Randall Dodd and Shari Spiegelunctad.org/en/docs/gdsmdpbg2420051_en.pdfIntroduction The massive...

UNITED NATIONS CONFERENCE ON TRADE AND DEVELOPMENT

G-24 Discussion Paper Series

UNITED NATIONS

Up From Sin:

A Portfolio Approach to Financial Salvation

Randall Dodd and Shari Spiegel

No. 34, January 2005

G-24 Discussion Paper Series

Research papers for the Intergovernmental Group of Twenty-Fouron International Monetary Affairs

UNITED NATIONSNew York and Geneva, January 2005

UNITED NATIONS CONFERENCEON TRADE AND DEVELOPMENT

INTERGOVERNMENTALGROUP OF TWENTY-FOUR

Note

Symbols of United Nations documents are composed of capitalletters combined with figures. Mention of such a symbol indicates areference to a United Nations document.

*

* *

The views expressed in this Series are those of the authors anddo not necessarily reflect the views of the UNCTAD secretariat. Thedesignations employed and the presentation of the material do notimply the expression of any opinion whatsoever on the part of theSecretariat of the United Nations concerning the legal status of anycountry, territory, city or area, or of its authorities, or concerning thedelimitation of its frontiers or boundaries.

*

* *

Material in this publication may be freely quoted; acknowl-edgement, however, is requested (including reference to the documentnumber). It would be appreciated if a copy of the publicationcontaining the quotation were sent to the Publications Assistant,Division on Globalization and Development Strategies, UNCTAD,Palais des Nations, CH-1211 Geneva 10.

UNITED NATIONS PUBLICATION

UNCTAD/GDS/MDPB/G24/2005/1

Copyright © United Nations, 2005All rights reserved

iiiUp From Sin: A Portfolio Approach to Financial Salvation

PREFACE

The G-24 Discussion Paper Series is a collection of research papers preparedunder the UNCTAD Project of Technical Support to the Intergovernmental Group ofTwenty-Four on International Monetary Affairs (G-24). The G-24 was established in1971 with a view to increasing the analytical capacity and the negotiating strength ofthe developing countries in discussions and negotiations in the international financialinstitutions. The G-24 is the only formal developing-country grouping within the IMFand the World Bank. Its meetings are open to all developing countries.

The G-24 Project, which is administered by UNCTAD�s Division on Globalizationand Development Strategies, aims at enhancing the understanding of policy makers indeveloping countries of the complex issues in the international monetary and financialsystem, and at raising awareness outside developing countries of the need to introducea development dimension into the discussion of international financial and institutionalreform.

The research papers are discussed among experts and policy makers at the meetingsof the G-24 Technical Group, and provide inputs to the meetings of the G-24 Ministersand Deputies in their preparations for negotiations and discussions in the framework ofthe IMF�s International Monetary and Financial Committee (formerly Interim Committee)and the Joint IMF/IBRD Development Committee, as well as in other forums.

The Project of Technical Support to the G-24 receives generous financial supportfrom the International Development Research Centre of Canada and contributions fromthe countries participating in the meetings of the G-24.

UP FROM SIN: A PORTFOLIO APPROACHTO FINANCIAL SALVATION

Randall DoddDirector

Financial Policy Forum

Shari SpiegelExecutive Director

Initiative for Policy Dialogue

G-24 Discussion Paper No. 34

January 2005

viiUp From Sin: A Portfolio Approach to Financial Salvation

Abstract

In this paper, we develop a proposal with the potential to greatly improvethe ability of developing countries to reduce their exposure to other countries�interest rate and exchange rate volatility and to lower their cost of raising capitalabroad. The key to achieving these goals is for developing countries to borrowin their own currencies and for investors to lend by creating portfolios of local-currency government debt securities that employ the risk management techniqueof diversification to generate a return-to-risk that competes favourably with othermajor capital market security indices. We show, based on data from the early1990s, that a portfolio of emerging market local currency debt can generaterates of return relative to risk that compete with those of major securities indicesin international capital markets. It bears noting that the early 1990s witnessedseveral severe shocks to international capital markets, including the crises inEast Asia, the Russian Federation and Brazil, and the failure of Long-TermCapital Management. We also analyse the implications of deploying such a policyfor attracting capital to developing countries, the impact on the stability of theirfinancial systems and on their costs of borrowing, and the implications for futuredevelopment of local capital markets.

ixUp From Sin: A Portfolio Approach to Financial Salvation

Table of contents

Page

Preface ...................................................................................................................................................... iii

Abstract ................................................................................................................................................... vii

Introduction ............................................................................................................................................. 1

Overview of the issue ............................................................................................................................... 2Market description: foreign currency claims ................................................................................... 2Consequences of foreign currency borrowing ................................................................................. 2Recognition of a problem ................................................................................................................. 4Contemporary insights into foreign exchange risk .......................................................................... 5

Salvation through diversification: a policy remedy ............................................................................. 6The portfolio approach ..................................................................................................................... 6Principles of financial economics applied to constructing an LCD portfolio .................................. 9

Track record ........................................................................................................................................... 11Experiments with investment in local currency debt ..................................................................... 11LCD track record............................................................................................................................ 11

Why the private sector has been slow to implement this strategy ................................................... 13

Alternative ownership structures ........................................................................................................ 14

Economic and development consequences .......................................................................................... 15

Conclusion .............................................................................................................................................. 16

Appendix A .............................................................................................................................................. 17

Appendix B .............................................................................................................................................. 18

Notes ........................................................................................................................................................ 18

References ................................................................................................................................................ 19

List of tables

1 Average correlation coefficients for 47 developing economy exchange rates, 1980�2004 ............ 72 Debt ratings: long-term maturities ................................................................................................... 83 Average correlation coefficients for ELMI+/EMBI+ ...................................................................... 8

List of charts

1 Risk-return profile, 1994�2003...................................................................................................... 122 Risk-return profile, 1993�2004...................................................................................................... 12

Introduction

The massive amount of foreign indebtednessis one of the most significant problems facing de-veloping countries today. The majority of this debtis denominated in the major currencies,1 and it hasleft developing countries with enormous exposuresto foreign exchange and foreign interest rate risk.External shocks transmitted through these exposureshave proven costly to absorb. Most of the emergingmarket crises over the past two decades were caused,or at least exacerbated, by foreign borrowings (al-though derivatives have added substantially to foreign-currency-denominated liabilities in some cases2).Concerns over foreign debt were raised soon afterthe 1973 oil price shock, and the problem becameapparent in the early 1980s after Mexico announcedit would be unable to meet its debt payments in Au-gust of 1982. More recently, outstanding dollarlinked bonds were one of the main triggers of the1994 Mexican crisis; dollar denominated private debtwas one of the factors in the 1997 Asian crisis; and

Argentina�s large issues of external debt was a pri-mary factor in its 2001 default.

Most of the economic policy research since thedeveloping country debt crisis broke out in 1982 hasfocused on solving the problem of foreign investorsmanaging their credit risk exposure and the problemof developing countries adjusting to the variabilityof foreign capital flows and financial crises. Lessattention was given to strategies to help developingcountries reduce their foreign exchange exposure.More recently, the issue of foreign exchange riskarising from foreign indebtedness has received greateras a result of a new body of economic policy researcharound the �original sin� hypothesis of Eichengreenand Hausmann (1999). There is now some newthinking about the challenge faced by developingcountries in moving away from foreign-currency-denominated borrowing.

In this paper, we take up that challenge by build-ing on the early experiments3 and providing a policy

UP FROM SIN: A PORTFOLIO APPROACH TOFINANCIAL SALVATION

Randall Dodd and Shari Spiegel*

* This paper has been prepared with financial support from the Ford Foundation and the International Development ResearchCentre (IDRC) of Canada. The authors thank Stephany Griffith Jones for comments on an early draft, Ira Handler for help indeveloping the EM LCD local currency portfolio and Claire Husson for research assistance.

2 G-24 Discussion Paper Series, No. 34

analysis of what we hope will prove to be a majornew financial innovation to financing development.It improves market efficiency by overcoming someexisting market imperfections. It requires negligi-ble statutory or regulatory changes. And it does notnecessarily require the involvement of such interna-tional financial institutions as the IMF or WorldBank, although it may well benefit from some offi-cial or public sector sponsorship.

Our proposal is to raise capital in internationalmarkets by forming diversified portfolios of emerg-ing market local currency debt (LCD) issued bysovereign governments. We show that the returnson such a portfolio are sufficiently independent toallow a substantial reduction in portfolio variancethrough diversification, to produce a risk-return pro-file � measured in dollars � that is competitive withmajor United States and European security indices.

In the following section we will discuss thecontext of this problem and the economic literaturethat addresses it. Following that, we develop the fi-nancial economic basis for the LCD portfolio andanalyse its financial performance relative to somefamiliar alternative investments. We then analyse thedevelopment and economics implications of this newfinancing facility for developing economies and of-fer concluding comments.

Overview of the issue

One key difference between advanced and de-veloping economies4 is that the latter generallycannot borrow in international capital markets intheir own currency.5 While other distinguishing fea-tures include the capital-labour ratio, productivity,education levels, and sophisticated financial systems,the inability of developing countries to borrow in-ternationally in their own local currency is a criticalelement affecting their financial stability.

Many developing countries need capital inflowsto augment domestic savings so as to obtain a paceof investment consistent with rapidly rising growth.In addition to foreign direct investment, developingcountries need capital investment in the form ofcredit. If they are unable to negotiate terms denomi-nated in their own local currency, then they mustenter into a Faustian compact in which the joys of

lower interest rates are held captive by the obliga-tion to repay in foreign currency denominations.

Market description: foreign currency claims

While foreign debt can play a useful economicrole in development by supplementing domestic sav-ings, its currency denomination can create unwantedexposure to exchange rate risk. Consider the break-down of developing country long-term external debt.The latest year for which data are available on cur-rency composition is 2000 (from the World Bank�sGlobal Development Finance database). The totallong-term external debt that year was $2,047.7 bil-lion (it would rise to $2,644 billion through the endof 2003). Of that, 64 per cent was denominated indollars, 12 per cent in yen, and 9.5 per cent is inDeutsche mark, French francs and pound sterling.The six currencies comprise at least 85.5 per cent ofthe long-term indebtedness of developing countries;another 7.6 per cent was debt in �multiple curren-cies� (these six currency likely make up a large shareof this figure) and 7.2 per cent in all other curren-cies.6

Two particular examples, drawn from Mexicoand the Republic of Korea just prior to their finan-cial crises, illustrate the predominance of majorcurrencies, and the lack of their own local currencies,in denominating their foreign debts. In Decemberof 1994, 61 per cent of Mexico�s long-term foreigndebt was denominated in dollars, 21 per cent in �mul-tiple currency� formulas, and 9 per cent in yen.Measured at year-end 1996 � the last data point priorto the financial crisis � the dollar composition oflong-term debt of the Republic of Korea was 79 percent, while that in yen and multiple currencies stoodat 13 per cent and 5 per cent respectively. Thus,92 per cent of the foreign debt of the Republic ofKorea was in two major currencies: the dollar andthe yen.

Consequences of foreign currency borrowing

The consequences of foreign currency denomi-nated indebtedness are a major source of developingcountry�s exposure to international disruptions anddisturbances, as well as their vulnerability to domes-tic fiscal solvencies and exchange rate systems. This

3Up From Sin: A Portfolio Approach to Financial Salvation

the more fragile economies of developing countriessubject to monetary policy changes in the country(or countries) whose currency denominates their in-ternational debt. If foreign central banks tightenmonetary policy and raise interest rates, this willincreases foreign currency interest payments for thedeveloping countries on both variable rate debt andnew issuances while they will also see the marketvalue of their foreign currency assets reduced. Forexample, in the 10 months prior to the financial cri-sis in Mexico in December 1994, the Federal Reserveraised short-term interest rates six times, from 3 percent to 5.5 per cent and the dollar rose by more than7 per cent against other Group of Ten (G-10) cur-rencies. This put enormous external pressure, on thepeso while diminishing the Mexican central bank�scapacity to defend it. Similarly, it was the rapid risein United States interest rates in response to the oilprice hike in the late 1970s that precipitated the LatinAmerican debt crises of the 1980s.7

An appreciation in the value of the foreign cur-rency of denomination would also raise the cost ofservicing foreign debt and put pressure on the inter-national value of the local currency. In the monthspreceding the first East Asian crisis, the dollar ap-preciated substantially against the Japanese yen.Starting in August 1996, the dollar rose 8.2 per centby January 1997; it was up 19 per cent by May 1997when the Thai baht was hit by speculative attacks;by December 1997, the dollar had risen 25.5 per centover the previous 14 months. This had the effect ofsubstantially raising the value of Thailand�s dollardenominated debt, and in comparison, lowering thevalue of their yen-denominated exports.

The accumulation of international debt denomi-nated in foreign currency poses a danger to morefragile financial systems, which often have dollar-denominated liabilities and local currency assets.Borrowing in foreign currency causes a �currencymismatch� that exposes the developing country todevaluation or some other international disruptions;and borrowing short-term causes �maturity mis-match� that exposes the country to other volatileeconomic sources that contribute to creating a frag-ile financial system. It creates a economic precipiceby dramatically raising the cost of devaluation, andit makes policymakers more reluctant to let the cur-rency devalue even when it becomes overvalued.8

Foreign-currency borrowing also reduces acountry�s ability to pursue independent monetary and

fiscal policy. The country�s monetary policy is con-strained through its impact on exchange rates and,in turn, its impact on the local currency cost of serv-icing foreign debt. Thus, a central bank that wouldotherwise respond to a contractionary shock by eas-ing credit conditions would be hampered from doingso for fear of the reducing its currency value andthereby raising the cost of servicing its foreign debt.

Foreign-currency borrowing also underminesthe credibility of central banks in developing coun-tries because their foreign reserves, which mightamount to a significant proportion of their importsand trade balances, can be dwarfed by the country�sforeign-currency debt obligations. Foreign-currencyindebtedness therefore necessitates increased hold-ings of foreign reserve, which can be quite expensive.9

The ability to engage in fiscal policy of for-eign-currency borrowers is also affected. It is con-strained because any increase in the fiscal deficit willwiden the country�s credit spread and further pushup the cost of borrowing. High foreign indebtednessforces countries to follow pro-cyclical macroeco-nomic policies, raising interest rates and tighteningfiscal policy during a recession.10 Foreign-currencyborrowing also undermines the fiscal credibility ofdeveloping countries because so much of their sov-ereign debt is known to be denominated in foreigncurrency.

On the other hand, borrowing in local currencyadds to the tax base for seigniorage revenue in thehome country owing to the increase in demand forthe local currency for the purpose of trading in thelocal currency instrument, as well as in the processof making and receiving payments on those securi-ties.

Yet an additional benefit is that the develop-ment and growth in the market for the local currencygovernment securities plays a critical role � as a foun-dation or skeletal structure � in promoting more�mature� domestic securities markets.

The danger from amassing foreign currencydebt is so great that it calls into doubt the efficiencyof international capital markets in distributing risk.Why is the foreign exchange risk so dispropor-tionally held by those least able to bear it? If thisdistribution of risk signals a market imperfection,then it highlights the need for an innovation or newpublic policy to rectify it.


Recognition of a problem

The recognition of the problem of borrowingin foreign currency is not new. Soon after develop-ing countries� indebtedness began to rise rapidlyfollowing the 1973 oil price hike, and well beforeMexico�s payment crisis erupted in August 1982,there were public policy discussions of this concern.One notable contribution to the debate came fromGerald Pollack (1974), an Exxon Corporation ex-ecutive writing in Foreign Affairs, who offered anearly warning of the consequences of large, dollar-denominated borrowing by developing countries. Hestated, �the Eurocurrency market has several defectsfor present purposes� and particularly �the Euro-currency market is not well suited to resource-pooror politically unstable developing countries with lowcredit standing.� He warned, �[t]his question of fi-nancial instability may turn out to be the biggest ofthe threats posed by the energy crisis.�

Commenting upon the build-up of large foreigndebts after the first and second oil price hike, WalterJ. Levy (1980) stated, �The debt problem could, ofcourse, be solved if the values of the currencies inwhich the debts are incurred decline ... [which] infact did occur between 1974 and 1978 ... But it seemsnow that the jig is up.�

Writing just prior to the 1982 payments crisis,Bacha and Diaz-Alejandro (1982) expressed con-cerns about prospective borrowing conditions in the1980s that were �moderately pessimistic relative torepeating the favourable performance of the 1970s.�With regard to dollar-denominated debt, they stated,�A major uncertainty for LDC borrowers looking atthe 1980s is whether the low or even negative realrates of interest prevailing during the 1970s will re-turn.� In particular they cited the risk of tighteningmonetary policy in the United States as a factor lead-ing to higher interest rates and a higher dollar.

Much of the academic economics literature isfocused on the issue of the credit risk faced by theinternational banks acting as intermediaries in thisrecycling process. Two noteworthy contributionscame from Laurie Goodman (1980 and 1981). In a1980 study for the New York Federal Reserve Bank,she provided an exemplary descriptive analysis ofthe financial instruments used for recycling petro-dollars, and she then analysed the pricing of thesesyndicated, dollar-denominated, variable-rate bankloans. The principal concern at the time was whether

the loans, which were priced as a spread aboveLIBOR, generated sufficient returns to banks� port-folios in proportion to their credit risk for the banks.

In a follow-up study, Goodman (1981) analysedthe possible problems of credit risk from the pointof view of lenders. She empirically tested a modelof diversification to show that banks could gain bydiversifying their credit risks across countries. Al-though only limited data were available at that time,Goodman showed that the country risks associatedwith lending to various developing countries weremore different than similar � that is, the unique riskis greater than the common or systematic risk. Sinceit is the unique risk that can be reduced through di-versification, Goodman suggested that internationalbanks could successfully manage the risk on theirloans to developing countries through diversifica-tion.

One of the first articles in the economics lit-erature to discuss the market risk associated withforeign-currency borrowing (as opposed to creditrisk) in the context of international lending to de-veloping countries was written by Lessard (1983).Writing in the wake of Mexico�s debt crisis in Au-gust of 1982, Lessard pointed out that the form ofdebt issued to developing countries was �unsound�,and that criticism had been too focused on the quan-tity of foreign borrowing and not sufficient attentionwas being paid to the fact that �[foreign lending] isstructurally unsound and is likely to result in mis-judgments and misbehaviour on the part of lendersand borrowers.� He went on to clarify what he meantby structurally unsound, stating that �a financial sys-tem that relies overwhelmingly on bank credit isunlikely to be an ideal system in terms of worldwelfare ... it involves debt service patterns that varyperversely with LDCs� net foreign exchange earn-ings ... [and] it shifts risks from LDCs to world capitalmarkets only through default [i.e., bankruptcy].�

While drawing needed attention to key featuresof foreign debt, such as the problems caused by theirvariable interest rate structure, Lessard failed tostrictly identify foreign exchange risk as a majorsource of the �perverse variability of debt serviceobligations�. Perhaps he took for granted that for-eign debt would be denominated in foreign currency.Lessard did, however, foresee the more recent trendin development finance research, discussed below,by recommending that debt service be stabilizedthrough the use of price level index-linked loans.


Contemporary insights into foreignexchange risk

Whereas the problems associated with devel-oping countries� exposure to foreign exchange riskhave been recognized by some people for quiteawhile, they have attracted greater attention in re-cent years. This can be attributed to the research andprovocative title � the �original sin� hypothesis11 �developed by Hausmann, Eichengreen and others.Their work in Hausmann et al. (2001 and 2002) andEichengreen et al. (2002 and 2003) has made a sig-nificant contribution to the study of developmenteconomics and sparked a much needed debate onexposure to foreign exchange risk. It has also con-tributed to the policies designed to raise livingstandards in the developing world by focusing at-tention squarely on the risks associated with theinvestment vehicles that conducted capital to devel-oping countries. (A summary of their EmergingMarket Index, as well as a discussion of its short-comings, are included below in appendix A.)

Simply put, Hausmann and his colleagues ar-gue that the principal problem facing developmentfinance is that developing countries cannot borrowin their own currency. Instead they have borrowedin �hard� currencies such as the dollar, euro (and itpredecessors), sterling, and yen and this has exposedthem to foreign exchange rate fluctuations not en-tirely within their control.12 Had these countries beenable to borrow in their own currency, Hausmann etal. conclude, they would have been better able tohandle shocks and other policy errors.

Dodd (2001) identifies currency mismatches asa major source of vulnerability affecting financialsector stability in developing countries and analy-ses the role of derivatives in increasing the magnitudeof the mismatch. Developing country financial in-stitutions were able to resort to derivatives, oftenunregulated or under-regulated as off-balance sheetitems, to achieve greater leverage and circumventrestrictions on their balance sheet�s currency mis-matches. The result was that they did not hedge theirlong dollar and short local currency mismatch, butrather took larger short dollar positions in order tocapture the gains from the substantial interest ratedifferential between the dollar and their local currency.Dodd pointed out the difficulty for governments, aswell as private sector investors, in monitoring thevolume of open interest and trading volume in a

market that has no regulation to require reportingand disclosure.

Expanding on a problem they identified in1996,13 Goldstein and Tucker (2004) provide a com-prehensive policy analysis that addresses the issueof currency risk head on, that is, the exposure toforeign-currency mismatches leaves developingcountries vulnerable to financial shocks, deepens theimpact from the shock, and hampers the use of mon-etary policy in preventing a contraction or crisis. TheGoldstein and Tucker thesis � although largely dis-tracted by its authors� effort to refute the �original sin�hypothesis that bad things happen to good countries� does provide a useful explanation for why cur-rency mismatches are important and how to bestmeasure them. Unfortunately the book�s policy rec-ommendations are little different from the adviceoffered before these insights were established. In-stead of trying to address the problem directly, theauthors advocate inflation targeting, floating insteadof fixed exchange rates regimes, and fiscal rectitudein government budgets.

Allen et al. (2002), of the IMF�s Policy Devel-opment and Review Department provides anauthoritative and comprehensive analysis of the bal-ance sheet aspects of developing country financialcrises. They identify the major types of risk that char-acterize a nation�s balance sheet vulnerability andsite foreign currency mismatch as a major factor,saying that �Almost all recent crisis episodes weremarked by currency mismatch exposures.� One oftheir key conclusions is that �The currency and ma-turity structure of the outstanding debt stock is almostas important as the total size of the debt stock.� An-other is that �A currency mismatch anywhere in theeconomy constrains the government�s capacity to actas a lender of last resort in domestic currency.�

The policy recommendations of Allen et al.include the need to develop local currency equityand debt (especially long-term debt) markets inorder to raise capital while limiting financial vul-nerability. This includes developing derivativesmarkets � although they recommend that �attentionmust be paid to the risks incurred by those who aresupplying the hedging instruments� (Allen, 2002).Allen et al. also cite the need to hold larger foreignreserves and the need for great external official lend-ing (i.e., IMF, World Bank, and bilateral lending).

However, as discussed earlier, holding foreignreserves can be expensive for developing countries.


Furthermore, additional external borrowing in for-eign currencies � whether it be by the IMF, WorldBank, or the private sector � will only to exacerbatethe problem of currency mismatch associated withexternal debt. The international financial institutionsneed to look instead towards the more radical ap-proach of lending in local currencies. In the followingsections of this paper we will analyse this possibilityin more detail, and show why lending in local cur-rencies is not a high risk strategy for foreign creditors.

Salvation through diversification:a policy remedy

We believe that there is a better way to financedevelopment. In this section, we develop a policyremedy that is both economically more efficient and,if politics is truly �the art of the possible�, morepolitically feasible. One important testament to itsviability is that, as described below, it has alreadybeen successfully pursued by at least one major as-set manager.

We will call this approach the �emerging mar-ket local currency debt portfolio� (or LCD). Theoriginal idea for this portfolio-based approach drawsheavily on the work of Shari Spiegel, based on herexperience in creating and managing a diversified,local currency developing country debt portfolio forLazard Freres (from May 1995 to January 2003).14

The fund was operated with the goal of capturinghigh rates of return paid on local currency securitieswhile reducing risk through diversification.

The LCD proposal does not require that inter-national financial institutions such as the IMF orWorld Bank play a pivotal role in its success. Theseofficial financial institutions could, however, makehighly productive contributions (see below).

Although the LCD policy is low cost and fea-sible, its potential benefits are substantial. It offersdirect help to countries to enable them to borrow intheir own currencies � both at home and abroad � inwhat could become a seamless market. The gainsfrom this alone are potentially enormous. Develop-ing economies will benefit also from greater stabilityas a result of their reduced exposure to changes inforeign exchange rates and interest rates, which areall the greater because of their correlation, in the

dollar and other major currencies. They will also gainfrom increased seigniorage and from potentiallylower costs of borrowing in their local currency.

Staying with the theme of Christian theologyinspired by Hausmann�s work, the LCD portfoliowill lead to � if not outright redemption of �originalsin� � absolution of worldly sins so as to facilitateprogressive steps toward financial salvation.

The portfolio approach

The core idea of the LCD approach is to applythe insights of portfolio theory,15 part of the disci-pline of financial economics � toward enablingdeveloping countries to borrow in their own curren-cies. The insight offered by portfolio theory is that aportfolio consisting of different securities whose re-turns are sufficiently independent (and especially soif they are negatively correlated) can yield superiorrisk-adjusted rates of returns than the individual se-curities. In other words, the volatility of the wholeis less than that of the sum of its parts.

The LCD portfolio would work by buying lo-cal currency government debt16 instruments frommany different developing countries and combiningthem so as to produce a portfolio whose return andvariance would be competitive in international capi-tal markets. The market risk, which consists of theuncertainty of domestic interest rates (i.e., interestrates in local currency assets) and exchange rates ofeach local currency security, is often significant.From 1994 to 2003, the average volatility of individualcountry returns on local currency debt instrumentswas nearly 16 per cent.17 At the same time, yields onlocal currency debt were also high, at 13.7 per centon average, but not high enough to compensate forthe risk. Hence investing in any one local currencymarket was not attractive.

Combining the returns on individual countrysecurities into a portfolio, however, does producedesirable results. As we will show below, returns ona diversified portfolio range from 8�10 per cent an-nually while the risk of a diversified portfolio dropssubstantially to approximately 5.5 per cent (whichis in line with United States investment grade bonds).

Note that this approach does not involve hedg-ing the currency risk. It is very expensive to reduce


this market risk by hedging with derivatives becausethe cost of hedging is equivalent to the differentialbetween foreign and local interest rates and, as dis-cussed above, local interest rates tend to be very high.Another reason why hedging costs may be expen-sive is that there are a disproportionate amount ofshort-hedgers in the market and a relative shortageof speculators willing to speculate on long-positionsin local currencies.18

In other words, the costs of the hedge over-whelm the benefit of cross-border borrowing orinvesting. In this context, the most cost-effectivemethod for mitigating risk exposure is achievedthrough diversification across different countries�local currency debt. It is the unique aspect of thisrisk that can be substantially reduced through diver-sification. The reason for this is that, historically,currency devaluations have had extremely low cor-relations.

We began our analysis by examining the ex-change rates of 46 developing countries, (leavingout Latvia from above). The countries were selectedprimarily on the basis of the availability monthlydata since 1980; we then eliminated some countriesthat had extraordinarily stable exchange rates suchas Saudi Arabia and Oman and we dropped Zimba-bwe for the opposite reason. The annual rate ofchange in all of these countries� exchange rates be-tween January 1995 and March 2004 was 9.8 percent; and of the 37 countries that had data doingback to January 1990, the average rate of changewas 10.2 per cent over the longer 14-year period(with highs reaching -78.4 per cent and -43 per centannually for Brazil and Romania respectively).

We than analysed the monthly rates of changefor 47 developing country currencies from January1980 to March 2004. The average correlation be-tween the rates of change for those 47 countriesduring that time period was 0.0713 � in other words,not a high degree of correlation on average. Theaverage correlation coefficients for the individualcountries are listed in table 1.

Diversification is also a means of reducingcredit risk. For most local currency securities, how-ever, credit risk is not the primary risk. Most devel-oping countries are more highly rated for debtobligations in their own currencies than in foreigncurrencies. Table 2 shows several recent examples �Brazil is the only exception � of how local currency

Table 1

AVERAGE CORRELATION COEFFICIENTSFOR 47 DEVELOPING ECONOMY

EXCHANGE RATES, 1980�2004

Argentina 0.013Bangladesh 0.040Botswana 0.123Brazil 0.059Bulgaria 0.025Chile 0.033Colombia 0.040Côte d�Ivoire 0.087Croatia 0.150Czech Republic 0.180Ecuador 0.022Egypt -0.012Estonia 0.217Ghana 0.042Hungary 0.153India 0.063Indonesia 0.038Israel 0.085Jamaica 0.027Jordan 0.097Kenya 0.077Latvia 0.078Lebanon 0.003Lithuania 0.060Malaysia 0.073Mauritius 0.144Mexico -0.020Morocco 0.190Namibia 0.109Nigeria 0.022Pakistan 0.061Peru 0.042Philippines 0.045Poland 0.082Republic of Korea 0.028Romania 0.041Russian Federation 0.021Slovakia 0.205Slovenia 0.206South Africa 0.110Sri Lanka -0.017Thailand 0.086Trinidad and Tobago 0.008Tunisia 0.169Turkey 0.054Ukraine 0.006Venezuela -0.014

Data: IMF, International Financial Statistics.


credit ratings are usually two notches above that onforeign currency debts. With the exception of theRussian Federation, default on local-currency gov-ernment debt has been extremely rare. The primaryrisk to investing in local-currency government debtis the domestic market risk (interest rate and ex-change rate uncertainty).

From an investor�s perspective, credit risk isthe primary risk on external debt, such as long-termEurobonds, issued by developing countries. Considerthe comparison of credit risk on dollar-denominatedsecurities versus market risk on local currency de-nominated securities. These are captured by theaverage correlation coefficients for the EmergingLocal Markets Index (ELMI)+ and the Emerging Mar-kets Bond Index (EMBI)+, calculated by JP Morgan.The ELMI+ index measures the rates of return(change in price plus interest payments) on local-currency debt securities in each of the 25 listedcountries. The EMBI+ index measures the rate ofreturn on United States dollar denominated debt secu-rities in the 20 countries listed. There are 13 countriesfor which there are both ELMI+ and EMBI+ data.19

The correlation coefficients are presented intable 3. Some of the correlation coefficients have

high p-values which suggests that there are some pe-riods of time in which the correlations are higher (aswell as lower) than the coefficients would indicate.

The correlation coefficients under the columnELMI+ measure the average correlation between thecountry (in the column to the immediate left) andthe other 24 counties for which there are data (empty

Table 3

AVERAGE CORRELATION COEFFICIENTSFOR ELMI+/EMBI+

(May 1993�May 2004, monthly)

ELMI+ EMBI+ Difference

Argentina -0.011 0.396 0.407Brazil 0.115 0.495 0.380Bulgaria 0.469Chile 0.165China 0.129Colombia 0.059 0.419 0.360Czech Republic 0.103Ecuador 0.454Egypt 0.044 0.487 0.444Hong Kong, China 0.110Hungary 0.092India 0.170Indonesia 0.131Israel 0.002Malaysia 0.151 0.390 0.239Mexico 0.135 0.553 0.418Morocco 0.533Nigeria 0.525Panama 0.539Peru 0.533Philippines 0.133 0.466 0.333Poland 0.174 0.514 0.339Republic of Korea 0.159 0.405 0.246Russian Federation 0.202 0.431 0.228Singapore 0.187Slovakia 0.078South Africa 0.065 0.126 0.061Taiwan Prov. of China 0.186Thailand 0.230Turkey 0.074 0.334 0.260Ukraine 0.360Venezuela 0.121 0.465 0.344Average 0.120 0.445 0.312

Note: ELMI+ adjusted for euro/dollar rates for CentralEurope.

Table 2

DEBT RATINGS: LONG-TERM MATURITIES

Foreign Localcurrency currency

Brazil B+ B+Chile A- A+Colombia BB BBB-Costa Rica BB BB+Egypt BB+ BBBEstonia A- A+Malaysia BBB+ AMexico BBB- BBBMozambique B B+Peru BB- BB+Philippines BB BB+Poland BBB+ A+Republic of Korea A AA-South Africa BBB A-Thailand BBB A-

Data: Fitch Ratings, February 2004.


rows indicate there are no data for that country ineither ELMI+ or EMBI+). The average correlationis pretty low and is negative for Argentina.

The coefficients under the EMBI+ column arethe same measures but for dollar denominatedforeign debt. Note that they are, on average, sub-stantially higher than for local currency debtsecurities. For the 13 countries for which there arecommon data the average correlation coefficient is0.312 higher for the credit risk reflected in theEMBI+ than for market risk reflected in the ELMI+series.

The financial economic lesson to be drawn fromthis comparison is that the potential reductions inmarket risk through diversification are greater thanthe potential reductions in credit risk.

Principles of financial economics applied toconstructing an LCD portfolio

Given these properties of exchange rates andmarket returns, a portfolio of local currency debtsecurities can be constructed so as to provide for-eign investors with an attractive investment vehicle.

Four interrelated decisions must be made inchoosing the securities to be included in the portfo-lio: the number (n) of securities; which countries toinclude among the n in the portfolio; which securi-ties in each country to include; and the weight ofeach security in the portfolio.

The decision to choose certain securities forinclusion in the portfolio will depend upon their re-turns, the distribution of those returns, and thecorrelation with other securities in the portfolio. Ifthe distributions of returns on the n securities havesufficient independence, then their combined yieldand variance will produce a portfolio return and vari-ance that compete with benchmark fixed incomeportfolios from the advanced capital markets in themajor currency economies � e.g., United States cor-porate bonds � and have substantially less volatilitythan such equity indices as the Standard and Poor�s(S&P) 500, DAX, the Financial Times Stock Ex-change (FTSE) 100, and the Nikkei 225 stock indexes.

An expression for the country returns � com-prised of changes in the exchange rate, interest

payments, and the change in price of the local debtinstrument � can be derived from the following equa-tion for interest rate parity, where r is the dollar ormajor currency rate of return (coupon payments pluschange in market price), r* is the local current rateof return, and e is the exchange rate in consecutiveperiods.20

(1)

Equation (2) represents the portfolio rate ofreturn (rp) from period t to period t+1 from invest-ing in n countries� local currency securities, whereeach country is weighted by a factor x in the portfolio.

(2)

The role of diversification in forming the port-folio is critical. The LCD portfolio would consist ofa sufficient number of different securities from dif-ferent countries so as to reach acceptable levels ofmarket and credit risk.

The variance of the portfolio (σp) is determinedaccording to the following equation (3), where x cor-responds to the weights of the securities in the port-folio, ρ is the correlation coefficient, σ is the standarddeviation, the subscript p denotes portfolio, and thesecurities in the portfolio are represented by i and j.

(3)

The main idea is to combine securities withdifferent distributions so that the variance of thecombination is potentially less than that of each ofthe individual securities. If the securities have a nega-tive correlation, then the variance of the combination� that is, the portfolio � can be very small. If theyare not negatively correlated but nonetheless havelow correlations, then the portfolio variance can stillbe greatly reduced, so that the variance of the wholeis less than that of the sum of the parts.

Up to now, we have assumed that market riskincludes both interest rate and currency risk. Thetwo can, however, be unbundled and analysed sepa-rately which is crucial to the decision of whichmaturities to include in the portfolio. The data arelimited because in many countries there are no ro-

]/)1[(1 1−∗+=+ tt eerr

1/)1( 11

−+= −∗

=∑ tti

n

iip eerxr

ijji

n

ij

n

jip xx ρσσσ ∑ ∑

= =

=1 1

2


bust data available on �long-term� interest rates.Experience in trading these securities indicates thatcorrelations are low across local interest rate mar-kets as well as currency markets. However, there isa correlation between interest rates and the currencywithin a country that is taken into account in the fol-lowing formula.

The additional risk added to a portfolio (σD)from extending the maturity or duration21 of invest-ments in the ith local currency securities can bemeasured by the following formula:

σDi = (w2σC2 + D2σr

2 + 2wρDσcσr)½ (4)

where w is the country weight, D is the contributionto duration, ρ is the correlation coefficient, and σcand σr are the standard deviations of the exchangerate and interest rate.

The decision of the optimal number of securi-ties to include in a portfolio depends on the marginalbenefit of the nth security to the portfolio variance.There are perhaps 40 or more countries whose localcurrency securities are suitable for inclusion in anLCD portfolio. Both theoretical reasoning and em-pirical testing support the case that the marginalbenefit to diversification declines as the number ofsecurities is increased. The rate at which the benefitsdiminish will depend on the degree of independenceof the returns so that a portfolio of completely inde-pendent securities will quickly overwhelm the benefitsof diversification by eliminating all �unique� riskand leave only �systematic� or common risk. In actualmarkets there are degrees of independence and in-terdependence. As the marginal benefit diminishes,the transactions costs and portfolio managementcosts rise with the number of securities.

Empirical studies of the United States equitymarkets, as discussed in Sharpe (1970), estimate thata portfolio with 10 securities will have 7 per centmore risk than the minimum (i.e., that of the marketportfolio without diversifiable risk) and a portfolioof 20 securities will have only 3 per cent more thanthe minimum. The actual determination of the optimalnumber of securities will depend on the correlationsand the transactions costs required to obtain andmanage such a portfolio.

Our preliminary analysis of data from emerg-ing markets in the 1990s indicates that the benefits

to diversification flatten off after the nineteenth se-curity. Thus, a portfolio of 20 securities would serveas a good estimate of the minimum number of securi-ties necessary for diversification to be fully effective.

The decision on the weighting of securities inthe portfolio can be driven by different motivations.One approach would be to treat the portfolio as if itwere an index in which country weights replicatedthe share of the respective economy or the marketsize of the security�s issuance relative to the othersrepresented in the portfolio.22 Most, but not all, se-curity indices in advanced capital markets are basedon weights that reflect the market capitalization ofthe security. However, the size of local-currency debtmarkets is generally quite difficult to measure. Mostsecurities are traded over the counter and includebank paper and even derivative products that are noteasy to quantify. In addition, government securitiesmarkets can be extremely illiquid, with a proportionof the securities held by the central bank or the Treas-ury, so that the number of outstanding issuances isnot a good indicator of the tradable size of the market.

Another approach, followed by the JP MorganELMI/ELMI+, is to weight countries by the size oftheir foreign exchange markets, as measured by to-tal exports and imports. This approach avoids theproblems in trying to estimate the size of the localcurrency markets. But it ignores the fact that liquid-ity in the local securities market has very little to dowith the size of foreign exchange trading. TheELMI+ attempts to address this by including meas-ures of liquidity in the index.

One of the results of this formulation is thatcountries in Asia represent nearly 40 per cent of theindex, since they have the largest share of trade ofcountries in the investment universe. Yet, from apolicy perspective, countries in Asia have the leastneed for foreign capital because of their high sav-ings rates and because they have the least amount ofsovereign debt denominated in foreign currency. Thiswould reduce the effect of the proposal in solvingthe problem of �original sin�. From an investmentperspective, yields in most countries in Asia tend tobe low. Asian currencies also tend to be highly cor-related owing to their high regional trade links. Thus,a portfolio based on this approach will tend to be ahigher risk/lower return portfolio. The ELMI+ ap-proach is consistent from the perspective of buildinga publishable index, but it is not the foundation ofan optimal portfolio.


The optimal approach would be to chooseweights that produce the desired return to risk trade-off in the portfolio. Under this approach, securitieswith high yields per unit of risk, or with low covariancewith the portfolio, would be weighted more heavilythan those with lower returns or higher covariance.This might result in a portfolio that contains rela-tively more securities from a small country or a coun-try that exhibits greater volatility but whose returnsare substantially uncorrelated from those of othercountries. One important implication of this last pointis that it has the positive effect of rewarding coun-tries that successfully pursue independent policiesand achieve independent results.

The complicating issue with this approach isthe difficulty in estimating future returns, volatility,and correlations since the historical figures are notalways good predictors of future returns.23 This isespecially problematic for currencies that have hadfixed-exchange rate systems, because the past cor-relations and volatility do not convey the full riskinherent in the positions. The volatility of a fixed-exchange rate currency is historically zero, even ifthe risk of devaluation is significant. There are sev-eral ways to address this mathematically, and thereare market indicators that can also be used to assessvolatility and expected returns.24

Even if, for simplicity, the securities wereequally weighted � that is, with weights of 5 percent for a portfolio with 20 securities � a set of suf-ficiently independent security returns can outperformthe return-to-risk of major security indexes. This isshown below.

Track record

Experiments with investment in localcurrency debt

The potential gains from investing in local cur-rency debt securities were apparent by the mid-1990sand at least some parts of the private sector wereaware of them. From 1993 to just prior to the Rus-sian Federation crisis in July 1998, short-term yieldson local currency debt averaged nearly 27 per cent.The primary risk to investing in local currency debtis currency volatility and so yields were high to com-pensate investors for the risk of devaluation. Yet the

actual average currency devaluation over the sameperiod turned out to be only 14.5 per cent (authors�calculations using list of 46 countries, as shown inappendix B).

Whether it was a fundamental market failureor a matter of incomplete markets giving rise toarbitrage opportunities, there were cases of privatesector investors attempting to take advantage of it.Many broker-dealers set up local currency tradingdesks and several asset management financial institu-tions, that is, �buy-side firms�, launched local cur-rency funds.25 In the summer of 1996, JP Morgan in-troduced the Emerging Local Markets Index (ELMI).26

Yet the ELMI, and later the ELMI+, never be-came a widely followed index. By the end of 1998many of the new local market funds had shut down.Most had taken large and concentrated positions ina few developing countries � notably some large longpositions in the Russian Federation� and when theRussian Federation defaulted the local market fundscame to an end.

In contrast to the funds that made concentratedbets in a few countries, those that were constructedfrom a risk-control perspective survived the crisisrelatively unscathed, with positive returns and lowvolatility.27 The successful strategy lay in recogniz-ing that, despite much of the popular discussion aboutcurrency contagion, correlations across emergingmarket currencies had been actually quite low. Forexample, returns from investing in local currencydebt of 25 countries from JP Morgan�s ELMI+ in-dex returned an averaged 3.03 per cent monthly or43 per cent annual rate of return over the five monthsfrom August through December 1998.

Highly diversified portfolios that were con-structed to take advantage of the low correlationsacross emerging market currencies produced strongresults, relatively high yields and low variance,throughout the various emerging market crises.

LCD track record

In order to better illustrate the market risk ofinvesting in local markets, we have created a sam-ple LCD portfolio using JP Morgan ELMI+ data.For simplicity, our sample portfolio equally weightsthe countries in the index. We did not use the ELMI+


Chart 2

RISK-RETURN PROFILE, 1993�2004

(Percentage)

LCD5.5 / 9.3 ELMI+

8.0 / 8.3

EMBI+16.8 / 13.9

0

2

4

6

8

10

12

14

16

0 5 10 15 20

Risk (per cent)

Ret

urn

(per

cen

t)

weights because, as we explained earlier, the ELMI+is highly concentrated in Asia and thus is not opti-mal diversification. In addition, the ELMI+ includesmajor currency risks implicit in emerging marketcurrencies tied to the euro and the yen. For examplethe local currency securities in Poland, Hungary andthe Czech Republic embody exchange rate risk �when measured in dollar � that is closely tied to thevalue of the euro. As a result they are more highlycorrelated with each other than is desirable for thepurpose of diversification.

The ELMI+ dataset starts in 1994, so we beginour analysis in that year. The first striking result ofour LCD portfolio is that there is only one calendaryear in which returns were negative. That year is1997, the timing of the East Asian crisis, when theLCD portfolio was down -0.78 per cent. In 1994 (theyear of the Mexican crisis), 1998 (the year of theRussian Federation crisis), and 2001 (the year of theArgentinean crisis) returns are all positive, at 3.2 percent, 17.8 per cent, and 5.3 per cent. In contrast, theEMBI+ generated negative returns for those yearsof -18.9 per cent, -14.4 per cent, and -0.8 per cent.

The second striking result is that the averagerisk of the LCD portfolio was only 5.5 per cent from1994 through 2003, significantly lower than theEMBI+ risk of 19.3 per cent over the same period.For comparison, we have also listed the risk/returnprofile of the ELMI+, S&P 500, and Morgan StanleyCapital International (MSCI) Emerging Market FreeIndex of equities (chart 1).

It is important to note that the start date of 1994has a significant effect on measured returns on allasset classes. If the data were to start a year earlier,or a year later, returns would be higher for all emerg-ing market indices (chart 2).

Another important point that is not shown inchart 2 is that the volatility of the LCD portfolio isremarkably consistent over time. We tested the datafor different start dates and found that the volatilityof the LCD portfolio never goes above 6 per cent.This further exemplifies the power of diversifica-tion at lowering risk in this asset class. In contrast,the volatility of the EMBI+ jumps from slightly over9 per cent to nearly 20 per cent, depending onwhether the start date is before or after a crisis year.

Chart 1

RISK-RETURN PROFILE, 1994�2003

(Percentage)

MSCI27.6 / -6.6

LCD5.5 / 8.7

ELMI+7.6 / 8.1

EMBI+19.3 / 9.6

S&P 50021.4 / 7.3

-8

-6

-4

-2

0

2

4

6

8

10

12

0 5 10 15 20 25 30

Risk (per cent)

Ret

urn

(per

cen

t)


Another comparison � and indeed contrast �that illustrates this point is one with the dollar re-turns on the FTSE � a sterling-denominated equityindex. This portfolio is diversified across the 100securities that make up the FTSE index, but it is notdiversified with regard to foreign currencies becausethe dollar exchange rate with the sterling introducesadditional risk to the FTSE equity index. The aver-age annual returns over the same period (January1995�January 2003) were 7.52 per cent and thestandard deviation was 17.73 per cent. This repre-sents a lower yield and higher risk than the LCDportfolio, and a lower yield for about the same levelof risk for the S&P 500 � or alternatively about thesame yield but much higher risk than a portfolio ofthe United States Aaa corporate bonds.

Why the private sector has been slowto implement this strategy

If the LCD policy is such a good idea, why hasthe private sector been so slow to implement it?There are several answers. One is the disappointinghistory of local currency funds in the mid to late1990s, which led investors to be wary of this assetclass. Most of that disappointment can be attributedto funds that took large concentrated bets in a fewcountries, especially the Russian Federation, and thusdid not maintain a diversified portfolio. Nonethe-less, these have served as a bad example of localcurrency portfolios.

Another reason is that the one local currencyindex, the JP Morgan ELMI+, did not perform aswell as a more diversified portfolio.

More recently, local currency capital marketsin some countries have grown, spurring new inves-tor interest in the asset class. A recent study by theIMF (2004a) documents several positive develop-ments in the issuance and trading in local currencysecurities in developing countries. The study citesimprovements in clearing trades in local currencysecurities and the settlement of payments for suchtrades. It also describes expanding liquidity in sec-ondary markets for these securities, and it identifiesthe development of inter-dealer markets as a sign ofgreat liquidity.

Another IMF report (2004b) describes a newinitiative in Asia, called the Asian Bond Market Ini-

tiative. This initiative includes settlement and ex-change regulation, a credit guarantee mechanism,issuance of local currency bonds by non-domesticissuers, and local rating agencies. Similarly, Mexicohas begun to develop a liquid local currency bondmarket. The development of local pension funds inseveral Latin American countries, such as Chile, hasalso stimulated the development of domestic markets.

As local currency bond markets continuedto develop, there has been more interest in localcurrency investments. This was apparent at the De-cember 2003 Emerging Market Traders Associationmeeting at which there was a significant amount ofdiscussion about the asset class. It is unclear, how-ever, how much of any increased investment will gointo diversified funds and how much will go intoconcentrated positions in the few large markets thatare easily accessible to foreign investors.

Two of the biggest complaints by foreign inves-tors are the difficulty of accessing some developingcountries� local currency securities markets and thehigh transaction costs in those markets. Several coun-tries have complicated tax structures, inefficientsettlement mechanisms, high custodial costs, andoutright restrictions to foreign investment. Foreigninvestors have pressured developing countries toreduce these transaction costs, but the countries haveresisted doing so. Brazil, for example, has one ofthe largest domestic securities markets, but it is alsoa market where foreign investors find access cum-bersome.

So, we need to ask the flip side of the questions,if local debt is so good for developing countries, whyhave they sometimes been slow to respond to for-eign interest?

We believe that there are two main answers tothis question. First, foreign currency financing isoften cheaper than domestic financing the risks in-herent in foreign currency financing are ignored.Second, it is possible that countries are wary of in-ternational institutional investors. These investors areseen as myopic, bringing in short-term hot money,overwhelming small capital markets in the goodtimes, pulling out in the bad times and leading toboom-bust scenarios or currency crises. Such flowscan add volatility to a local financial markets anddisrupt if not destabilize the economy. In fact, theproblems associated with short-term capital flowshas been the focus of much economic policy litera-


ture in recent years.28 Similar to concerns that ex-cessive dollar debt is linked to crises, so too haveopen capital markets.29

In this light, market regulations and local se-curity market inefficiencies can be viewed as a meansof restricting capital flows. For example, in the early1990s foreigners were not allowed to purchase Hun-garian Government debt. Later, as the country movedtowards joining the European Union, the restrictionswere eased so that foreigners could buy debt securi-ties with remaining maturities of more than oneyear.30 In this Hungary was able to use controls onits local bond market to regulate short-term capitalinflows.

On implication of this insight is that �originalsin� is in some cases the result of policy decisions.Faced with a choice between two evils, opening theirdomestic markets to the vicissitudes of short-termcapital flows or borrowing from abroad in foreigncurrency, developing countries have often chosen thelater.

The question then becomes, is it possible tostructure local currency portfolios to keep funds ina country long-term? Can we construct the portfolioin a way that still gives policymakers the option tocontinue to selectively use capital account regula-tions to limit short-term inflows? The example ofHungary gives us a hint of how this can be done.Even when foreigners were not allowed to accessthe local bond market directly, they were given thepermission to buy into long-term closed-end funds.31

Alternative ownership structures

Another virtue of the LCD portfolio approachto financing development is its flexibility and abil-ity to take on many different structures to theownership of the portfolio. The minimum size foran economically viable portfolio is not very large,so this would not sharply constrain its use by a vari-ety of financial institutions. For example, a portfolioconsisting of $2 million in securities from each of20 countries would amount to a $40 million portfolio.

There are many financial institutions in ad-vanced economies that are large enough to purchaseand manage a portfolio such as the LCD portfolio,and most of these have the regulatory authority for

cross-border and foreign currency investments.These institutions could construct such a portfolioand hold it as an asset on their balance sheets. Pen-sion funds and insurance companies could alsooperate such a portfolio as a portion of these assets.Similarly, mutual funds and hedge funds could offermore direct ownership of such portfolios. In all thesecases, the most important point is that managersshould stick to a diversified approach, and this fea-ture could be enforced by including diversificationas a requirement in the managed fund prospectus orpension fund investment guidelines.

Multilateral development banks such as theWorld Bank could employ this approach to managethe risk of lending to its developing country bor-rowers. The Bank�s portfolio is already diversifiedacross lenders, and so this analysis shows how itcould safely lend in local currency instead of its ex-isting policy of lending in major currencies. Had theBank been pursuing this approach all along, it wouldhave played a much more stabilizing role in devel-opment finance than it has.

Yet another approach to ownership structurewould be to securitize the LCD portfolio in a man-ner similar to that successfully pursued by UnitedStates mortgage lending titans Fannie Mae, FreddieMac, and Ginnie Mae. They collect together finan-cial assets into a pool that is held by a trust; the trustthen issues securities known as mortgage-back se-curities that represent ownership shares of the poolof assets. These shares can then be traded in a liquidand transparent secondary market so that the proc-ess results in price discovery of the return and riskof the underlying mortgage investments. This attractsadditional capital to the market, provides investorswith greater diversification, and, most of all, helpsthe development of good standard market practicesthat are often lacking in the market for the underly-ing assets or securities.

A similar security that would be backed by othercurrency debt securities, call it an �LCDBS�, couldbe denominated in dollars or any currency. The trustcould be structured to reinvest all proceeds in localcurrency debt and then authorized and empoweredto convert the various local currency proceeds intothe currency of denomination (e.g., dollars) in orderto coupon payments on each scheduled payment dateplus the payment of principal at maturity.32 Thechoice of reinvestment versus regular coupon pay-ments could vary so as to suit the preferences of the


investors. Some investors, such as pension funds andinsurance companies, want the regular payments forcash flow or tax purposes, while other investorsmight prefer reinvestment at the internal rate of re-turn of the portfolio.

As the great wit and sage Yogi Berra warned,�If you can�t copy them, don�t imitate them.� In thiscase, however, the application of this experience withmortgage-backed securities to local currency devel-oping country debt is fairly straightforward and thuslends itself to imitation.

The high performance of the LCD portfolio,which compares favourably with the dollar returnon major United States equity indices, United Statescorporate bond returns, and sterling-denominatedFTSE returns, will attract portfolio capital from in-dividual investors, managed funds, and financialinstitutions.

Of course, the LCD portfolio might attract ad-ditional investor interest if shares were sold ondifferent classes of the portfolio. This is a commonpractice in the securitization of debt through struc-tured securities, known as collateral debt obligations,where different tranches of the pool � based on thepriority in which debt service payments fulfill debtobligation � are sold separately. This generates dif-ferent classes of shares based on creditworthiness,such as a class AAA, a class BB, and junk class (i.e.,speculative grade debt). However, since this LCDportfolio would not be burdened with very muchcredit risk � most developing countries enjoy highratings for debt payments in their own currency �the different classes of shares would be differenti-ated based on market risk. Thus, Class A might offera very high likelihood of providing an 8 per centreturn in dollars, while Class B might offer a stronglikelihood of 10 per cent and Class C might offer aspeculative return of 14 per cent. Such structureshave proven time and again in other areas of ad-vanced financial markets to add significant value tothe portfolio by tailoring the risk-return profile ofthe instrument.

Economic and development consequences

The primary goal of helping developing coun-tries borrow in their own currency is to reduce theirexposure to foreign exchange risk, which has helped

trigger many financial crises in recent decades. Butapart from this, there are additional important eco-nomic benefits of a widespread adoption of LCDportfolios and more local currency (and less foreigncurrency) debt issuance.

The macroeconomic benefits include:

� enabling developing countries to attract moreforeign capital, and to do so in more steady vol-umes because the returns and risks will be moreakin to those found elsewhere in advancedeconomies� capital markets;

� raising the demand for local currency govern-ment securities, and by association other localcurrency securities, so as to lower the cost ofcapital in those markets;

� promoting improvements in local financialmarkets in the areas of clearing, settlements,and secondary market trading;

� stimulating investment and growth by lower-ing local currency interest rates and increasingthe maturity and depth of local credit markets;and

� increasing the government�s revenue from sei-gniorage by increasing the use of local currencyfor trading and servicing local-currency-de-nominated assets; this together with lowerborrowing costs, will improve the fiscal posi-tion of developing country governments.

The microeconomic benefits of the LCD port-folio include:

� it will create a new benchmark in internationalfinancial markets;

� the portfolio, by eliminating most if not all�unique� risk from each security, will establishthe price of �common� or �systematic� risk indeveloping country debt markets;

� the new benchmark would �price� the �marketrisk� � inflation, nominal interest rates and ex-change rate risk � of investing in developingcountry debt instead of the credit risk;

� this benchmark rate of return will have the ef-fect of sharpening competitive pressures oninternational investors by identifying wherelocal-currency rates of return exceed that justi-fied by the benchmark; and


� the local currency interest rates and exchangerates will have the excess risk premiums pricedout of them.

This new power to borrow abroad must come,however, with a strong warning. Some developingeconomies in the recent past have suffered becauseof their governments� fiscal imbalances � wherebylarge and persistent deficits were financed by bor-rowing that was subsequently monetized as part ofmonetary policy. The consequences for inflation,output and growth were sometimes dire. In this con-text, relaxing the constraints on developing countrygovernments� ability to borrow in their currency isexpected to raise concerns about the possible haz-ard of eliminating the usual discipline of foreignborrowing.

While expanding the market for local-currencyfinancial instruments, the LCD policy will not nec-essarily reward any particular country for pursuingreckless fiscal or monetary policies, as it does notguarantee any particular country the right to sell itslocal currency securities abroad. It does thoughoffer a reward for being able to do so, and the magni-tude of the reward will be proportional to a country�sability to produce a stable economic environmentwith lower interest rates and low rates of currencydepreciation. In this regard the policy does not elimi-nate any disciplinary �stick�, but rather it adds�carrots�.

Conclusion

In this study, we describe how to construct aportfolio of emerging market local-currency denomi-nated debt. This portfolio can generate United Statesdollar rates of risk-adjusted return that are competi-tive with familiar financial market benchmarks suchas the S&P 500 and FTSE indexes.

This is a feasible proposal that can be readilyadopted by private financial markets for its profit-able opportunities, and yet it can also be promoted

at negligible expense through public policies by theG-7 Governments, the Finance for Developmentprocess at the United Nations and international fi-nancial institutions such as the World Bank. Thelatter could play a helpful role in establishing a dem-onstration project that would produce marketinformation on prices, returns and risk that couldstimulate others to follow. In addition, the projectcould help finance develop structures that would bevaluable long-term investments.

By comparison to Hausmann�s plan for an EMIndex, this proposal stands as a simple and morestraightforward policy, as it does not require the ap-proval and active participation of major currencygovernments or international financial institutionssuch as the World Bank and IMF. It is also moreflexible in that the portfolio approach can be used toconstruct a variety of types of portfolios that can bedirectly owned or securitized and can be sold off inuniform shares or in risk-related tranches.

This proposal stands as an implicit criticism ofofficial financial institutions that for decades havefailed to identify this policy opportunity that wouldallow them to use their already country diversifiedportfolio to lend in local currencies. Instead they havecontinued practised a policy of lending in major cur-rencies that has lead to financial crises that resultedin more debt in major currencies to help developingcountries adjust to shocks from changes in foreignexchange and interest rates.

Private financial markets too have overlooked,for the most part, this investment opportunity. In anindustry that prides itself on its innovation and itstop-flight financial analysis of arbitrage opportuni-ties, this is a large, squandered opportunity. It is anexample of the proverbial $20 bill on the ground,which theoretical economists believe cannot exist.

This LCD portfolio approach has enormouspotential to promote the maturation of local finan-cial markets as a development policy. It has thecapacity to not only reduce developing countries�vulnerability to financial crises, but also to fostergreater stability and sustained development.


Appendix A

Summary of EM Index proposal

The recent literature on the causes, costs, anddangers of foreign currency borrowing, that is, the�original sin� literature, develops a policy proposalfor redemption. The proposal in Hausmann et al.(2002) consists of three steps. First, construct an in-dex (called an �EM Index�) to determine the rate ofreturn on index-linked financial instruments (bankloans, bonds, and interest rate derivatives). The in-dex is calculated using the changes in each country�sexchange rate and inflation rate and by weightingeach currency in the index by the real output of therespective country.33 The proposal calls for the WorldBank to calculate the official index.

Second, the IFIs (official international financialinstitutions such as the World Bank and Inter-American Development Bank) would issue at leastsome of their debt denominated in the EM Index.This would enable those IFIs to lend in local emerg-ing market currencies while maintaining a matchbetween the currency denomination of the assetsmatch and that of their liabilities. Similarly, the G-10countries, would be called on to issue some of theirdebt in the EM Index.

The IFIs and G-10 borrowers, it is argued,would be motivated by the advantages of diversify-ing their liabilities or alternatively capturing savingsafter swapping their EM Index obligations back intotheir respective C-5 currency (dollar, euro, poundsterling, yen and Swiss franc). This would have thedesirable consequence of creating an otherwise rareshort position in the local currency. The third stepwould thus consist of having the developed coun-tries swap out of C-5-currency-denominated debtback into their own local currency.

Problems with the EM proposal

Following are some limitations of the EM pro-posal, as we see them:

1. The proposal is overly complicated relative tothe problem it seeks to solve. It requires indus-

trial country governments to change their debtmanagement policies to include the use of EM-Index-linked securities, foreign currencies,foreign-currency swap transactions, and add-ing exposure to swap related credit risk fromcounterparties. While the Government of Swe-den has already been successfully developingmore sophisticated debt management policies,these policy recommendations stand in starkcontrast to the debt management policies insuch countries as the United States. Such achange in policy might not prove to be directlybeneficial to the United States, Japan, and eurozone Governments. For example, the policychange would disrupt, and thereby diminish thebenefits from, regularly scheduled debt auc-tions. Also, the proposal would require theG-10 Governments to borrow in a foreign cur-rency and swap back into their own currencyat time when it might � as well as when it mightnot � be cheaper to borrow in their own cur-rency.

2. Individual developing countries would still notbe borrowing directly in their own currency.

3. The most immediate benefit to developing cur-rencies from the emergence of the EM Indexwould be more (long positions in) foreign cur-rency swaps that would enable them to swapout of their hard-currency foreign debts andother obligations.

4. The EM Index proposal does not clarify howthe index-linked debt and securities are to bebought, sold, or in general traded. The transac-tion must be denominated in some singlecurrency unless the proposal�s proponents wantthe entire basket of currencies to be used. Itwould be impractical to buy and sell securitiesthat require multiple currencies for engagingin price quotes, negotiations, and settlement. Ifthe EM Index is comprised of 22 currencies, aslanguage in the literature suggests, it will re-quire a basket of 22 currencies to buy or sell.Otherwise, if it is traded in dollars then it willnot directly help emerging market countriesborrow in their own currency or reduce theirforeign currency exposure � although it wouldgive them a more diversified exposure.


Appendix B

List of 46 developing countries included in for-eign exchange rate analysis:

Argentina Bangladesh BotswanaBrazil Bulgaria ChileColombia Côte d�Ivoire CroatiaCzech Rep. Ecuador EgyptEstonia Ghana HungaryIndia Indonesia IsraelJamaica Jordan KenyaLebanon Lithuania MalaysiaMauritius Mexico MoroccoNamibia Nigeria PakistanPeru Philippines PolandRep. of Korea Romania Russian FederationSlovakia Slovenia South AfricaSri Lanka Thailand Trinidad & TobagoTunisia Turkey UkraineVenezuela

Notes

1 The term major currency refers to the C-5, the dollar,euro, yen, pound sterling and Swiss franc.

2 See Dodd (2001) for an analysis of this dimension of theproblem.

3 One of the authors of this paper, Shari Spiegel, devel-oped and managed a successful risk-controlled diversi-fied local currency debt portfolio in 1995 at Lazard As-set Management.

4 The article will use IMF definitions for terms such asadvanced economies and emerging market economies,but will use the term developing countries to refer toboth emerging and developing countries.

5 This is also true for most other emerging market econo-mies and some newly industrialized economies.

6 The numbers add to slightly more than 100 per cent dueto rounding.

7 Stiglitz (2003).8 For a more thorough discussion of these points see Dodd

(1989), Allen et al. (2002), and Goldstein and Tucker(2004).

9 Funds that could be used for development are insteadheld as reserves. For example, a country might borrowat 10 per cent, but then need to hold reserves against thisdebt in the United States Treasuries at, say, 3 per cent.For more on this, see the upcoming Initiative for PolicyDialogue Overview book on Macroeconomics and Capi-tal Markets Liberalization, forthcoming from OxfordUniversity Press.

10 For a detailed discussion of this, see the upcoming Over-view Volume on Macroeconomics and Capital MarketLiberalization by the Initiative for Policy Dialogue:

Ffrench Davis R, Nayyar D, Ocampo JA, Spiegel S, andStiglitz J, forthcoming from Oxford University Press.

11 The term was coined in Eichengreen and Hausmann (1999).12 Eichengreen and Hausmann show econometrically that

a weak policy framework is not the primary reason coun-tries have been unable to issue debt in domestic curren-cies; hence the term, �original sin�.

13 Goldstein M and Tucker P (1996).14 The Lazard Strategic Yield Fund, May 1995 to April

1997. The Lazard Emerging Income Fund 1, from April1997. The strategy was developed to maintain exposureto local emerging market currencies, and minimize thecurrency risk through diversification.

15 The foundation of portfolio theory in the words of oneof the great contributors to the theory is Sharpe (1970).

16 The term security will be used to describe governmentdebt instruments, although the portfolio could potentiallycontain government loans.

17 Yield and volatility data are taken from average yieldspublished by the JP Morgan as part of the ELMI+ index,1994�2003.

18 The term short-hedger refers to an investors whose busi-ness normally involves a long position in the local cur-rency and needs to hedge that exposure by taking a shortposition in the derivatives market.

19 Our data for these series were monthly, starting in Janu-ary 1994 and ending in May 2004, although some coun-tries do not have observations for the entire period.

20 The dollar is chosen as an example. The same would betrue for other currencies. The rate of return refers to thesum of interest payments and any change in price of thesecurity. This equation yields a continuous time version:r = r* + e. + r*e..

21 Duration is the weighted average of the maturity of thesum of the bond�s cash flows, with the weights calcu-lated by the present value of the corresponding cashflows. It reflects the marginal interest rate risk owing toconvexity.

22 This approach to weights is in contrast to that of Haus-mann et al. for their EM Index approach (appendix A),in which the index weights are decided by GDP. Thatwould result in the largest countries having the largestimpact on the index and their currencies being in greaterdemand in subsequent transactions. This would replicateand reinforce one of the causes of �original sin�, namely,the size of the economy.

23 A succinct warning could be drawn from Yogi Berra�sobservation that �the future is not what it used to be�.

24 One tested method is to use a jump diffusion model in-stead of a normal distribution. While this approach isuseful in determining optimal weights, the results do notalter the general portfolio thesis.

25 To name a few: Lazard Asset Management launched aLocal Currency Trust in 1995, ANZ Asset Managementand Morgan Stanley Asset Management launched fundssoon thereafter, and other asset managers launched re-gional funds.

26 The ELMI was later supplemented by the ELMI+. Thenew index includes more countries and uses currencyforwards to estimate local yields.

27 The Lazard Asset Management Local Currency Trust andaffiliated funds is the one fund the authors know of thatmaintained a diversified investment strategy, based on arisk-management perspective.


28 Rodrik D and Velasco A (2000).29 World Bank (2000); and Demirguc-Kunt A and Detragiache E

(2001).30 Eventually the market was made fully open.31 One of the authors, Shari Spiegel, also managed one of

the first local currency fixed income closed-end fundsin the Hungarian domestic market in the early 1990s.One tranche of the fund was open to foreign investors.

32 Note that mortgage-backed securities involve a similarprocess of collecting receipts from the many mortgagesin the portfolio, whose payments fall across many dif-ferent dates, and then efficiently managing these cashflows until they are paid out on regularly scheduled pay-ment dates.

33 Real output is measured by the purchasing power parity(PPP) value of the gross domestic product (GDP) andinflation is measured by the consumer price index (CPI).

References

Allen M, Rosenberg C, Keller C, Setser B and Roubini N(2002). �A Balance Sheet Approach to Financial Crisis�,IMF Working Paper (Policy Development and ReviewDepartment) WP/02/210, Washington, DC.

Bacha E and Diaz-Alejandro C (1982). �International Finan-cial Intermediation: A Long and Tropical View�, Essaysin International Finance, #147, Princeton: PrincetonUniversity.

Demirguc-Kunt A and Detragiache E ( 2001). �CML is corre-lated with banking crises�, �Financial Liberalization andFinancial Fragility�, in Gerard Caprio, Patrick Honohanand Joseph E. Stiglitz (eds.) Financial Liberalization,How Far How Fast?, Cambdrige University Press.

Dodd R ( 2001). �The Role of Derivatives in the East AsianFinancial Crisis�, In Lance Taylor and John Eatwell (edi-tors), International Capital Markets: Systems in Transi-tion. Oxford University Press (2002), The Financier,Vol.7, Issue 4, 2001.

Dodd R (1989). �Risky Foundations of Development Finance�,manuscript presented at the Eastern Economics Asso-ciations conference in Baltimore, MD, March 1989.(available at www.financialpolicy.org/dscwriting.htm).

Eichengreen B and Hausmann R (2003). �Original Sin: TheRoad to Redemption�, unpublished manuscript, avail-able at http://ksghome.harvard.edu/~.rhausma.cid.ksg/Original%20Sin/roadtoredemption2003.pdf.

Eichengreen B and Hausmann R (1999). �Exchange Rates andFinancial Fragility�, Paper presented at symposium, New

Challenges for Monetary Policy, hosted by Kansas CityFederal Reserve Bank.

Eichengreen B, Hausmann R and Panizza U (2002). �OriginalSin: The Pain, the Mystery, and the Road to Redemp-tion�, unpublished manuscript, November 2002, avail-able at http://ksghome.harvard.edu/~.rhausma.cid.ksg/.

Goldstein M and Tucker P (2004). �Controlling Currency Mis-matches in Emerging Markets�, Washington, DC, Insti-tute of International Economics.

Goldstein M and Tucker P (1996). �Banking Crises in Emerg-ing Economies: Origins and Policy Options�, BIS Eco-nomic Papers, No. 26 (October).

Goodman L (1980). �The Pricing of Syndicated EurocurrencyCredits�, Quarterly Review Federal Reserve Bank ofNew York (Summer).

Goodman L (1981). �Bank Lending to Non-OPEC LDCs: AreRisks Diversifiable?� Quarterly Review Federal ReserveBank of New York (Summer).

Hausmann R (2002). �Good Credit Rations, Bad Credit Rat-ings: The Role of Debt Denomination�, unpublishedmanuscript, Harvard University, available at http://ksghome.harvard.edu/~.rhausma.cid.ksg/.

Hausmann R, Panizza U and Stein E (2001). �Why Do Coun-tries Float the Way They Float?� Journal of Develop-ment Economics, 66: 387�414. Also, Inter-AmericanDevelopment Bank Working Paper #148, May 2000,available at http://ksghome.harvard.edu/~.rhausma.cid.ksg/WP/pubWP-418.pdf.

International Monetary Fund (2004a). Emerging Local Secu-rities and Derivatives Markets, World Economic and Fi-nancial Surveys (from the Global Financial StabilityReport), Washington, DC.

International Monetary Fund (2004b). Global Financial Sta-bility Report: Market Developments and Issues, Wash-ington, DC.

Lessard D (1983). �The Implications for Multinational Bank-ing�, Journal of Banking and Finance, 7, pp. 521�536.

Levy W (1980). �Oil and the Decline of the West�, ForeignAffairs (Summer).

Pollack G (1974). �The Economic Consequences of the En-ergy Crisis�, Foreign Affairs (April).

Rodrik D and Velasco A (2000). �Short-Term Capital Flows�,in Annual World Bank Conference on Development Eco-nomics, World Bank, 1999, Boris Pleskovic and JosephStiglitz (eds.), Washington: World Bank.

Sharpe W (1970). Portfolio Theory and Capital Markets, NewYork: McGraw-Hill Book Company.

Stiglitz J E (2003). �Whither Reform? Towards a New Agendafor Latin America�. The 2003 ECLAC Prebisch lecturepublished in CEPAL Review, No. 80, August, pp.7�38.

World Bank (2000). Global Development Finance 2000, �Short-term flows and vulnerability�, Washington: World Bank.


G-24 Discussion Paper Series*

Research papers for the Intergovernmental Group of Twenty-Four on International Monetary Affairs

No. 33 November 2004 Ilene GRABEL Trip Wires and Speed Bumps: Managing FinancialRisks and Reducing the Potential for Financial Crisesin Developing Economies

No. 32 October 2004 Jan KREGEL External Financing for Development and InternationalFinancial Instability

No. 31 October 2004 Tim KESSLER and Assessing the Risks in the Private Provision ofNancy ALEXANDER Essential Services

No. 30 June 2004 Andrew CORNFORD Enron and Internationally Agreed Principles forCorporate Governance and the Financial Sector

No. 29 April 2004 Devesh KAPUR Remittances: The New Development Mantra?

No. 28 April 2004 Sanjaya LALL Reinventing Industrial Strategy: The Role of Govern-ment Policy in Building Industrial Competitiveness

No. 27 March 2004 Gerald EPSTEIN, Capital Management Techniques in DevelopingIlene GRABEL Countries: An Assessment of Experiences from theand JOMO, K.S. 1990s and Lessons for the Future

No. 26 March 2004 Claudio M. LOSER External Debt Sustainability: Guidelines for Low- andMiddle-income Countries

No. 25 January 2004 Irfan ul HAQUE Commodities under Neoliberalism: The Case of Cocoa

No. 24 December 2003 Aziz Ali MOHAMMED Burden Sharing at the IMF

No. 23 November 2003 Mari PANGESTU The Indonesian Bank Crisis and Restructuring: Lessonsand Implications for other Developing Countries

No. 22 August 2003 Ariel BUIRA An Analysis of IMF Conditionality

No. 21 April 2003 Jim LEVINSOHN The World Bank�s Poverty Reduction Strategy PaperApproach: Good Marketing or Good Policy?

No. 20 February 2003 Devesh KAPUR Do As I Say Not As I Do: A Critique of G-7 Proposalson Reforming the Multilateral Development Banks

No. 19 December 2002 Ravi KANBUR International Financial Institutions and InternationalPublic Goods: Operational Implications for the WorldBank

No. 18 September 2002 Ajit SINGH Competition and Competition Policy in EmergingMarkets: International and Developmental Dimen-sions

No. 17 April 2002 F. LÓPEZ-DE-SILANES The Politics of Legal Reform

No. 16 January 2002 Gerardo ESQUIVEL and The Impact of G-3 Exchange Rate Volatility on Felipe LARRAÍN B. Developing Countries


* G-24 Discussion Paper Series are available on the website at: www.unctad.org. Copies of G-24 Discussion Paper Series maybe obtained from the Publications Assistant, Macroeconomic and Development Policies Branch, Division on Globalization andDevelopment Strategies, United Nations Conference on Trade and Development (UNCTAD), Palais des Nations, CH-1211 Geneva 10,Switzerland.

No. 15 December 2001 Peter EVANS and Organizational Reform and the Expansion of the South�sMartha FINNEMORE Voice at the Fund

No. 14 September 2001 Charles WYPLOSZ How Risky is Financial Liberalization in theDeveloping Countries?

No. 13 July 2001 José Antonio OCAMPO Recasting the International Financial Agenda

No. 12 July 2001 Yung Chul PARK and Reform of the International Financial System andYunjong WANG Institutions in Light of the Asian Financial Crisis

No. 11 April 2001 Aziz Ali MOHAMMED The Future Role of the International Monetary Fund

No. 10 March 2001 JOMO K.S. Growth After the Asian Crisis: What Remains of theEast Asian Model?

No. 9 February 2001 Gordon H. HANSON Should Countries Promote Foreign Direct Investment?

No. 8 January 2001 Ilan GOLDFAJN and Can Flexible Exchange Rates Still �Work� in FinanciallyGino OLIVARES Open Economies?

No. 7 December 2000 Andrew CORNFORD Commentary on the Financial Stability Forum�s Reportof the Working Group on Capital Flows

No. 6 August 2000 Devesh KAPUR and Governance-related Conditionalities of the InternationalRichard WEBB Financial Institutions

No. 5 June 2000 Andrés VELASCO Exchange-rate Policies for Developing Countries: WhatHave We Learned? What Do We Still Not Know?

No. 4 June 2000 Katharina PISTOR The Standardization of Law and Its Effect on Develop-ing Economies

No. 3 May 2000 Andrew CORNFORD The Basle Committee�s Proposals for Revised CapitalStandards: Rationale, Design and Possible Incidence

No. 2 May 2000 T. Ademola OYEJIDE Interests and Options of Developing and Least-developedCountries in a New Round of Multilateral Trade Nego-tiations

No. 1 March 2000 Arvind PANAGARIYA The Millennium Round and Developing Countries:Negotiating Strategies and Areas of Benefits

G-24 Discussion Paper Series*

Research papers for the Intergovernmental Group of Twenty-Four on International Monetary Affairs

Randall Dodd and Shari Spiegelunctad.org/en/docs/gdsmdpbg2420051_en.pdfIntroduction The massive...

Documents

Transcript of Randall Dodd and Shari Spiegelunctad.org/en/docs/gdsmdpbg2420051_en.pdfIntroduction The massive...