and the market. The risk of debt management and cash management being downplayed by a large institution like the Ministry of Finance could result in the commercial needs of the business being inadequately funded. This issue was also factored in while taking the debt office outside the Ministry of Finance and setting it up as an autonomous agency. In order to build a sound risk management culture, an autonomous structure, under the supervision of a board of directors responsible for managing the government’s interests is likely to adopt a more commercial approach in reviewing the need for additional expenditure on systems, training, salary compensation and recruitment of skilled staff. In particular, an autonomous agency could maintain a flexible management and career path structure, and link the pay scale of its staff to that of private sector practitioners. Such flexible pay structure would allow the debt office to attract highly qualified staff, that are, knowledgeable in the increasingly complex financial instruments and markets. 4.3.5 The degree of autonomy in such debt offices, vary from country to country. Although the debt offices are autonomous in nature, most of them either report to the Minister of Finance or to the Parliament. Thus, while the formulation of debt policy like level of the debt, limits on domestic and foreigncurrency borrowing is a political decision and therefore should rest with the government, the actual management of sovereign debt can be extracted from the political domain by assigning such responsibility to an autonomous institution. Under this arrangement, the Ministry of Finance, based on its objectives, risk 45

preferences and macroeconomic and institutional constraints of the country, defines the medium-term strategy for debt management; while the debt office implements that strategy and administers the issuance of domestic and foreign currency debt. 4.3.6 The issue of conflict in objectives between debt management and monetary management has led to many central banks, transfer its debt management responsibility back to the government or to an autonomous agency outside the Ministry of Finance. Instances of such conflicts are as under: o A central bank may be reluctant to raise interest rates to control inflationary pressures to avoid any adverse effect on its domestic liability portfolio. o A central bank may be tempted to manipulate financial markets to reduce interest rates at which government debt is issued or to inflate away some of the value of the nominal debt. o A central bank may also be tempted to inject liquidity in the market prior to debt refinancing, or to bias the maturity structure of the debt profile according to the stance of its monetary policy. 4.3.7 Management of public debt by the central bank is further constrained due to conflict in objectives between debt management and exchange rate management. Such instances include: o Daily management of the liquidity of the foreign currency debt by the central bank in the foreign exchange market i.e. converting foreign debt proceeds into local currency or converting local currency funds into foreign currency debt repayments, may conflict with the intervention policy of the central bank. o Similarly, the central banks sale and purchase of securities to meet foreign currency debt requirements could also be perceived by financial markets as having a signalling effect on its exchange rate policy, thereby undermining its effectiveness. 4.3.8 In view of the potential conflicts in objectives between debt management, monetary policy and exchange rate management, Austria, Hungary and United Kingdom have recently shifted their debt management responsibility out of the central bank. In New Zealand, all debt management functions carried out by the central bank, as agent of the debt office, have been conducted without reference to monetary policy considerations since 1988. Moreover, in South Africa, after a thorough review of debt policy, the central bank, which until recently, has been the government’s agent for marketing its debt instruments, was made accountable to the Department of Finance on all matters related to debt management. Funding activities undertaken by the central bank on behalf of the government were ring-fenced from monetary policy operations.

46

4.3.9 Denmark is the only exception, where in 1991 the Danish government decided to regroup assets and liabilities management under the central bank’s authority. The rationale behind the decision was to improve the coordination of the management of the public debt and the foreign exchange reserves, and to reduce the net exposure of the government to exchange rate risk. Some governments recognise the fact that the central bank has more staff with market transaction experience. The efficiency advantage has led to some central banks (United Kingdom and Brazil) undertake the government’s foreign currency borrowing. For countries, which have already established their debt offices, the central bank, still retains some debt management functions like book-keeping, registry services etc under an agency agreement between the Ministry of Finance and the central bank. For some countries, cash management and foreign exchange transactions related to foreign currency debt are also undertaken by the central bank. 4.3.10Internationally, independent set-up and the Ministry of Finance are regarded as somewhat watertight compartments for locating the debt office. In reality, however, there is a very thin line between an independent set-up and the Ministry of Finance. The Ministry of Finance always exercises some measure of control over the operations of the debt office, irrespective of its location. This is unavoidable because it is the liability of the Government that is to be managed by the debt office. Therefore, even among the most independent set-ups like National Treasury Management Agency of Ireland and the Swedish National Debt Office which are entrusted with day-to-day management responsibilities, the Ministry of Finance determines the policy, sets the operational guidelines and the benchmarks under which the debt office is required to operate. This institutional mechanism is usually, safeguarded, by public debt legislation or legal statutes. 4.3.11On the other hand, in many countries, the Ministry of Finance recognised the need to build professionalism and capacity for debt management, by locating the debt office outside the mainstream Ministry as a separate entity. The main objective is to accord sufficient degree of autonomy to the debt office, even though it is located within the Ministry. The degree of freedom varies across countries with respect to its own budget provision, recruitment policy, salary structure and legal framework. 4.3.12For the fifth category of debt offices, public debt management responsibilities are typically split, with the Ministry of finance in charge of the public external debt management and the central bank is responsible for domestic debt management. Table 4.5 shows the split nature of debt management responsibilities for some emerging market economies till recently. Among the emerging market economies, some countries like Colombia, Hungary, Mexico and South Africa have already established a full-fledged debt office under the Ministry of Finance. Further more, some countries like 47

Argentina, Brazil, China and Thailand have started initiatives to set up a debt office under the Ministry of Finance. To establish a public debt office, the starting point for all these countries has been to strengthen the capacity building of the debt office by establishing a middle office.

48

Table 4.3. Institutional Location of Sovereign Debt Management Responsibility Under the Ministry of Finance or Treasury* Advanced Economies Australia Austria Belgium Canada Denmark Finland France Germany Greece Ireland Italy Japan Netherlands New Zealand Portugal Spain Sweden Switzerland United Kingdom United States Emerging Economies Argentina1 Brazil 1 China1 Colombia Hungary Korea 1 Mexico South Africa Thailand1 Turkey

Located within the Central Bank

Located elsewhere as an autonomous entity  

                 

         

1: Establishment of sovereign debt management offices is currently underway in these countries. * : In many countries, although debt offices are under the Ministry of Finance or Treasury, the debt office is set up as an autonomous entity with sufficient operational independence. Source: Draft document on “Sound Practices in Sovereign Debt Management”, FPS Department, The World Bank, March 2000; OECD as mentioned in “Risk Management of Sovereign Assets and Liabilities”, Working Paper, WP/97/166, IMF, December 1997 and national authorities.

49

Table 4.5 : Institutional Framework for Foreign Currency Debt Management in Emerging Economies Countries

China India Indonesia Korea Singapore Thailand Argentina Chile Colombia Mexico Peru Venezuela Czech Republic

Hungary Poland Russia Israel

South Africa

Central Govt.

State & Local Govt.

MOF MOF MOF MOF None DMO under MOF MOF MOF MOF MOF DMO under MOF MOF None DMO under MOF MOF MOF MOF DMO under MOF

Not allowed Not allowed Not allowed Own responsibility

State Owned Enterprises SOEs

None Own responsibility Not allowed State Govts./MOF State Owned banks DMO under MOF Not allowed MOF

SOEs SOEs MOF MOF SOEs/MOF MOF MOF Not allowed

SOEs None

Own responsibility Regional agencies Own responsibility Not allowed

SOEs MOF

Source: “Managing foreign debt and liquidity risks in emerging economies : an overview”, John Hawkins and Philip Turner, as excerpted in “Managing Foreign Debt and Liquidity Risks”, BIS Policy Papers, No. 8, September 2000.

Table 3.3 : Degree of Autonomy for Debt Offices Countries

Degree of Autonomy

Ministry of Finance Belgium Canada Germany Greece Japan Mexico Netherlands New Zealand Switzerland Turkey United Kingdom

Not independent Independent except for independent matters Dependent Independent within the broad objectives of the Development Plan Independent No specific independence Independent, except for restriction on some type of institution Independent under normal circumstances Independent within limits set by the remit

Autonomous Agency Australia Austria Ireland Sweden

Highly independent Highly independent Independent in some broad guidelines drawn by the MOF Independent except for foreign exchange

Central Bank Denmark

Borrowing program is approved by the MOF

Source: OECD as mentioned in “Risk Management of Sovereign Assets and Liabilities”, Working Paper, WP/97/166, IMF, December 1997.

50

Table 3.4 : Institutional Arrangement of Debt Offices and Annual Borrowing Authority or Debt Ceiling Limit Institutional Arrangement Ministry of Finance

Countries

Limit on Annual Borrowing Authority

Debt Ceiling Limit

Belgium



Canada



Finland



France



Germany



Greece



Hungary



Italy



Mexico



Morocco



New Zealand



× × × × × × × × × × ×

United States United Kingdom

× 



Australia



Ireland



Portugal



Sweden



× × × ×

Denmark

×



×

Autonomous Agency

Central Bank

Source: Draft Guidelines for Public Debt Management, SM/00/135, IMF.

Table 3.5 : Legal Framework for Debt Offices Countries

Limit for Domestic Borrowing

Decides new limits

Limit on the cost of borrowing Yes, Borrowing Authority Act Yes, a limit is set by federal legislative authorisation (Budget Law) No, except for the limit to T-Bills Yes, a limit is set by Budget Law Yes, a limit is set according to the Federal Budget Implicit limit (budgeted borrowing requirement) No legal limit No legal limit Only for govt. bonds the limit is twice the budget deficit Limit by the funding remit

The Parliament The Parliament The Parliament

MOF may alter the program For govt. bonds, the Parliament -

Yes, financial year budgetary need Yes, the limit is set by the Financial Law No Limit only for foreign exchange funding

DMO and the Treasurer The Parliament -

Limit on the level of debt outstanding

The Parliament

Ministry of Finance Belgium Canada Germany Greece Japan Mexico Netherlands New Zealand Switzerland Turkey United Kingdom

The Parliament The Congress -

Autonomous Agency Australia Austria Ireland Sweden

Central Bank Denmark

51

Source: OECD as mentioned in “Risk Management of Sovereign Assets and Liabilities”, Working Paper, WP/97/166, IMF, December 1997.

4.4

Scope of role and functions of a debt office

4.4.1 In a number of countries, the scope of debt management operations of the central government broadened in recent years. The scope of a debt office involves all contractual liabilities of the central government, including contingent liabilities, in some cases. Few debt offices also attend to manage the pension liabilities of the government. Thus, while the public sector debt, which is included or excluded from the central government’s mandate over debt management varies from country to country depending on the nature of the political and institutional framework, some debt offices include publiclyguaranteed debt in the sovereign debt management policy. Debt management by a public debt office, often involves, the oversight of liquid financial assets and potential exposures due to off-balance sheet claims on the government, including contingent liabilities such as state guarantees. On the other hand, some debt office exclude publicly-guaranteed debt from their debt management policies until the guarantee is invoked. It is, however, desirable that all public debt is centralized under the management structure of the debt office so that the risk exposure of the sovereign debt is adequately reflected and managed as a single portfolio. 4.4.2 For a public debt agency, the main tasks or responsibilities involved generally include:  issuance or contracting debt obligations  redeem or service debt obligations  manage the day-to-day risk exposure like liquidity risk, market risk and credit risk of the sovereign debt portfolio,  secondary market operations  cash management  issue and manage guarantees  evaluate performance  report periodically to the authority through publication or data dissemination  book-entry and record keeping  clearing and settlement system 4.4.3 Where debt management office exist within the Ministry of Finance or Treasury, the roles of the debt office has sometimes been merged into a broader organisational structure, expanding the role of debt offices to manage both asset and liability of the government. Debt offices in New Zealand and South Africa have been established as part of the asset-liability management office. In addition, the debt offices may also advice on a range of public sector financial management issues including privatisation and restructuring of state-owned 52

enterprises; and management of a wide range of commercial, contractual and litigation risks on behalf of the government. The main benefits from such a merged structure enables the debt office to better understand the risks of the government’s balance sheet and identify natural hedges. 4.5

Organisational Structure

4.5.1 Sound governance considerations suggest that debt management functions should be organised as separate units given their different objectives, responsibilities and staffing needs. Accordingly, the business of sovereign debt management has changed rapidly since the late 1980’s, with many governments investing heavily to create a specialised debt management office by adopting an organisational structure similar to that found in leading corporate and banking treasuries and in the reserve management departments of central banks. The usual practice has been to divide functional responsibilities and accountabilities by establishing a separate front, middle and back office, and maintain separate reporting lines to the Chief Executive of the debt agency. Although all the debt offices might not have achieved such a complete structure, most debt offices have established the main elements of such functional entities. 4.5.2 The front office is typically responsible for executing transactions in financial markets, including the management of auctions and other forms of borrowing, and all other funding operations. For most of the debt offices, the front office is normally responsible for the analysis and efficient execution of all portfolio transactions, consistent with the portfolio management policy of the agency. These transactions usually include short and medium term borrowing in domestic and foreign currencies, management of trading positions and hedging transactions, the investment of foreign currency liquidity and any excess cash balances associated with the government’s daily departmental cash management. Within the front office, individual portfolio managers are assigned different functional responsibilities usually on an instrument, market, or currency basis. Given the market-related nature of operations of portfolio managers, they can also provide a wide range of portfolio management services like designing funding and pricing strategies, assessing fair value on individual transactions and exploring market opportunities to help move the actual debt portfolio closer to the strategic benchmarks. The latter could be achieved through terms for new borrowing, hedging strategies and buyback operations. They can also provide advice on government initiatives to foster the development of the primary and secondary government bond market and possible market reaction to new fiscal information. 4.5.3 A separate middle office has been set up both within a full fledged debt office as well as where debt management responsibilities are split between the Ministry of Finance and the central bank. The key role of the middle office is to develop a debt management strategy by establishing a cost and risk 53

management framework for the central government’s debt portfolio. For this purpose, the middle office could report to the head of the management operation on the objectives for sovereign debt management and on the cost and risk tradeoffs of different portfolio management strategies. Based on the government’s preferences in respect of expected cost and risk, middle office staff would develop a set of portfolio management policies and seek approval of the Minister of Finance for these. This is usually done through evolving strategic benchmarks for the portfolio for managing different risks. The main policies developed by the middle office to address different portfolio risks include: • For managing market risk, the middle office identifies the preferred currency composition, duration of the portfolio, along with the decision rules for transacting to transform the actual portfolio closer to the strategic portfolio over time. Questions as to whether tactical trading should be permitted would need to be examined and, if undertaken, position limits and loss limits need to be established. • For management of refinancing risk, the middle office is required to develop policies specifying the acceptable maturity profile of the portfolio and the degree of refinancing risk in any single year. • Similarly, management of credit risk would require establishing limits on the government’s exposure to individual counterparties through swap transactions or the investment of excess liquidity. Acceptable limits for credit exposure are often based on the credit rating assigned by sovereign credit rating institutions, the marked-to-market exposure of the position, the type of instrument involved and the time to maturity. Sub-limits for specific transactions may also be established. • Liquidity risk management policies specify minimum levels of foreign currency liquidity, the instruments and currencies that this liquidity can be held in, and tactical trading benchmark for the investment of this liquidity. Liquidity levels may also need to be established for domestic currency debt portfolio if the government is not able to borrow domestically at short notice to meet day-to-day liquidity needs. • Settlement risk management policies involve designating acceptable transaction bankers, custodians, clearing brokers and fiscal agents and the setting maximum amount of exposure to any settlement institution. Provision of overdraft facility up to specified limit may also be necessary for situations where timing mismatches means that the funds may be few hours late in arriving while the loan agreements specify that the foreign currency debt servicing payments must be made early in the business day. Since settlements payments could be very large, debt offices should have a detailed settlements procedures manual. • Operational risk management policies should include policies, through crossreference to an operations manual, to manage operating risk around transactions, management information systems, the legal framework and 54

other threats to the organisation’s business continuity and reputation. While the most common operational risks tend to lie on the transaction side through errors in confirming and settling trades, even with comprehensive management controls, the most serious operating risks generally relate to fraudulent breaches of controls and systems failure. This is why a sound risk management culture, well-defined controls and separate reporting lines are so important. 4.5.4 The middle office is also responsible for monitoring compliance against strategic benchmarks. This separation helps to promote the independence of those setting and monitoring the risk management framework and assessing performance (the middle office) from those responsible for executing transactions (the front office). Debt offices in some countries like Belgium, Ireland and Sweden also have a legal office inside the middle office. Information technology functions, which tend to be located in the back office, are sometimes a middle office responsibility. Some middle offices (e.g., in Belgium, Brazil, Colombia and New Zealand) also carry out investor relation functions, which are usually the front office responsibilities. 4.5.5 Although separation in responsibilities between the front and middle office helps to promote the independence of those setting and monitoring the risk management framework and assessing performance (the middle office) from those responsible for executing transactions (the front office); tensions may arise between them while discharging their duties. Tension may arise in deciding the appropriate types of transactions or new instruments to introduce, the setting of benchmarks, and measuring and reporting value added. It is therefore, essential to ensure that the middle office functions are not constraining the initiative of the front office staff in their search for various ways to add value to the debt management operation, lower debt servicing costs, increase returns from their liquidity management, and reduce risk. Similarly, the integrity of the middle office staff and their capability in their risk management design and monitoring and control functions can be seriously compromised if this group is not given sufficient freedom to develop their professional capacity. Such tensions could also be resolved by ensuring that there is an agreement on the essential roles of the two offices, and their capacity to develop strong synergies within a clear set of accountabilities. 4.5.6 Back Office 4.6

Governance: Legal Framework and Accountability

Sound governance practices are an integral part of sovereign debt management. In order to establish appropriate accountability for managing the sovereign debt portfolio, most governments have in place legislation pertaining to powers to borrow, invest, issue guarantees and undertake transactions on 55

behalf of the government. This overcomes the need to request specific authorisations from the Parliament or the constitutional authority for individual transactions, which can introduce and a range of political factors into the decision making and considerable delay the execution of transactions. For e.g., a study by IBRD suggests that more than three-fourth of the member countries surveyed, on an average, took more than a week to obtain approval to execute a foreign borrowing transaction. This implies that such countries, in their endeavour to manage the debt portfolio, can only operate with a lag to achieve a transaction in case there are any desirable market movements in the international financial market. Such impediments, in a volatile international financial market, could ultimately prove to be imprudent decision finally when the transaction is entered into or entail a loss in terms of opportunity cost for not being able to execute a transaction at the opportune moment. The authority to issue new debt is normally stipulated in the form of either borrowing authority legislation with a pre-set limit or a debt ceiling. Legal arrangements should also be supported by delegation of appropriate authority to debt managers. A common feature of this type of legislation is that the authority for borrowing or financial transaction decision rests with the Minister of Finance or Treasury and requires the Ministry to be accountable for these decisions to the Parliament. The Minister, in turn, delegates the decisionmaking authority to the head of the debt office. All delegations pass through the head of the debt agency to individual portfolio managers and any other staff with operational responsibility. For example, in Portugal, a statue specifies responsibilities, administrative and supervision framework, delegation of powers, distribution of tasks and the overall financial structure. Objectives for government debt management should be clearly specified, publicly disclosed and included in legislation, wherever possible in order to reduce uncertainty as to the government’s willingness to trade-off cost and risk. Unclear objectives can often lead to poor decisions on how to manage the existing debt resulting in a potentially risky and expensive portfolio. Lack of clarity might also create uncertainty within the financial community leading to a higher risk premia. Thus many debt offices have made disclosure of their primary objective in terms of cost-risk tradeoff (Table 4.6). Disclosure may also be made in respect of secondary objectives like such as maintaining the liquidity of government issues at various points on the yield curve so as to provide a pricing benchmark for private issuers, or for emerging market economies, to promote the development of domestic debt market through a gradual extension of the maturities of government debt and development of new instruments. The above arrangement should be supplemented with establishment of a risk management framework. The strategic benchmarks for portfolio management of the sovereign debt, in terms of the currency, interest and maturity mix, produced by the debt office needs to be approved by the Minister 56

of Finance on an annual basis. For countries where the debt agency is outside the Ministry of Finance or Treasury, the recommendations made to the Minister of Finance should be made by the Ministry of Finance in conjunction with the debt office. This requires a counterpart unit in the Ministry of Finance, which is usually compact, to supplement the recommendations of the debt office while approving the strategic benchmarks. Once the benchmarks are approved, the debt office would be independent to operate for achieving such strategic benchmarks. For few debt offices pursuing active portfolio management, tactical trading limits are imposed on the portfolio manager. The Table below provides arrangement for countries enacting legislation for clearly specifying debt management objectives and strategic benchmarks. The legislation should also ensure that there is appropriate auditing of the financial transactions undertaken by the debt office to ensure that they comply with generally accepted accounting practices and the portfolio management policies of the debt office. There should also be comprehensive reporting of financial performance to the Minister of Finance and/or Parliament. Irrespective of the institutional structure of the debt office, legal arrangements should clearly specify the organisational framework for debt management including the mandated and roles of the debt office. Authorising an outside body of advisors (constituting of a board of directors, advisory bodies etc.) are used frequently to provide quality assurance on debt management on a regular basis to the head of the debt office and the minister of finance or the head of the finance ministry. Thus, autonomous debt agencies in countries like Sweden and Portugal, are managed by boards, appointed by the government and chaired by the head of the debt office. Advisory boards with mainly nongovernmental members work with the autonomous debt agency in Ireland. In countries like Belgium, Colombia, Hungary and South Africa where the debt office is located within the Ministry of Finance, committees staffed mainly from the Finance Ministry; other government agencies, including the central bank; meet regularly with the government debt managers to discuss broader government debt and asset management issues. On the other hand, for the debt office in New Zealand, which is located in the Treasury, the Advisory Board comprises mainly of non-governmental members. Such members are experts in risk-management theory and practice, meets four times a year to provide advice and oversight across a broad range of strategic and operational risk management issues and to establish greater transparency in the decision making and supervision process. 4.6

Co-ordination with fiscal policy authority and monetary policy authority

4.7

Capacity Building in Sovereign Debt Management

57