Debt equity conversion analysis : a case study of the Philippine program

WDP-76 76 E World Bank Discussion Papers Debt Equity Conversion Analysis A Case Study of the Philippine Program John D. Shilling, Anthony Toft, Woonki Sung, and Wayne Edisis  RECENT WORLD BANK DISCUSSION PAPERS No. 25. The Poor and the Poorest: Some Interim Findings. Michael Lipton No. 26. Road Transport Taxation in Developing Countries: The Design of User Charges and Taxes for Tunisia. David Newbery, Gordon Hughes, William D.O. Paterson, and Esra Bennathan No. 27. Trade and Industrial Policies in the Developing Countries of East Asia. Amarendra Bhattacharya and Johannes F. Linn No. 28. Agricultural Trade Protectionism in Japan: A Survev. Delbert A. Fitchett No. 29. Multisector Framework for Analysis of Stabilization and Structural Adjustment Policies: The Case of Morocco. Abel M. Mateus and others No. 30. Improving the Qualitv of Textbooks in China. Barbara W. Searle and Michael Mertaugh with Anthony Read and Philip Cohen No. 31. Small Farmers in South Asia: Their Characteristics, Productivity, and Efficiency. Inderjit Singh No. 32. Tenancy in South Asia. Inderjit Singh No. 33. Land and Labor in South Asia. Inderjit Singh No. 35. Global Trends in Real Exchange Rates. Adrian Wood No. 36. Income Distribution and Economic Development in Malawi: Some Historical Perspectives. Frederic L. Pryor No. 37. Income Distribution and Economic Development in Madagascar: Some Historical Perspectives. Frederic L. Pryor No. 38. Quality Controls of Traded Commodities and Services in Developing Countries. Simon Rottenberg and Bruce Yandle No. 39. Livestock Production in North Africa and the Middle East: Problems and Perspectives. John C. Glenn LAlso available in French (39F)J No. 40. Nongovernmental Oraanizations and Local Development. Michael M. Cernea LAlso available in Spanish (40S)J No. 41. Patterns of Development: 1950 to 1983. Moises Syrquin and Hollis Chenery No. 42. Voluntary Debt-Reduction Operations: Bolivia, Mexico, and Beyond... Ruben Lamdany No. 43 Fertility in Sub-Saharan Africa: Analysis and Explanation. Susan Cochrane and S.M. Farid No. 44. Adjustment Programs and Social Welfare. Elaine Zuckerman No. 45. Primary School Teachers' Salaries in Sub-Saharan Africa. Manuel Zymelman and Joseph DeStefano No. 46. Education and Its Relation to Economic Growth, Poverty, and Income Distribution: Past Evidence and Further Analysis. Jandhyala B.G. TilaK No. 47. International Macroeconomic Adjustment, 1987-1992. Robert E. King and Helena Tang No. 48. Contract Plans and Public Enterprise Performance. John Nellis LAlso available in French (48F)j No. 49. Improving Nutrition in India: Policies and Programs and Their Impact. K. Subbarao No. 50 Lessons of Financial Liberalization in Asia: A Comparative Study. Yoon-Je Cho and Deena Khatkhate No. 51 Vocational Education and Training: A Review of World Bank Investment. John Middleton and Terry Demsky (Continued on the inside back cover.)  7 6 World Bank Discussion Papers Debt Equity Conversion Analysis A Case Study of the Philippine Program John D. Shilling, Anthony Toft, Woonki Sung, and Wayne Edisis The World Bank Washington, D.C.  Copyright C 1990 The World Bank 1818 H Street, N.W Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing March 1990 Discussion Papers are not formal publications of the World Bank. They present preliminary and unpolished results of country analysis or research that is circulated to encourage discussion and comment; citation and the use of such a paper should take account of its provisional character. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board ofExecutive Directors or the countries they represent. Any maps that accompany the text have been prepared solely for the convenience of readers; the designations and presentation of material in them do not imply the expression of any opinion whatsoever on the part of the World Bank, its affiliates, or its Board or member countries concerning the legal status of any country, territory, city, or area or of the authorities thereof or concerning the delimitation of its boundaries or its national affiliation. Because of the informality and to present the results of research with the least possible delay, the typescript has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to Director, Publications Department, at the address shown in the copyright notice above. The World Bank encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to photocopy portions for dassroom use is not required, though notification of such use having been made will be appreciated. The complete backlist of publications from the World Bank is shown in the annual Index of Publications, which contains an alphabetical title list and indexes of subjects, authors, and countries and regions; it is of value principally to libraries and institutional purchasers. The latest edition is available free of charge from Publications Sales Unit, Department F, The World Bank, 1818 H Street, N.W, Washington, D.C. 20433, U.S.A., or from Publications, The World Bank, 66, avenue d'1ena, 75116 Paris, France. John D. Shilling is manager of, and Anthony Toft and Woonki Sung staff members of, the Financial Advisory Unit of the World Bank's Cofinancing and Financial Advisory Services Department; Wayne Edisis is with the Foreign Investment Advisory Services of the Intemational Finance Corporation and Multilateral Investment Guarantee Agency, both affiliates of the Bank. ISSN 0259-210X Library of Congress Cataloging-in-Publication Data Debt equity conversion analysis : a case study of the Philippine program / John D. Shilling ... let al.]. p. cm. -- (World Bank discussion papers ; 76) ISBN 0-8213-1515-3 1. Debt equity conversion--Philippines. 2. Debts, External- -Philippines. 3. Loans, Foreign--Philippines. I. Shilling, John D., 1943- . II. Series. HJ8802.D42 1990 336.3'4'09599--dc2O 90-34372 CIP  ABSTRACT The paper analyzes the potential impact of implementing a debt equity conversion program on selected macroeconomic variables of a debtor country and discusses some pragmatic issues related to the implementation. It shows that benefits of conversions, primarily the promotion of additional investment and debt reduction, may come with short-run costs in monetary expansion carrying inflationary potential. But it demonstrates that these costs are more than offset by reductions in monetary expansion and lower service payments over the following several years due to the debt reduction. The paper suggests that should the government decide to establish a program, it integrate the likely effects conversions into the overall macroeconomic planning. In regard to the implementation of a program, the paper demonstrates that various mechanisms can have a great effect on the program's efficiency in meeting its objectives, and that the best way to promote additional investment would be to maintain a steady program of sufficient duration.  This paper was written by Messrs. John D. Shilling, Anthony Toft and Woonki Sung of the Financial Advisory Unit of the Cofinancing and Financial Advisory Services Department, World Bank, and Wayne Edisis of the Foreign Investment Advisory Services, IFC/MIGA whose involvement was funded by UNDP under Project INT/87/021. The views and interpretations in this document are those of the authors and should not be attributed to the World Bank, its affiliated organization or to any individual acting in their behalf.  TABLE OF CONTENTS Page No. I. Introduction ............................................ 1 II. Analysis of Debt Equity Conversion ..................... 3 Objectives and Trade-Offs .. 3 Investment Promotion .. 4 Macro-Economic Impacts .. 6 a. The Burden of the Debt and Conversions. 6 b. Potential Inflationary Impact. 8 c. Impact on the Balance of Payments .................. 15 d. Integrating the Debt Equity Swaps into the Macro Plan .19 e. The Asset Privatization Program .20 f. Aggregate Effects .23 III. The Mechanics of Debt Equity Conversions .24 Measures to Increase Additionality .24 Approval Process and Auction System .26 Preventing Round-Tripping .30 IV. Conclusion .32 ANNEX I. Analysis and Projections of Macro-Economic Impacts   - 1 - I. INTRODUCTION 1.01 In the course of the continuing debt crisis, a number of highly- indebted countries have initiated debt equity swap programs. Among them, Chile, Mexico, Brazil, Argentina and the Philippines have implemented more active programs than others. These countries aimed to achieve several objectives through their programs. Recently, however, some of the countries restricted substantially or suspended the programs primarily on the grounds that, as they perceived, conversions would entail macroeconomic costs and provide unjustified subsidies to foreign investors. 1.02 In a typical debt equity conversion deal, an investor, foreign or national, purchases title to a foreign currency-denominated debt at a discounted price on a secondary market. The investor then presents this title to the government authorities in the country which originally contracted for the loan. The national authorities next issue the face value in local currency to the investor, a portion of which may be retained by the government as a means of capturing for itself part of the secondary market discount. Finally the investor spends the local currency to implement an approved equity investment project. 1.03 Debt equity conversions provide a debtor country with an opportunity to prepay its foreign debt at a discount within the framework of debt restructuring agreements and to promote foreign investment through a preferential exchange rate. The conversion creates a cash bonus for an investor equal to the difference between the secondary market price (which represents the investor's expenditures) and the redemption price received after the government takes its share (which represents the investor's payout). The bonus can be effective for attracting new foreign investment. Conversions can also facilitate public service restructuring programs, such as privatization programs, channel investment funds to priority sectors, and encourage the return of flight capital. However, debt equity conversions are a double-edged sword. Depending on the structure of the deals, they may have macroeconomic costs, primarily adverse monetary/ inflationary and balance of payments effects. 1.04 The paper critically analyzes the potential impact of implementing a debt equity program on selected macroeconomic variables of a debtor country and discusses some pragmatic issues related to the implementation. The analysis is "partial" in the sense that it is not based on a full macro model of the economy. Nor does it try to "solve" for optimum values of the major macro economic variables. An important element of the analysis is placing a conversion program in a medium-term framework to identify the longer term impacts, positive and negative, on debt and payment flows, rather than just the first year effects. It highlights the  -2- trade-offs that decision makers will have to weigh if they decide to have a program. The approach in the paper is generic in the sense that the main elements of the analysis can be generally applied to conversion programs of a number of debtor countries although its analytical focus is primarily on the Philippine conversion program. l 1.05 The outline of the paper is as follows. Section 2 discusses several objectives that can be achieved through a conversion program and possible trade-offs among the objectives. It investigates how a program promotes investment. This is followed by an analysis of macro-economic impacts including debt reduction, potential inflationary and the balance of payments effects. Section 3 reviews the mechanisms of debt equity conversion processes and suggests ways to increase their effectiveness. In particular, it addresses several measures to improve additionality, the approval process, the auction system and safeguards against abuses of conversions. Section 4 concludes. The annex has projections of macro economic impacts of conversions. 1I The Philippine conversion program was announced through Central Bank Circular No. 1111, 1986 Series, Program for the Conversion of Philippine External Debt Into Equity Investments, of August 4, 1986 (the Circular).  - 3 - II. ANALYSIS OF DEBT EQUITY CONVERSION OBJECTIVES AND TRADE-OFFS 2.01 A debtor government could have several objectives to achieve through a debt equity conversion program. The stated objectives of the Philippine program, for example, as stipulated in Central Bank Circular No. 1111 are: to promote foreign investment, to encourage the repatriation of flight capital, to promote investment in priority sectors, and to reduce the external debt burden. Although the program can contribute to achieving all of the objectives, there may be conflicts among them in that achieving one may affect the capacity to achieve others. There are trade-offs between using conversions to stimulate investment and to retire foreign debt. Further, while all conversions entail macroeconomic costs, primarily adverse balance of payments and monetary/inflationary effects, the structure of a program has a bearing on the nature and magnitude of the costs involved. 2.02 If the government wishes to achieve primarily debt reduction through conversions, its conversion program should be structured to be flexible in screening and processing conversion projects. It should encourage a large volume of conversions for investment, additional or non- additional, and capture a large share of secondary market discounts for the government. The larger a portion of the market discount that the government captures, the more it can reduce its liabilities for a given expenditure, but the less of the discount will be left to stimulate additional investment. As a result, a large volume of debt may be retired but less additional investment may be promoted, and conversely. The less additional investment that occurs, the greater will be the short term drain on foreign exchange revenues, although this decline will be reversed in the medium term due to savings on future debt service payments. 2.03 On the other hand, if the primary objective of a program is to promote additional investment, then a larger share of the discount may need to be left to the investor and the program made more selective. To the extent the program is effective in promoting investment, there will be added benefits to the economy from the higher level of investment than would occur without the program. The government is likely to capture less of the discount, but the initial adverse effects on the balance of payments would be less than with little additionality. Depending on specific assumptions, the impact on the creation of reserve money will be favorable about the same, but the government will retire less debt per unit of expenditure if it increases the bonus to raise additionality. If a program were geared too strongly toward additionality, it would run the risk that the conversion volume would remain too small, with much lower impact on debt reduction.  -4 - 2.04 If a program is predominantly oriented to the promotion of foreign investment in priority sectors, it may help channel investment funds to the privileged sectors, which would not occur in the absence of the program. However, as experienced in some countries with a special program in place for incentive projects, enthusiasm for investment in priority sectors with low profitability may turn out to be difficult to sustain, even with a substantial exchange rate bonus to the investor. If this were the case the result could be an inactive program with insignificant amounts of foreign investment promoted and foreign debt retired, very large subsidies to investors participating, and the government capturing a very small, if any, portion of the discount. The potential for inefficiency resulting from such distortion is also large. 2.05 If the repatriation of flight capital is the primary objective, a program can be structured to provide a de facto amnesty to national investors to encourage the return of funds previously sent abroad. While there may be some political considerations to be taken into account, this process of using flight capital reflows to retire foreign debt through the swap program has a more favorable impact on the balance of payments because they are least likely to generate future transfer streams abroad. INVESTMENT PROMOTION 2.06 Additional Investment Although investors are not greatly influenced by post investment benefits, such as tax incentives, they tend to respond favorably to pre-investment advantages, such as the bonus generated by a debt-equity conversion. This is because investor are assured of receiving "up front" the full benefit of the bonus. Particularly, the bonus both lowers risk by reducing the total investment expenditure and heightens reward by creating a better rate of return. An analysis of conversion programs in Chile, Argentina, Brazil and Mexico has demonstrated the appeal of this bonus; it was estimated that over 602 of all debt-equity investments were 'additional" -- due to the existence of the program.1 However, it appears that additionality under the Philippine program has been low compared with the experience in Latin America. In other words, many of the investments made through debt equity conversions would have occurred anyway. 1/ Debt/Equity Swaps and Foreign Direct Investment in Latin America, J. Bergsman and W. Edisis, IFC Discussion paper Number 2, The World Bank, 1988, p. 8. In this paper, additionality is defined as meaning that the particular investor would not have made the investment if a swap were not available. Partial additionality means that something like the investment would probably have happened sometime without the swap, but that the swap made the investment happen sooner and/or made it larger.  -5- 2.07 The low additionality under the Philippine program to date should not be surprising. Most investment projects of consequential scale, especially when undertaken by foreign investors, require more than a year of advance planning. Consequently, an 'incubation period' is necessary before a conversion program can begin to attract substantial additional investments. The Philippine program using Central Bank paper was stopped, however, before many investors were past this critical time. Thus, most of the investment undertaken with CB swap program may have not been additional in the strict sense. It does appear that the program has contributed to making the size of the investment significantly larger than otherwise would have been the case, inducing new technologies or advanced equipment, and advancing the timing of foreign investment. 2.08 The conversion program may have produced other more important benefits to the Philippines. Its existence created confidence in the new Government's determination to restore a vigorous investment climate and to manage the country's large outstanding debt. In fact, when a country such as the Philippines emerges from a crises period, it may be useful to grant investment incentives without worrying about free-riders. The program does appear to have played a role in the overall recovery of private investment in the Philippines. Net foreign investment in the Philippines rose from the level of $17 million a year in 1984 and 1985 to $140 million in 1986 when the program was announced, $205 million in 1987 and $500 million during the first nine months of 1988. Out of $705 million investment recorded during the period of 1987 and the first nine months of 1988, over $640 million worth, representing 781 of the investments, was transacted through debt equity conversions. 2.09 Debt Reduction and Priority Investments. As of August 30, 1988 debt-equity conversions under the program retired the external debt of the Philippines in the amount of $660 million, representing about 62 of the country's debt owed to commercial banks eligible for conversion. Of the $660 million, about 902 involved Central Bank paper, with the balance involving private sector debt (92) and non-CB public sector debt (12). As a result, the country's interest payments will be reduced by approximately $70 million annually, assuming LIBOR of 9.5Z, which represents about 62 of the total interest payments to commercial banks. Of the conversions deals closed, about 502 with a value of $259 million as of September 30, 1988 were concluded by local investors. Although it is difficult to quantify, a majority of these national investors appear to have repatriated flight capital through the program. Of the transactions closed, 942 of the total value were Schedule 2 investments in priority sectors, with export producers, assets for privatization, and agricultural producers leading the list. 2.10 The perceived adverse monetary and inflationary effects have been a limiting factor for the Philippine program. However, the impact of conversions on inflation is a very complex issue requiring careful review of other factors leading to inflation before any firm conclusions can be drawn. These macroeconomic issues will be discussed in the next section.  -6- MACROECONOMIC IMPACT (a) The Burden of the Debt and Conversions 2.11 The debt equity swap program offers a debtor government the opportunity to reduce, at a discount, the absolute amount of debt that the economy has to service, to convert a portion of the fixed term external liabilities (debt) into contingent liabilities (equity), and to transfer some of the repayment liability from the government/Central Bank to the private sector while at the same time promoting private investment. Retiring debt at a discount has a high yield. In the case of the Philippines, in comparison to the payments stream under the current rescheduling, prepaying the debt at a 30% discount over face value yields a rate of return to the economy of about 22Z in current dollar terms. A 40Z discount yields 27Z rate of return.2 The difference between these discounts and the secondary market discount of about 50X of face value is the bonus that accrues to the investor. Central Bank debt swaps, if carefully designed, can generate cumulative gross savings on balance of payments outflow of about $250 million during 1989-93 per $100 million per year unit of debt retirement.3 In addition, swaps of government debt in the Asset Privatization Trust (APT) program will have a net positive impact on the government budget within two to four years and generate an internal rate of return to the budget of 40X or more. 2.12 The costs of the program are, depending on the type of swaps undertaken, the requirement to generate local currency resources to prepay part of the debt being swapped, the increase in foreign ownership of real rather than financial assets in the country, the additional demand for foreign exchange in the short run,4 and the possibility of fraudulent gains. The relative weight of these benefits and costs will be analyzed below for the principal kinds of swaps being considered in the Philippines. 2.13 There are generally three types of debt in the Philippines that could be used for swaps, and each has different characteristics and impacts on the macroeconomic equilibrium. Debt owed by private debtors may be swapped for equity of non financial private sector firms or prepaid. The primary effects are on the balance sheet of the debtor, who may reap a bonus from part of the secondary market discount; on the ownership and management of the debtor if equity is transferred; and on the net foreign exchange 2/ These rate of return calculations assume the Philippines will eventually repay the full value of its external debt. 3/ Counting DFI foregone in addition to savings on other flows gives a positive total cumulative impact in the sixth or seventh year, depending on other assumption. 4/ Including DFI flows implicitly used for debt prepayment and use of reserves for capital imports.  position of the banking sector, which will be higher over the term of the retired debt by the amount of the discount.5 The government has recently directed most proposed swaps to private debt to avoid the inflationary implication of using Central Bank debt. As a result of this increased demand in relation to a relatively small stock of debt, there is little private paper available in the market, and the discount has been sharply reduced. 2.14 Debt owed by the central government under current regulations may be converted and used in the purchase of public assets being privatized. Although this conversion is analytically the swap of a debt instrument for equity and thus a balance sheet transaction with no budgetary implications, the actual mechanism makes it more complicated. The prepayment of the debt (at a discount implicit in the 30Z fee charged by the government) is treated as a line item expenditure in the government budget. The investor then uses the funds received for the debt surrendered to pay the government for the assets he is purchasing. This payment is entered as a revenue item. The transaction is a wash as far as the current budget balance is concerned and has no direct inflationary impact, but the amount of swaps possible is determined by the budget allocation to that line, which is set by the legislature in the annual budget.6 After the swap, the government is no longer responsible for servicing that debt, which is extinguished. The private investor, if a foreigner, will service the equity liability through repatriation of dividends and eventually his capital, so there remains a future claim on foreign exchange, but not on the government budget. If the investor is a national, there is no further specific right or expectation to expatriate earnings or capital, and the foreign exchange obligation is also canceled. To the extent that the capital account is open, the Filipino investor would have the same right of any Filipino to convert Pesos into foreign exchange in the future. 2.15 Debt owed by the Central Bank may be converted into domestic currency for approved investments or into peso denominated debt instruments. The Central Bank has no physical assets to convert and must issue fiduciary instruments (cash or bonds). This is potentially the largest part of the swap program. This program has been temporarily suspended due to the high level of demand and the concern over the inflationary impact of the domestic 5/ To the extent that private firms can arrange to prepay their debt in tranches just prior to the maturity of each payment, they benefit from the discount and must do very little prepayment. This however requires complex intermediation from the banking sector as well. 6/ There is often a temptation to count revenues from asset sales as a source of current income in the budget. While technically possible, it is not usually wise since it is not a permanent source of income and in the absence of full asset accounting, tends to offer a misleading picture of government financing since the net decline in public wealth is not recorded. Swapping assets for debt, however, enhances public net wealth by the amount of the discount. And the fee thus generated can be used as exceptional income.  -8- currency that must be issued by the Central Bank to redeem the debt being swapped. Resources to prepay the debt must be generated by revenue raised by the government, by credit expansion, by money creation, or by issuing Central Bank debt in the local market. The first has an impact on the budget deficit, the next two on inflationary pressure, and the last on domestic capital markets, which are already pressed by the financing requirements of the government budget. These possible effects will be analyzed below. (b) Potential Inflationary Impact 2.16 When external debt is used directly to acquire domestic physical assets (private or government debt for assets), equity of the debtor (private debt for equity), or peso debt of similar or longer maturity (private or Central Bank debt for debt), there is no current element of the transaction to increase domestic demand or to expand money and credit. Thus there is no direct inflationary consequence. Even in the case where a private debtor prepays his debt in cash to an investor who then acquires other assets, there is not likely to be an increase in demand unless overall credit to the economy is allowed to expand to finance that transaction. The liquidity is generally transferred from the debtor to the investor. In the case of debt-for-debt swaps, however, when there is substitution of a domestic creditor for a foreign creditor, this may put pressure on domestic capital markets. In the case where the Central Bank retires its debt by issuing cash to be used for investment, there a potential inflationary impact. In view of the potential amount of these swaps, this issue deserves careful consideration. 2.17 The extent to which the monetary creation7 resulting from the swaps of Central Bank paper translates into inflation depends on several factors: the inflationary momentum, other demands for monetary expansion (e.g. the size of the consolidated government deficit or private loan demand), the degree of excess capacity in the economy, and the supply response of increased production and exports from the investment. The issue is an intertemporal one. The resources would have to be generated to purchase the foreign exchange to pay interest currently and later to pay amortization on the external debt when it falls due in 1993 and thereafter. The question is to what extent the economy can absorb the impact of generating the resources now (for retiring debt at a discount) or must postpone that until later (to repay at full value). As will be demonstrated below, the total net requirement to create additional money, or equivalent, is the same over a 6 year period for a swap program or for normal debt servicing; only the swap program front loads the monetary expansion. The internal rate of return to the Central Bank's budget of the swap would be 22-27Z, if the resources could be found to initiate the program. The net resource requirement declines after the first year. 7/ Budget financing would increase the deficit, and that would eventually require Central Bank financing in the current economic environment, with similar impact.  2.18 An analysis of the potential inflationary impact must take place in the context of the macroeconomic planning that sets growth and inflation targets within a consistent framework. That planning process must weigh a variety of factors, costs and benefits of alternative strategies, and the possible effects on expectations and levels of confidence in the government's program. The size and format of a debt equity conversion program is but one, relatively small, element in that planning process. Judgements about its potential utility over time might justify some modifications in the overall macroeconomic targets, but the swap program cannot drive the other targets. What is important is that the program be integrated into the multi-year macroeconomic framework, which is taken up later. 2.19 The actual impact on inflation will depend, of course, on other demands on credit creation. The most important of these is the expected size of the consolidated government deficit and its financing. The larger the level of credit or monetary creation required to finance the government deficit, the less that would be available for the prepayment of debt through a swap program within any overall target on credit expansion. The government deficit may also be financed by domestic or foreign borrowing. This would reduce the need for monetary creation, but would put pressure on the domestic capital market or strain the limits of the governments access to foreign resources. The size of the government deficit has been influenced by the recent political changes and the restructuring of major national development banks and public enterprises. As the impact of these factors is absorbed and passes, the government will be able to exercise greater control over the deficit and its financing. 2.20 To integrate a debt equity swap program into the macro framework, the government must weigh its impact on base money and the money supply, on supply and demand conditions in the economy, and on the credit markets. Resources for the swap can come from increased revenues, monetary expansion, or public sector borrowing. Use of either current revenues or monetary expansion definitively retires debt and yields a higher rate of return than borrowing, which still leaves the government with a debt servicing liability, albeit only in local currency. Assuming revenues cannot be raised in the short run, that leaves monetary expansion or borrowing. 2.21 One could take a dynamic view of the net impact on monetary expansion over time. Service payments are saved over time after debt is retired, the net demand on monetary expansion declines, and eventually results in a "reduction" in the demand for monetary expansion compared to the case with no swap. Under the assumptions used, the net demand shifts from an increase to a decrease in the fifth year, and the cumulative impact on reserve money is negative in the seventh year. That is, over a seven year period, a program of debt equity swaps has a zero net impact on monetary expansion. This is illustrated in Graph 1 where the annual and cumulative net impacts on reserve money are shown.8 The intertemporal element of the swap is important, as demonstrated in these figures.9 8/ For a unit swap program of $100 million per year for 5 years, 30Z discount captured by the Central Bank and no additionality. 9/ The same analyses and relation would apply if the analysis were based on base money rather than reserve money.  - 10 - Graph 1: PHILLIPINES Impact: Debt Equity Swap on Res. Money Bllion Pesos 10 5 .................................................................................................................................................. 0 5 . ................... ............................................ .............. ..............I....................... .....I....... ..........I. ............. ................................................................\.... _5 . .............................................................................................................I..............I.............. .................... .................................................................\.......... . -161 ..... ..' - Net Cumm. Change -Net Change -20 l l l 1989 1991 1993 1995 1997 1999 Year The effect of a unit swap program of $100 million p.a. for five years  - 11 - 2.22 Since contracts in the Philippines do not tend to be indexed and the economy is not hypersensitive to inflation, as is the case in Latin America, cost push factors stemming from other possible policy actions, such as a modest devaluation or raising the tariffs of public enterprises, should not be additive to the inflation resulting from monetary expansion over the expected ranges of these variables. In fact, those price increases would tend to reduce real demand and the public sector deficit, and thus they would lower the demand pressure on prices. 2.23 The above analysis assumes that the economy has little excess capacity and is near full employment. Any increment to demand would then have a greater impact on prices than production. To the extent that there is excess capacity in the economy, then the demand created by the swap program will have less impact on prices and more on increasing production. For example, if the investment encouraged by the swap program increased demand for construction to enlarge capacity, and if there are unemployed resources in the construction sector, production rather than prices will go up in the first round. Output in the construction sector in 1987 (the most recent data available) was 9Z below output in 1983 in nominal terms, and about one half the 1981-83 average in constant 1972 Pesos, so there should be substantial capacity to be tapped. Since labor is unlikely to be a bottleneck, it is likely that the extra investment demand generated by a swap program would have a larger impact on production rather than prices in the first round. 2.24 The income generated in the construction sector however would spill over into the rest of the economy and could drive up other prices unless there were general excess capacity. This second round inflationary pressure would be attenuated to the extent that the new investment increased the resources available to the economy domestically or for export. The second round impact will also be attenuated by lags and leakages, so directing the incremental resources into investment activities is likely to have a lower impact on inflation than general monetary expansion. More generally, any investment in additional, efficient, productive capacity will increase supply in the economy over time and dampen the inflationary pressure of the expenditure that created the investment. 2.25 The Central Bank has the option of sterilizing the monetary creation resulting from the debt equity swaps by issuing an amount of bonds equal to the payments to redeem the debt. This offsets the impact of the new money issued and effectively transfers the cost of prepaying the debt to the private financial markets in the short run. In the long run, however, the Central Bank will still have to raise the resources to retire the domestic debt, and in the meantime, the Central Bank will still be paying  - 12 - interest to domestic creditors. Whether the domestic interest payments will be more or less expensive depends on the discount the Central Bank is able to get when it redeems the debt and the relative levels of domestic and foreign interest rates. As shown in Table 3.1, if the interest rate on the rescheduled debt is 102 and the Central Bank redeems the debt at a 30% discount, its interest payments will be less as long as the domestic interest rate it pays is less than 14.3%, and similarly for other values of the discount. However, as long as the Central Bank is paying interest, the internal rate of return of the swap is lower than if the Central Bank had paid off the debt entirely. TABLE 8.1 EQUJlVALENT DOMESTIC INTEREST RATE Rate on Discount R-ceivod by Contral Bank /a External Debt 6% lOX 15X 20% 26% 30% 35% 40X 46X 60X 66% Sox 8.6X 8.9X 9.4 10.OX 10.6 11.3% 12.1 X13.1% 14.2% 16.X5 17.01 18.9X 21.3X 9.0X 9.5X 10. OX 10.6 11.3X 12.0% 12.9% 13.8% 16.0X 16.4X 18.0X 20.0% 22.5% 9.SX 1O.OX lO.1 X 11.2X 11.9X 12.7% 13.6X 14.6 X15.8X 17.3 X19.0X 21.1X 23.8X 10.OX 10.5X ll.lX 11.8%X12.5X 13.3X 14.3X 15.4 18.7X 18.2X 20.0X 22.2X 26.0X 10.5X ll.lX 11.7X 12.4 13.1 X14.0X 15.0X 16.2X 17.5X 19.1X 21.0X 23.3X 26.3% ll.OX ll.8X 12.2% 12.9X 13.8X 14.7X 16.7X 18.9X 18.3X 20.0X 22.0X 24.4% 27.6X /I 5X discount implios that the govornmont receives 5 por 100 of tho converted debt faco valuo.  - 13 - 2.26 The extent that the authorities can use this option depends on other demands on the capital markets and the target level of the interest rate. The larger the government deficit that must be financed by issuing treasury bills and notes, the less scope there will be for additional bond issues to finance swaps for any given interest rate target. 2.27 The Central Bank may also offer to swap foreign debt for long term local currency denominated debt to be held or used to repay domestic debt rather than to be turned into investment. This option is primarily appealing to firms with debts to the the Central Bank, to owners of flight capital, or to other potential investors who are looking for long term assets (e.g. pension funds). A program of debt for debt swaps with limited transferability in the short term might offer some possibilities for retiring debt if the combination of discount and interest rate can be made attractive enough. For example, if an investor were able to buy Philippine debt at a 50Z discount and redeem it at 701 of face value for a debt instrument paying 14Z, that would be equivalent to a nearly 201 return on the actual investment, which may be attractive. 2.28 The question of issuing long term bonds to be resold in the market to raise investment funds for current disbursement is more problematic. In a fully integrated capital market, this is no different than the Central Bank issuing the bonds and paying the investor directly, except there is more intermediation. It would have the same effect on the market: other borrowers would be squeezed and interest rates increased. Only if one can show that the market is segmented by time preference can one argue that this might raise otherwise unavailable funds. And it must be further argued that private investors would be better able to tap this long term market than the Central Bank. 2.29 In determining the size of a swap program, the authorities must consider the amount of money creation, debt expansion and reflow of rescheduled debt to the Central Bank that can be allocated to debt equity swaps, consistent with the expected rate of return on swaps and other macroeconomic targets, particularly inflation. In the following quantitative analysis, the effects of a unit program of expenditures of $100 million per year for five years on debt equity swaps are presented under different assumptions on other aspects of the program. If the Central Bank collects a portion of the discount in the debt conversion (through an auction or other mechanism) and retires that amount of money, or equivalently only redeems the debt at a percentage of face value, the amount of debt retired and the benefits will be larger for a given expenditure than through redemption at face value. 2.30 The experience elsewhere and the Philippine experience with the private debt suggest that the Central Bank should be able to capture 30-40X of the face value of redeemed debt when the secondary market discount is around 50Z. It is currently between 50Z and 60% off face value. Depending on the conversion rules, this could leave the investor with a bonus of up  - 14 - to 40-20Z on his invested funds, some of which will accrue to intermediaries. This should be a sufficient bonus for investors. With these parameters, a debt reduction program of $100 million equivalent per year in CB expenditures would retire $143 million in debt per year with 30Z of face value to the government and $167 million per year with 40Z to the government. The net money creation requirement in the first year, after taking account of reduced interest payments, would be about Pesos 1.8-1.9 billion ($90 - 95 million equivalent), depending on the level of conversions and discount to the Central Bank. TABLE 3.2 PHILIPPINES IMPACT OF 5 YEAR CENTRAL BANK SWAP PROGRAM ON MONETARY EXPANSION Expenditure on Debt Retirement ($mil/yr) 100 100 Percentage share of face value to CB (Z) 30Z 40Z Debt Retired ($millyr) 143 167 Debt Retired after 5 years ($ million) 714 833 Net Monetary Creation Year 1 (P. bil) /a /c 1.9 1.8 Net Monetary Creation Year 5 (P. bil) la Ic -0.4 -0.8 IRR (Z) on Reserve Money Creation /b Ic 22Z 27Z NPV at 10Z (P. bil) on Reserve Money Creation /b /c 4.3 6.2 (a Assumes 20 P/$. /b Compared to servicing the debt on the current rescheduled maturities. /c Results slightly more favorable if some additionality is assumed.  - 15 - 2.31 The reduction of future service payments that result from debt retirement considerably reduce future expansionary impact of a program as the Central Bank would not have to create or raise the domestic currency to buy the foreign exchange to make the service payments. For such a program, the net impact on monetary expansion turns negative in the fifth year. That is, the reduction in demand for resources to service debt is greater than the increase in demand for resources for the swap program. So if the inflationary potential of the first couple of years can be absorbed, the expansionary impact of the swap program will reverse itself and free monetary expansion capacity of the Central Bank for other needs.10 In addition, expenditure leakages into imports of capital goods will reduce the inflationary potential. If the imports of incremental investments are financed by using banking sector foreign reserves, then overall domestic demand could be reduced by Pesos 600 million. The Internal Rate of Return (IRR) on reserve money creation is 22-27% and the net present value of the increased expenditures (at 10Z) is Pesos 4.3-6.2 billion. (c) Impact On The Balance Of Payments 2.32 Debt equity swaps give rise to direct foreign investment (DFI) without an actual inflow of additional foreign exchange. External liabilities are reduced instead, and any difference between the value of the debt reduction and the size of the investment (due to the discount or fee captured by the Central Bank) is accounted as a capital gain on asset values. Alternatively, one can assume that the foreign exchange that would have been used to buy Pesos to make the investment is immediately used to prepay foreign debt at a discount. The prepayment saves future foreign exchange payments by the Central Bank and has a high rate of return. But, the economy does not gain access to the foreign exchange that otherwise would have flowed in with the DFI, which reduces the availability of free foreign exchange in the short run. The extent to which this intertemporal advancing of payments can be made depends on the overall tightness of the foreign exchange markets and the additionality of the swap financed investment. 2.33 The question of the availability of foreign exchange must be resolved in the country's overall economic planning. The example of $100 million per year in expenditure on a swap program would substitute for a like amount of DFI inflows and reduce debt by $143-167 million per year. The net impact of foregone DFI inflows and lower service payments on gross 10/ In this analysis, it has been assumed that the Central Bank issues money to prepay the debt through the swap program, or to purchase the foreign exchange needed to service the debt without the swap program. Obviously other resources (budget transfers) could be used in either case, but assuming the government is strapped for resources, the analysis is going to come to the same point, e.g. increased government transfers will increase the budget deficit and require more Central Bank financing. If the resources were available without increasing the deficit, then they could be used in the swap and eliminate the inflationary impact.  - 16 - revenues ranges from $71 to 93 million in lower reserves in the first year due to lower service payments, but the negative impact declines and becomes positive in the fifth year.11 (See Table 3.3 and the detailed tables in Annex I.) The total cumulative impact on the balance of payments becomes positive after the sixth year if there is no additional investment. With 30% additionality, this break even period may be shortened by one year.12 This level of foreign exchange use in the first year is less than 3? of exports and less than one third the annual average improvement in net international reserves of the banking system and Central Bank in 1986-87. Nevertheless, since the level of reserves is quite low, this use will have to be evaluated against alternative demands for reserves. In making such comparisons, it should be noted that the internal rate of return on this use ranges between 22Z and 33Z. The Net Present Values are between $213 and 336 million per $100 million of program. TABLE 3.3 PHILIPPINES IMPACT OF 5 YEAR CENTRAL BANK SWAP PROGRAM ON BALANCE OF PAYMENTS a b c d Expenditure on Debt Retirement ($mil/yr) 100 100 100 100 Percentage Share to Central Bank (Z) 30? 30? 40? 40? Additionality factor (X) o0 30Z 0? 30? Debt Retired ($mil/yr) 143 143 167 167 Debt Retired after 5 years ($mil) 714 714 833 833 B of P Saving ($mil/yr) year 1 7 -2 8 -1 B of P Saving ($mil/yr.) year 5 121 103 142 124 Net B of P (inc DFI foregone $mil/yr) year 1 -93 -72 -92 -71 Net B of P (inc DFI foregone $mil/yr) year 5 21 33 42 54 IRR B of P 22Z 27? 27? 33? NPV at 10? B of P ($mil) 213 237 312 336 Note: Column heading letters refer to sets of tables in Annex I which show greater detail over the period 1989-2000. 11/ This takes account of dividend remittances on additional investment as well, which are a burden on the balance of payments but not the Central Bank's domestic resources. Data is limited on the rate of profit remittances in the Philippines some firms reinvest their earnings rather than repatriate them for tax and business reasons. There is strong evidence that much of the swap investment is return of flight capital, and that investment would not generate a demand for repatriation of earnings. 12/ To the extent investment is additional, the foregone DFI is, of course, lower.  - 17 - 2.34 When there is no additional investment, the reduction of the external debt is greater than the prepayment cost to the Central Bank (in proportion to the discount received by the Central Bank) and the future repayment liability on the "reduced" foreign claim is transferred to the private sector (or eliminated if the swap is made with flight capital). The amount of foreign exchange that would otherwise have flowed in is reduced in favor of a reduction in debt at a discount. When there is some additionality to total investment, there will be added benefits from the higher level of investment and growth fostered by the swap program.13 The lack of free foreign exchange due to the additional investment is not lost foreign exchange, since that DFI would not have occurred without the program. There will be additional current demand for imports to realize the investment and future demands for dividend and capital repatriation, which would put an additional demand on available foreign exchange. This demand can be partly reduced by the design of the program and will be partly offset by the foreign exchange savings in service payments. The rest of the increased demand for foreign exchange will have to be absorbed through using reserves. Available data from the Central Bank indicates that 23Z of investment project costs go to imports in the Philippines, experience elsewhere suggests that about 40Z go for imports. To be conservative, 30% of project costs are assumed to be imported. 2.35 Table 3.3 shows the full impact on the balance of payments of a five year, $100 million per year Central Bank swap program with no additionality and with 301 additionality. It assumes the current repayment schedule of rescheduled debt and a 10 of assets dividend transfer rate 2 years after the investment is undertaken. A 10? interest rate is assumed. Half of interest payments is saved in the year the debt is retired, and all interest is saved in subsequent years. 2.36 Some countries have tried to mitigate the additional demand for foreign exchange by requiring that the investor to use swapped funds only for local expenditures and that he finance his imports separately. However, since the Philippines capital account is relatively open, the net impact would be to reduce the amount of debt retired through each project in the program and the share of the discount obtained by the government. If the investor can only use the swap for a smaller portion of his investment, he will demand a larger share of the discount on that portion to get the same overall bonus. It would probably be better to set the level of annual swaps at an amount that can be absorbed, capture as much of the discount as possible, and dispense with as many constraints as possible. 2.37 Table 3.4 compares a straight foreign investment with (a) the case of a Full Swap with no additionality, where the investor is not required to 13/ It may also be argued that the increased investment brings technology transfer and other disembodied gains to the Philippines, but that analysis is beyond the scope of this paper.  - 18 - put up additional foreign exchange for imports, (b) a Domestic Cost Swap with no additionality, where the investor is required to pay separately for his imports, (c) a Full Swap that is fully additional, and (d) a Domestic Cost Swap that is fully additional. A constant 302 bonus to the investor in each case has been assumed. Other assumptions are as stated in the table. For simplicity a single currency unit is used for both domestic and foreign currency. This results in no loss in generality. TABLE 3.4 PFiILIPPINES COMPARISON OF EFFECTS OF VARIOUS IMPORT REQUIREMENTS No Additionality Full Additionality (,)7 (b) (C) (d) No Full Domestic Full Domestic Investor Impact Swap Swap Cost Swap Swap Cost Swap (1) Investment Project Cost 100 100 l00 100 100 (2) Investment Financed by Swap - 100 70 100 70 (3) Debt Retired [(2)/(1-(9))] - 154 98 164 93 (4) Investor FX to Rotire Debt ((3)s(1-(a))J - 77 47 77 47 (6) Use of Investor FX for Imports ((1)*(c)] 30 - 30 - 30 (6) Use of Domestic FX for local costs 70 - - - - (7) Total Inv-stor FX Costs ((4)+(6)+(6)] 100 77 77 77 77 (8) Foe to Government t(3) -(2)] - 54 23 54 23 (9) As Share of Debt Retired t(8)/(3)] - 3S% 26X 5% 25% (10) Reserve Money Used to R-tir- Debt (net) ((2)] - 100 70 100 70 (11) Reserve Mon-y/100 of Debt Retired [(2)/(3)*100] - 65 75 a6 75 Balance of Payments Impact (12) Inflow of FX for Investment 100 - - - - (13) Outflows of FX for Imports [(1)*(c)] 30 30 - 30 - (14) Net Change in Reserves [(12)-(13)] 70 -30 - -30 - (15) Net Change Compared to No Swap - -100 -70 -30 - (16) Net Change per 100 of Debt Retired [(16)/(3)*100] - -86 -76 -20 0 Assumptions (a) Market Discount 30% (b) Investor Conuo 30% (c) Import Content 30%  - 19 - 2.38 For an investment project of 100, foreign investment without a swap brings in a net of 70 in foreign exchange (100 investment less 30 used for imports). A non-additional Full Swap will retire debt worth 154, but uses 30 in reserves to pay for imports. The investor keeps a bonus of 23 in local currency equivalent (100-77) and pays fees to the government of 54 (154-100). The difference between regular DFI and the swap is a net lower level of reserves of 100, less the reduced stream of debt service payments, which are not included here. In the Domestic Cost Swap, only 93 of debt is retired. Because of the smaller amount of debt retired and the smaller base for the bonus, the government is only able to collect a fee of 23 equivalent (93-70). The net impact on reserves is that they are lower by 70, but less debt is retired. It would take additional investment projects to retire as much debt as in the previous case. Under these assumptions, it takes 65 net use of reserves and emission of reserve money to retire 100 of debt in a Full Swap, but 75 with a Domestic Cost Swap. With a Domestic Cost Swap program, the amount of fee the government can collect is very limited for a given level of bonus and market discount. If the market discount fell to 40Z, the government would not be able to collect a fee if the 302 bonus, and a lower amount of debt could be retired for a given level of Central Bank expenditure. (See Annexes I for detailed analytic proof.) 2.39 If the investment project is fully additional, the relation between the fee and bonus remains the same, as does the use of reserve money. However, the balance of payments situation improves. With a Full Swap, the net impact on reserves is -30, while for the Domestic Cost Swap, there is no immediate impact on reserves. However, this favorable result assumes the investor acquires his foreign exchange for imports entirely from sources outside the Philippines. If he is able to use Pesos to acquire them from the foreign exchange holdings of Filipino banks, then country loses the favorable balance of payments impact and retires less debt than in the Full Swap case. In considering whether to require investors to finance their own imports, the government should consider that a Full Swap program will retire a larger amount of debt per unit of expenditure for prepayment and per project than a Domestic Cost Swap, but that for additional investment projects, the use of reserves will be greater in a Full Swap if the imports are fully financed outside the Philippines. (d) Integrating The Debt-Equity Swaps Into The Macro Plan 2.40 The above analysis shows that the benefits from the debt equity conversion program come with short run costs in terms of monetary expansion carrying inflationary potential, possible upward pressure on interest rates, and reductions in the availability of unrestricted foreign exchange. While these costs are more than offset by the reduction in monetary expansion and lower service payments that occur over the following several years due to the debt reduction, the short run impacts must be incorporated into the country's macroeconomic planning.  - 20 - 2.41 The government must first establish the priority of debt reduction and investment encouragement in its overall macro targets, based on the benefits described above. It must then determine how much of its "resources" from monetary expansion, increased resource mobilization, higher interest rates, and/or reduced reserves that it can afford to apply to debt reduction. These resources should be budgeted in the overall macroeconomic planning; the amount of the swap program firmly established in terms of the mechanisms described in the next section; and the program implemented in a consistent manner. By budgeting the resource in this way, the effects will be already included in the overall projections, and the government will avoid surprises. 2.42 In its decision making, the government must also decide whether to allow Filipinos to participate in the program and whether to require investors to bring in new money. Excluding Filipino investors would reduce the possibility of the Central Bank's or banking system's own reserves being used to acquire debt for conversion and increase the likelihood of limiting the program to non-national sources of foreign exchange. It would also reduce the opportunity to attract flight capital. However, given the relative openness of the capital markets and the ease of setting up dummy offshore corporations, such a prohibition is unlikely to have much impact on Filipino investors who really want to take advantage of the swap program, and those who do through offshore corporations will have rights for dividend and capital payments to the overseas entity nominally undertaking the investment. On the other hand, allowing Filipino participation is more likely to encourage the repatriation of flight capital, which does not give rise to future capital outflows. This also avoids creating incentives to evade the regulations. Careful monitoring of financing sources should be instituted to limit the ability of investors to borrow locally and acquire foreign exchange on the local market in order to buy debt for conversion. 2.43 The analysis above shows that requiring the investor to finance his imports with funds from sources outside the swap program will have a more favorable effect on the balance of payments only if the investment is strictly additional and if the resources are foreign exchange owned previously by the investor or borrowed externally without displacing other lending to the Philippines (e.g. using up ECA cover limits). Otherwise, that requirement will not improve the balance of payments impact compared to allowing the imports to be financed with proceeds of the swap. On the other hand, requiring the imports to be financed separately reduces the share of the market discount the Central Bank can collect for a given bonus to the investor and reduces the amount of total debt the Central Bank can retire for a given expenditure of reserve money. The authorities will have to take these factors into consideration in designing their program. (e) The Asset Privatization Program 2.44 It has been argued that using debt to privatize assets implicitly increases the government deficit because in the absence of the swap program, the government would have received cash. This is a narrow  - 21 - viewpoint. First, since the swap retires debt at a substantial discount, the use of the program to reduce debt has a high rate of return and significantly reduces future budget expenditures allocated to service the debt. Second, the government collects a fee for the conversion which reduces the budget 'loss'. 2.45 Three factors influence the attractiveness of the assets being privatized: the final cost to the investor, the investment code benefits he will receive; and the retained bonus on debt swapped in the transaction, if any. Under the current rules, the investor does not know if he will be able to use swapped debt in his purchase, so he must bid on the assumption that he will not. Any additional bonus from the swap is a windfall to the investor, except for the fee charged by the government. By redesigning the program to allow the potential investor to include the use of the swap in his price calculations, the government can increase its gain, either by getting a higher net price or obtaining a larger fee. In effect, this will reduce the bonus to the investor from the swap and finance a greater part of the cost of the privatization with the discount offered on the debt. 2.46 For example, under the current system, an investor might successfully purchase a Pesos 1 billion (book value) asset with a bid of Pesos 250 million. If he is then able to acquire his Pesos through a debt swap, he will have to pay $9 million (Pesos 180 million) to acquire Pesos 360 million face value of debt (at a 50Z discount) to swap for Pesos 360 million with which to pay the 30Z fee (Pesos 110 million) and end up with Pesos 250 million to pay for the asset. If the swap rights were included in the bidding, it is likely that the government could capture some of the Pesos 70 million bonus he made on swap. Assume that the government is able to capture Pesos 40 million of the bonus in a higher fee. The investor will then have to buy debt for $10 million (Pesos 200 million) and swap the debt for Pesos 400 million to pay the fee of Pesos 150 million and pay Pesos 250 million for the asset. The net foregone budget revenue in the first year would be Pesos 100 million, (250 foregone revenue less 150 fee) compared to the deal with no swap, or to Pesos 150 million under the current swap program. But the $10 million purchase of discounted debt retires $20 million in obligations, and the government saves interest payments of Pesos 40 million per year for the next 5 years, and larger amounts in subsequent years when amortization payments would begin. The net impact on the government budget turns positive after the third year. If the assets privatized were earning net revenues, the savings on future service payments would have to be reduced by the foregone revenues. 2.47 Tables 3.5 shows the impact of a four year program of $100 million in swaps per year on the budget for a 30Z and 402 fee respectively. The net impact on the budget is small if the fee is about 40Z, which would still leave the investor with a bonus of about 202 on his outlay. About one third of the government debt would be retired and almost half of the privatization program financed. The balance of payments impacts are relatively small.  - 22 - TABLE 3.5 PHILIPPINES IMPACT OF 4 YEAR APT SWAP ON BUDGET AND BALANCE OF PAYMENTS la Expenditure or Debt Retirement (P. bil/yr.) 2.0 2.0 Percentage Share to Government (Z) 30Z 40Z Debt Retired ($mil/yr.) 143 167 Debt Retired after 4 years ($ mil) 571 667 Payment to budget (P. bil) year 1 1.0 1.5 Payment to budget (P. bil) year 5 -2.3 2.7 Net impact on budget (P. bil) year 1 -1.0 -0.5 Net impact on budget (P. bil) year 5 2.3 2.7 Net B of P effect ($mil) year 1 -93 -92 Net B of P effect ($mil) year 5 114 133 IRR budget 42Z 87Z IRR B of P 22Z 27% NPV at 102 budget (P. bil) 6.4 9.6 NPV at 10X B of P ($mil) 185 268 /a If it is assumed that some APT operations are additional because of this program, the B of P effect becomes more favorable.  - 23 - (f) Aggregate Effects 2.48 Table 3.6 shows the combined impacts of the Central Bank and APT programs at the levels suggested above. The possible benefits are quite impressive. The amount of debt retired over the 5 year period would range from $1.3 to 1.5 billion or up to 10X to 122 of the commercial debt currently outstanding. Service payments would be reduced by up to $0.3 billion per year. The rate of return on resources used in this way is high. The costs to the economy are added pressure on monetary expansion (inflation), on interest rates (if more domestic debt is issued), and on reserves. These are short run impacts, as the resources to repay the debt would have to be mobilized at some point in any case. Although these costs will be offset by savings on debt service payments, it is important to exercise care that they do not disrupt the government's efforts to maintain price stability and appropriate real interest rates. If the swap program can be managed within the limits of prudent macroeconomic policy, it is likely that the program will stimulate additional investment and attract substantial amounts of flight capital back to the Philippines. These benefits will depend on the credibility and continuity of the program and on some modifications in its design suggested below. TABLE 3.0 PHILIPPINES AGGREGATE IMPACTS OF CENTRAL BANK AND APT SWAP PROGRAMS a b c d Expenditure on Debt Rotirement (Smil/yr) /a 200 200 200 200 Percentage Share to Central Bank and govornment (N) 30X 30% 401 40X Additionality factor (X) (Central Bank only) OX 301 OX 30% Debt Retired (Smil/yr) /n 286 286 333 333 Debt Retired after 6 yoars (Smil) 1,286 1,286 1,600 1,600 Not B of P (inc DFI foregone tmil/yr) yoer 1 -186 -135 -183 -132 Net B of P (inc DFI forogone Smil/yr) year 5 136 139 175 178 IRR B of P 221 29% 271 36X NPV at 10X 8 of P (Smil) 398 489 S80 651 Net Budget/Monetary Impact (P. bil/yr) year 1 /b 2.9 2.8 2.3 2.3 Net Budget/Monetary Impact (P. bil/yr) year 6 -2.7 -2.8 -3.5 -3.6 IRR Budgot/Monetary Impact 29X 29X 42X 43X NPV Budget/Monetary Impact (P. bil) 10.7 10.9 16.8 16.0 NB. Column hoeding refer to sets of table. in Annox I showing groat-r detail. /a For first 4 years, fifth year is less as the APT program in not functioning in that year. / (C) = inflationary impact.  - 24 - III. THE MECHANICS OF DEBT EQUITY CONVERSIONS 3.01 The concept behind debt-equity conversions is relatively simple, but the mechanics of designing and implementing a working program are complex. This complexity arises because governments often try to achieve several objectives through these transactions. At best, these objectives overlap; at worst, they conflict. For example, one trade-off in designing a program pits the goal of administrative ease against the objective of preventing round-tripping. A further area of potential conflict is the problem balancing the participation of local investors with the necessity to control the impact on the balance of payments. The result is that there is no "perfect" design for a conversion program. Each element requires compromises between competing goals, a trade-off which every government must decide for itself. The policy makers first should decide on their relative priorities among the objectives of debt reduction, investment encouragement, repatriation of flight capital, and priority sector development. Once these priorities are clear, a debt-equity swap program can be designed to conform to those relative priorities. Measures to Increase Additionality 3.02 In order to increase additionality in any program, the best course would be to maintain a steady program of sufficient duration to outlast the necessary incubation period of potential project preparation. A program that switches on and off would not achieve this. In order for potential investors to undertake feasibility studies, they need to know that a government has a long-term commitment to permitting debt-equity transactions. Consequently, if a debtor government decides to establish or resume a program, it should consider announcing a minimum of a five-year program. Although the government could suspend this commitment before the period expires in case of an emergency, the initial commitments would be a valuable signal in the meantime for stimulating interest among possibly additional investors. 3.03 Another significant factor affecting additionality is the size of the bonus that investors receive. If the incentive is too small, then it will not prove sufficient to attract the interest of uncommitted investors. It is, of course, difficult to determine how large a bonus is enough. There is, for instance, the interest among Philippine investors to acquire and convert private-debt paper, even when this produced a bonus of under 1O, but this may not be additional. However, given the large demand for swaps, the government should not allow too large a bonus, or the implicit subsidy to non-additional investors will be hard to justify and the distortions in investment incentives will be too great. Since Philippine debt now is selling at discounts of 50-60Z, the Central Bank should be able to capture a substantial portion of the discount and still leave a handsome bonus for the investor. This would help increase the amount of debt retired.  - 25 - 3.04 Several factors affect the size of the bonus actually received by the investor: the share of the discount claimed by the government for converting the debt to local currency, issuing a bond for the debt that has to be negotiated at less than face value, taxing the capital gain created by the conversion, or requiring "fresh money" in addition to the conversion. As analyzed above some, such as requiring "fresh money," reduce the efficiency of the swap program in reducing debt and offer few benefits. It is best to keep the conversion process as simple as possible. 3.05 If an auction is used, limiting too tightly the total amount of conversions may push the resulting bonus below the level needed to stimulate additional investment. This is because non-additional investors could bid down the amount of the bonus to a low level. To avoid this problem, the market created by an auction should be deep enough to prevent non-additional investors from "squeezing out' additional ones. The size of each auction needs to be set with this situation in mind. 3.06 Additionality may be increased by reserving certain benefits for 'regular' investments made without a debt-equity conversion. The idea is to strike a balance between the advantages available to normal investment and the bonus produced by a conversion. If this balance is achieved, then non-additional investors (i.e. those investors who do not need a conversion bonus) will be no worse-off if they elect to make a regular investment. In practice, striking a perfect balance will be hard because of the large advantage of a conversion. If the government considers that the incentives created by a conversion bonus plus the normal investment incentives, which conversion projects are currently entitled to are too large, the government could consider removing some of the normal incentives for investments made through conversions. In Argentina, a successful bidding under the country's conversion program requires the removal of tax benefits that could be available under different federal programs. This Argentinean policy does not appear to have deterred additional investments made through conversions. 3.07 Under Circular 1111, priority projects encountered no obstacle on dividend payments; they could remit earnings from the first year onward. Moreover, they could begin to remit investment principal beginning at 20Z in the fourth year. These aspects may be tightened without weakening the appeal of the conversion program for additional investors. Such restrictions are generally imposed under conversion programs in operation in Chile, Brazil and Argentina. In fact, interviews with over one hundred investors revealed that significant (although not excessive) restrictions on the timing of both dividend and capital transfers were not perceived to pose an obstacle for additional investment.1 11 Debt-Equity Swaps, op cit, p.25  - 26 - Such restrictions are found to have been responsible for convincing some non-additional investors to pursue a regular investment instead of applying for a conversion. It would therefore be advisable for the Philippines to consider the creation of a waiting period for dividends and to lengthen the existing time frame for repatriating principal.2 Approval Process and Auction System 3.08 The demand by investors for debt-equity conversions normally exceeds the limitations set by a government on the size of the program. Conversion rights are thus a scarce commodity, which a government needs to allocate. Different allocation methods are available, and the choice that a government makes among the alternatives depends on its priorities and will shape the overall nature of its program. 3.09 Conversions may be allocated by government fiat. Under this method, a government receives a pool of applications and then evaluates each project against a set of general criteria. These criteria reflect a government's judgment on the types of investments that are desirable. The goal is to identify the applications that meet these criteria and to reserve the conversion bonus exclusively for them. This case-by-case method enables the host government to exert a greater degree of influence over the direction of the investment. For example, in this approach, the government can encourage investment in certain sectors, ensure a higher level of technology transfer in appropriate cases, and insist that certain investments have a higher export orientation. Essentially, this is the system of Chile for debt/equity conversions under Chapter 19 and that which operated under Circular 1111 in the Philippines. 3.10 On the other hand, however, there are potentially both economic and administrative problems with allocating conversions by government decision. There is no guarantee that the approved projects will be economically efficient. The government may try to judge, ex ante, whether a project will be economically viable, but the government's ability to foresee the vagaries of the marketplace may be limited. Consequently, this system of allocating conversions may provide bonuses to unsound projects, while denying assistance to efficient investments. Furthermore, there is no guarantee that a system of government decision- making will be able to set fees that capture the largest available portion of the secondary-market discount. 3.11 Administratively, government decision-making creates a heavy workload for the bureaucracy, which must examine the details of every application. Often, a complete evaluation requires inputs from several government agencies, whose separate analyses must be coordinated. 2/ Such convertability limitations may require consultation with the IMF.  - 27 - Moreover, this process is time consuming, which forces investors to wait for a reply. Under the procedures of Circular 1111, investors often had to wait for up to two months before receiving a decision whether they could use the debt equity swap program, assuming that they had submitted fully complete applications. Finally, this method lacks transparency. No matter how professionally the government evaluates applications, the process still occurs behind closed doors, giving rejected and disgruntled investors the grounds for complaining about a lack of fairness. 3.12 An alternative method for allocating conversions is to create a competitive market through an auction. Applicants may bid against each other for access to a fixed amount of conversions. This method will automatically screen projects for economic efficiency, because the most profitable projects should be able to bid the highest. Also, it will deliver the largest possible portion of the secondary-market discount to the government. Additionally, this market-driven approach would be far simpler to administer than by government approval. It would require less manpower, produce results quicker, and be fully transparent. 3.13 To assure that certain government concerns about investment allocation under the market method of approval are met, a "negative list" may be prepared, specifying sectors that are off-limits to debt- equity conversions. The forbidden sectors would include any area of the economy from which private investment is normally excluded. Also, the list would set aside sectors where the present level of investment is already adequate, or where the government believes that a bonus is not needed. This would provide an a priori basis to exclude sectors in which the financial returns are already high enough to encourage investment. For example, given the broad level of trade protection in the Philippine economy, the government may wish to exclude some highly protected import substitution subsectors from the conversion program. 3.14 Brazil and Argentina have chosen market based auction procedures to allocate access to conversions. Since early 1988, there have been ten auctions in Brazil and eight in Argentina. Neither Brazil nor Argentina exclude domestically-oriented investments from the conversion process. From the experiences of these auctions, the following several observations can be made. 3.15 First, using auctions need not necessarily preclude screening of conversion projects by the government, although the screening process in this case tends to be lighter than under the negotiation approach. For example, in Argentina the Ministry of Economy screens eligible  - 28 - projects which may be presented to Central Bank auctions. This process may not be onerous but strict enough to screen out conversion applications for patently non-productive investments. Second, establishing different auctions for investment of different priority seems to work. In Brazil, there are two different auctions: one for "incentive" projects in some less developed, priority regions in the northeast and the other for "non-incentive" projects anywhere else in the country. The first auction has channeled almost as much investment funds as the second, although the auction discounts in the first were considerably lower than in the second. In Argentina there is a separate functioning auction which is limited to the prepayment of Central Bank medium-term rediscount facilities, in addition to the one for regular investments. 3.16 Third, commitments by the government to the market system and its efficient administration are crucial to the success of auctions. In both countries, the government establishes in advance conversion ceilings for each auction. In Argentina, for example, based on macro- economic considerations, the government has established auction amounts of $50 million in conversion value for every bi-monthly auction on a five-year basis. Presentation to the auction must be made through a local bank (Agent Bank) authorized by the Central Bank to operate foreign exchange. Acceptance of bids is decided strictly on the basis of the auction discount (Central Bank fee portion) offered for the conversion. In Brazil, the auctions start with an initial discount of 0.5Z. The minimum amount to be offered is $100,000, and only multiples of the minimum amount can be offered. If the total amount offered by the participants is greater than the ceiling of the Central Bank, the discount will increase by 0.5Z. This process will continue until the total offer is equal or lower than the ceiling. 3.17 Fourth, results under these auctions generally support the view that this method captures a larger share of the discount for the Central Bank than under the negotiation method. As Table 4.1 shows, although they are subject to variation, the auction discounts (central bank fee portion) as well as the government's share of the secondary market discounts have been on the rise as the market price of debt declined over the last several months. The average government share in the difference between face value and secondary market price was 69Z in Brazil and 77Z in Argentina, which were higher than in Chile (less than 30%) where the negotiated method was used.  - 29 - TABLE 4.1: Auction Discounts, Market Prices, Government's Share and Incentives to Investors (Z) Country Av. Auction Market Govt.'s Share/i Incentive to/2 and Date Discount Price of Market Discount Investors (of face value) (of face value) Argentina /3 Jan 20, 88 36.7 32.0 54.0 97.8 Mar 29, 88 53.9 27.0 73.8 70.9 Jun 10, 88 57.7 27.0 79.0 56.8 Sep 22, 88 66.1 24.0 87.0 41.1 Dec 15, 88 71.9 21.5 91.6 30.7 Mar 28, 89 72.0 18.0 88.0 56.0 Average 59.7 78.9 58.9 Brazil 14 Mar 88 27.0 46.0 50.0 58.7 Apr 88 32.0 50.0 64.0 36.0 May 88 22.0 54.0 44.9 44.0 Jun 88 13.5 52.0 28.7 66.3 Jul 88 24.0 52.0 58.7 40.4 Aug 88 29.5 47.0 55.7 50.0 Sep 88 34.5 47.0 65.1 39.4 Oct 88 38.0 47.0 71.7 31.9 Nov 88 50.0 42.0 86.2 19.0 Dec 88 59.0 39.5 81.0 29.1 Average 32.3 68.6 41.5 Chile Average 11.0 /5 59.0 26.8 50.8 Source: Prepared from the information in Debt Auction and Conversion Data, Latin Finance, January 1989 and a report of Banco Rio del la Plata S.A. IL Estimated using the formula: average auction discount/ (1 - market price of debt) /2 Estimated using the formula: [(1 - average auction discount)/market price of debt)] - 1 This ignores commissions, taxes and other expenses for arranging conversion transactions. Thus, actual incentive to the investor would be smaller than the figures indicated here. /3 Auctions for regular investments. 14 Free area auctions. (5 Central Bank fees.  - 30 - 3.18 Finally, these auctions appear to have been effective in achieving conversion objectives. As the conversion volume increased and the level of auction discounts rose, substantial amounts of foreign debt were retired. About $2 billion in Brazil and about $1.3 billion in Argentina, in face value of debt were retired during 1988 through the auctions. In Argentina where local participation is allowed, more local investors than foreign ones have participated in the auctions, repatriating flight capital to the country. Conversions have financed a variety of investment projects with total costs of approximately $700 million. Specific design of an auction in the Philippines, if it were so decided, would have to be the object of further discussion between the government and its financial advisors. The detailed design would be a function of the government's objectives, and the rules and customs of financial transactions in the Philippines. Preventing Round-Tripping 3.19 Debt/equity conversions create arbitrage opportunities (round- tripping), which, if left unchecked, could lead to serious abuses of the conversion program. For example, in debt equity swaps, investors that would otherwise retain earnings in the country are encouraged to take funds out and bring them back through swaps, taking advantage of the preferential exchange rate. Careful design and implementation of the conversion program can help reduce round-tripping. It may be minimized at the outset by checking the credentials of investors, but verifying their reliability in advance of the actual conversion is far from foolproof. Also, as in Chile, the differential between official and parallel exchange rates can be closely monitored by the authorities to control volumes of conversions and incentives for round-tripping. These measures will reduce the potential for round-tripping in general. 3.20 In specific projects, further monitoring can limit this danger. In the past, the Central Bank has guarded against round-tripping by controlling the disbursement of converted funds. Investors have had to present invoices and other valid receipts to the Central Bank before funds could be paid directly to the suppliers of goods and services. This arrangement cast the Central Bank into the role of paymaster for the conversion program. Its staff had to check the accuracy of every demand for payment and determine its legitimacy. Needless to say, this imposed an administrative burden, which represented a "hidden cost" of maintaining a conversion program. 3.21 By using an alternative method for disbursing funds, the Central Bank may be able to lighten its administrative burden without sacrificing the adequacy of controls against round-tripping. Following a system which was effectively utilized in Mexico, direct responsibility for paying-out funds could be transferred to Philippine commercial  - 31 - banks, which would be required to apply the same rigorous verification procedures now followed by the Central Bank. It should not be difficult for the commercial banks to handle the disbursement of funds efficiently; many of them have already adopted similar safeguards for the release of loans for projects they finance. In order to meet the costs of verifying invoices and providing other services, commercial banks could be permitted to charge investors a small fee. To the extent that the same commercial banks also operate as the investors' agents in conversions, they can include the disbursement fee in their overall charges. 3.22 Under this modified system, the focus of the Central Bank's job would shift away from checking each invoice toward overseeing the operations of the commercial banks. The Central Bank would want to keep tabs on the overall process, as well as occasionally to "spot check" specific projects. If the Central Bank uncovers an irregularity in disbursements, then it could penalize the commercial bank. The idea is to transfer responsibility for applying the correct procedures onto the shoulders of the banking community. An effective penalty for abuse may be to prevent banks which disburse funds incorrectly from acting as agents in the conversion program, thereby denying them an opportunity to earn lucrative fees. 3.23 By "privatizing" the disbursement process, both the Central Bank and the commercial banks will be able to operate in the ways they know best. The Central Bank has more experience monitoring the activities of the commercial banking system than in supervising investment projects. Commercial banks, however, have daily experience handling project expenditures. Under the proposed system, the Central Bank would be able to operate within its own area of expertise by monitoring the commercial banks to ensure that disbursements are properly made. At the same time, commercial banks would function according to their strengths at the project level. 3.24 Finally, an effective monitoring program during the post- investment phase can help to reduce round-tripping. The Central Bank should consider expanding and standardizing its current reporting procedures, which should be required of all projects financed by a conversion. Reporting on the use of conversion proceeds should occur on a frequent basis (maybe quarterly), in order to uncover discrepancies before it becomes too late to correct them. From time to time, the Central Bank will have to verify the accuracy of reports with on-site inspections. A system of reasonable penalties needs to be developed and clearly stated in a revised circular. A special "penalty tax" may need to be levied against projects found to be in violation of conversion procedures.  _ 32 - IV. CONCLUSION 4.01 Debt equity swaps require current cash resources to retire the debt presented by the potential investors. It follows that the amount of debt that can be retired through such conversions will be limited by the debtor country's ability or willingness to commit the resources required to effect what amounts to a buy-back of debt. Depending on how the current resources are generated, there can be inflationary or interest rate pressure. A sustained and predictable program can, however, have a significant impact over time in reducing the stock of debt.1 4.02 Debt-equity conversion programs provide an investment incentive, in the form of an up front bonus equivalent to the difference between the secondary market price of debt and the redemption price in local currency actually received. Such up front investment incentives would be of particular interest to investors because they both reduce total expenditure on the investment and raise the financial rate of return. 4.03 Debt equity swaps are attractive to creditor banks because they increase demand for debt in the secondary market and make it easier for banks wishing to sell their debt to do so at more favorable prices. Even the banks who choose to hold their debt benefit from the higher market value of the debt and the greater likelihood the remaining debt will be serviced as total debt is reduced. Thus, the issue of establishing or resuming a conversion program will probably be an element in the negotiations with creditor commercial banks on debt restructuring and new money facilities. 4.04 The potential for debt equity conversions to induce inflation depends on the type of debt being converted. Conversions of debt owed by private borrowers, the national government, or public sector enterprises financed by their own resources is not likely to lead to an increase in aggregate demand unless overall credit to the economy is allowed to expand to finance these conversion transactions. However, when the Central Bank retires its debt by issuing cash to be used for investment, there is a potential inflationary impact. The analysis indicates that, in the case of the Philippine program using a realistic range of assumptions, for each $100 million equivalent of expenditures, $145 to $170 million in debt would be retired, requiring about pesos 1.8 billion ($90 million equivalent) in net expenditures in the first year. However, the net impact on expansion of reserve money will decline over time and will turn negative in the fifth year as the reduction in expenditures to service debt is greater than the expenditures for the swap program. 1/ Chile has retired over $5 billion of its debt since 1985 through its debt conversion programs including debt-to-debt conversions under Chapter 18. Brazil retired over $8 billion of debt during 1988 through debt conversions. About $2 billion of debt were retired under the conversion program and about $6 billion through informal conversions.  - 33 - 4.05 Debt equity swaps give rise to direct foreign investment without an actual inflow of additional foreign exchange, which reduces the availability of free foreign exchange in the short run. For each unit of $100 million per year for five years of debt reduction payments, the analysis shows the net impact of foregone DFI inflows and lower service payments will range from $71 to 93 million in lower reserves in the first year, but the negative impact declines and becomes positive in the fifth year. The total cumulative impact on the balance of payments would become positive after the sixth year if there is no additionality. If there is some additionality to total investment, the initial adverse effects would decline more quickly. 4.06 The mechanisms for implementing a program can have a great effect on the program's efficiency in meeting the government's objectives. A consideration to be weighed is the size of a bonus to give the investor; a larger bonus may encourage more additional investment, but will reduce the benefit to the government and the amount of debt reduction for a constrained volume of conversion resources. The best way to increase additionality would be to maintain a steady program of sufficient duration to outlast the necessary "incubation" period of project preparation and investor confidence. A program that switches on and off would not achieve this. In order for potential investors to undertake feasibility studies and do adequate project preparations, they need to know that the government has a long-term commitment to permitting debt-equity transactions. 4.07 When the government's resources for swaps are limited, conversions may be allocated by a government fiat. Under this method, a government receives a pool of applications and then evaluates each project against a set of general criteria. This is the system of Chile under Chapter 19 and that operated in the Philippines under Circular 1111. An alternative method for allocating conversions is to create a competitive market through an auction. Under this system, applicants may bid against each other for access to a fixed amount conversions. This would help deliver the largest possible portion of the secondary market discount to the government. Such a market-driven approach would be simpler to administer than government evaluation of individual projects. 4.08 There could be, however, certain trade-offs. Because the investor is unable to determine his actual capital costs until after the auction, auctions may introduce some uncertainty into the investment decision-making process. Generally, this uncertainty is viewed as a disincentive, particularly in the case of large new projects. Moreover, the case-by-case method enables the host country to exert a greater degree of influence over the direction of the investment. For example, in this approach, the government can encourage investment in certain sectors, ensure a higher level of technology transfer in appropriate cases, and insist that certain investments have a higher export orientation. However, the case by case, negotiated method is costly in terms of government manpower, delays, and the lack of transparency, regardless of how scrupulous the negotiations.  - 34 - 4.09 From the experiences of recent auctions in Brazil and Argentina, several observations can be made which may be useful for a debtor government to consider in designing its program. Using auctions does not necessarily preclude screening of conversion projects by the government. The government can screen out projects which may be considered non- desirable. Different auctions for investment of different priority can be established. In Brazil, the separate auction for "incentive projects" in less developed, priority regions in the northeast has channeled as much investment funds as the other auction for 'non-incentive projects' anywhere else in the country. Commitments by the government to the market system and its efficient administration are important for the success of auctions. The government's share was considerably higher under the auction system than under the negotiated method. 4.10 A debt equity conversion program can generally contribute to achieving several government objectives. However, depending on its structure, there can be conflicts among those objectives in that achieving one may affect achieving others. For example, if a debtor government wishes to achieve primarily debt reduction, its conversion program would be structured with a screening procedure that passes projects up to the limit of resources available. This would encourage a large volume of conversions for investment, additional or not, and provide a large share of secondary market discounts to the government. On the other hand, if the primary objective were to promote additional investment, and/or investment in priority sectors, it is important to keep the level of incentive higher and include criteria to screen out investments that are not additional. If a program were oriented strongly to additionality, it would run the risk that the conversion volume would remain small, that bonus and supervision costs would be high, and that it would have much less impact on debt reduction. 4.11 On the basis of an analysis of potential benefits as well as costs of conversions, and an assessment of the capacity of the economy to absorb the short-term macroeconomic impacts, decisions have to be made by the debtor government as to whether and to what extent it wants to allow the conversion of its debt paper. Also, a decision has to be made as to what should be the primary objective it wishes to achieve through conversions: debt reduction or investment additionality. Should the government decide to establish a program, it would need to formulate long-range operational targets by types of conversion and integrate their likely impacts into the overall macroeconomic planning. This would allow the required resources to be budgeted and the program implemented in a consistent manner.  ANNEX 1 ANALYSIS AND PROJECTIONS OF MACRO-ECONOMIC IMPACTS (A) Projections of Economic Impacts (B) Demonstration of Relation Among Variables - 35 -  - 36 - AMMX TABLE I a PHILIPPtNES: INPACT OF SWAPS ON ULANCE OF PAYMENTS DfI* S100 APT* S100 GovSh a 30% Add OS lipShu 30X (USS eillionr at Current Prices) 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 (1) Debt Retirement 143 143 143 143 143 0 0 0 0 0 0 0 (2) DFI tl1) 1 - (b)) 100 100 100 100 100 0 0 0 0 0 0 0 (3) Inftlow foregon (2) (1 - (f))] 100 100 100 100 100 0 0 0 0 0 0