Mario de Zamaroczy and Sopanha Sa · 2. Empirical Models of Dollarization 3. Theoretical Models of Dollarization 4. Causality Analysis Between Inflation and Dollarization in Cambodia,

OCCASIONAL PAPER

Economic Policy in aHighly Dollarized Economy

The Case of Cambodia

Mario de Zamaroczy and Sopanha Sa

INTERNATIONAL MONETARY FUNDWashington DC

2003

©International Monetary Fund. Not for Redistribution

© 2003 International Monetary Fund

Production: IMF Multimedia Services DivisionTypesetting: Choon Lee

Figures: Theodore F. Peters, Jr.

Cataloging-in-Publication Data

Zamaroczy, Mario de.Economic policy in a highly dollarized economy : the case of Cambodia /

Mario de Zamaroczy and Sopanha Sa — Washington, D.C.: InternationalMonetary Fund

p. cm. — (Occasional paper ; 219)

Includes bibliographical references.ISBN 1-58906-189-6

1. Cambodia — Economic policy. 2. Dollar, American. I. Sa,Sopanha. II. International Monetary Fund. III. Occasional paper(International Monetary Fund); no. 219HC442.Z35 2003

Price: US$25.00(US$22.00 to full-time faculty members and

students at universities and colleges)

Please send orders to:International Monetary Fund, Publication Services

700 19th Street, N.W., Washington, D.C. 20431, U.S.A.Tel.: (202) 623-7430 Telefax: (202) 623-7201

E-mail: [email protected]: http://www.imf.org

recycled paper


http://www.imf.org

Contents

Acknowledgments

I Overview

II Economic and Financial Developments

III Dollarization in Cambodia

A Synopsis of the Concepts of Dollarization and ofCurrency Substitution

Measuring Dollarization in Cambodia

IV Costs and Benefits of High Dollarization in Cambodia

Drawbacks of High DollarizationBenefits of High Dollarization

V Implications of High Dollarization for MacroeconomicPolicy Design

Dollarization and the Banking SystemExecution of BudgetConsideration of a Possible Currency Board Arrangement

VI Conclusion

Appendix

Cambodia—A Simple Model to Estimate Dollars in CirculationOutside Banks

Bibliography

Boxes

1. Key Dates in Cambodian Political and Financial History2. Empirical Models of Dollarization3. Theoretical Models of Dollarization4. Causality Analysis Between Inflation and Dollarization in

Cambodia, Lao P.D.R., and Vietnam5. Measuring Seigniorage6. Baivkmg Reform m Cambodia7. Currency Board Arrangement Versus National Bank of

Cambodia Policies

Tables

1. Ratio of Dollars Circulating in the Economy2. Estimates of Seigniorage

2

5

55

14

1417

20

202223

26

28

32

267

111421

24

1215

v

1

iii


3. Interest Rates on Deposits and Loans, February 20024. Budget Execution in Foreign Currency5. Velocity of Broad Money I6. Maximum Likelihood Estimates for the State-Space Representation7. Velocity of Broad Money II8. Value of the Proportionality Coefficient

Figures

1. Foreign Currency Deposits2. Real GDP and Inflation3. Assets and Liabilities of the Banking System in

Foreign Currency4. Dollarization Ratios5. Ratio of Foreign Currency Deposits to Broad Money6. Foreign Currency Deposits in Cambodia, Lao P.D.R.,

and Vietnam7. Dollars in Circulation Outside8. Estimated Currency Substitution9. Ratio of Dollars in Circulation to Foreign Currency Deposits

10. Estimated Dollarization11. Official International Reserves12. Spread Between Official and Market Exchange Rates13. Riels per U.S. Dollar14. Riels per Thai Bhat and 100 Vietnamese Dong15. Effective Exchange Rates16. Co-Currency Circulation

162330303131

34

1010121213151717181822

The following symbols have been used throughout this paper:

. . . to indicate that data are not available;

— to indicate that the figure is zero or less than half the final digit shown, or that the item doesnot exist;

between years or months (e.g., 2000-01 or January-June) to indicate the years or monthscovered, including the beginning and ending years or months;

/ between years (e.g., 2000/01) to indicate a fiscal (financial) year.

"Billion" means a thousand million.

Minor discrepancies between constituent figures and totals are due to rounding.

The term "country," as used in this paper, does not in all cases refer to a territorial entity that isa state as understood by international law and practice; the term also covers some territorial enti-ties that are not states, but for which statistical data are maintained and provided internationallyon a separate and independent basis.

CONTENTS

889


Acknowledgments

We are indebted for useful comments and suggestions to Thomas Rumbaugh,Philippe Callier, Edward Frydl, Balazs Horvath, Vitale Kramarenko, Tola May,Johannes Mueller, Gabriel Sensenbrenner, and other IMF colleagues. We are gratefulto Chanthana Neav from the National Bank of Cambodia for providing data andinsightful comments, to Marc Paoletti (IMF resident advisor in Cambodia) for bud-getary data, and to Paolo Guarda (Central Bank of Luxembourg) for comments.Chenda Pich and Myrna Bas offered superb secretarial assistance. Jeff Hay den of theExternal Relations Department edited the manuscript and coordinated its production.We bear responsibility for any remaining errors. The views expressed in thisOccasional Paper are those of the authors and do not necessarily represent those ofthe IMF or IMF policy.



I Overview

I n recent years a number of countries have adoptedthe U.S. dollar as their official currency. In some

other countries the dollar has become a semi-officialcurrency. This "dollarization" occurs in a variety ofways. Some countries allow several currencies to cir-culate side by side and over time the dollar dominates(so-called "de facto dollarization"). Others make aconscious choice, opting for the dollar because of itstraditional stability. In Cambodia, this process hasbeen particularly rapid and extensive.

The signing of the Paris Peace Agreements onOctober 23, 1991 heralded the political and eco-nomic rebirth of the Kingdom of Cambodia aftermore than 20 years of continuous civil and interna-tional wars. The United Nations Transitory Authorityin Cambodia (UNTAC) oversaw the country's polit-ical and economic management from 1991 until1993, when free elections brought a civilian govern-ment to power. Since then the country has achievedan economic rebound, albeit from a very low base.Although the country achieved good economicprogress during 1994-95—among other things, inthe framework of an International Monetary Fund(IMF)-supported program—factional fighting brokeout briefly in July 1997, resulting in a temporary set-back in development and foreign investment. Elec-tions in 1998 brought a coalition government to thehelm and, through the surrender of the last KhmerRouge, the country returned to peace and stability atthe end of 1998. The coalition government has beenable to focus on economic and structural reforms,and embarked upon a new IMF-supported programin October 1999.

A notable feature of the Cambodian economy isits high level of dollarization, which presents achallenge for decision makers to devise the bestpolicy mix for sustainable growth, coupled withsteadfast poverty reduction. Dollarization was nei-ther sought nor encouraged by the monetary author-ities. Rather, it came from the "supply side" in theform of sudden and massive inflows of foreigncurrency—continuing to date—stemming from siz-able international assistance, private transfers, and

export earnings. Such large inflows of dollars fromoverseas, coupled, on the "demand side," with a lackof confidence in the domestic currency and politicaluncertainties, provided the impetus for speedy dol-larization, which is a unique feature of Cambodia'seconomic experience. Since the authorities haveadopted an open economy and a liberal exchangesystem, the U.S. dollar has become a de facto sec-ond legal tender along with the national currency,the riel. As a result, Cambodia has been confrontedwith multiple currencies circulating freely through-out its territory, to the point that the dollar hasbecome the dominant currency, with the riel playinga relatively minor role.

Cambodia achieved almost complete de facto dol-larization during 1991-95 and this condition hascontinued to prevail since then. The country islargely a cash-based economy, with a large amountof cash dollars circulating outside the banking sys-tem. The originality of recent Cambodian economicpolicy is that it has been akin to an "orthodox" cur-rency board arrangement, yet it has been imple-mented in a virtually fully dollarized environment.This policy has served Cambodia well since 1999,but a number of risks associated with the growingeconomy call for close monitoring of economicdevelopments.

This paper is organized as follows. Section II pre-sents a short description of economic, financial, andstructural developments in Cambodia since indepen-dence, but focuses on the decade ending in 2001,which serves as a backdrop for the discussion on theemergence of dollarization. Section III reviewsrecent developments in the literature on dollarizationand discusses the degree of dollarization in Cambo-dia. Owing to the unique way in which dollarizationwas introduced in Cambodia and to the specific char-acteristics of Cambodia's economy, an attempt ismade to provide an econometric estimation of cashforeign currency circulation. Section IV examinescosts and benefits of dollarization in Cambodia, andSection V discusses the ensuing macroeconomic pol-icy implications.


II Economic and Financial Developments

Cambodia became independent from France in1953 and experienced a 17-year period of rela-

tive political stability and steady economic growth,starting from a development base similar, if not supe-rior, to that of other countries in South East Asia(Box 1). During that period, the riel was the legaltender and was used both for financial transactionsand as a store of value. Following the March 1970coup d'etat that toppled the government, the countrywas quickly drawn into the international turmoil ofthe subregion, and eventually into a civil war. In1971, a flexible exchange rate system was intro-duced, while the exchange rates were unified. How-ever, owing to economic difficulties, a dual exchangerate system was reinstated in 1973 with a "basic"rate for most transactions and a "preferential rate"for aid-related imports and services.1

The extreme revolutionary experience (an "agrar-ian moneyless society") of the Khmer Rouge during1975-79 involved bans on private property, banking,and money, and brought the country to utter eco-nomic and human disaster, during which an esti-mated one and a half million people, about one-fifthof the population, perished.2 The Khmer Rouge weredriven out of most parts of Cambodia by the Viet-namese in 1919.3 Under the new transition regime,the riel was reintroduced in 1980 as the domesticcurrency, and an official dual exchange rate systemwas created. A state-owned monobank was set up atthe same time, whose role included central, commer-cial, and development banking activities, accordingto the socialist economic model.

Cambodia had its first experience with limited dollarizationduring the Lon Nol regime (1970-75), as increases in U.S. mili-tary personnel and assistance brought dollars into the country.

2Although the Khmer Rouge forcibly emptied the capital of allinhabitants a few days after seizing it, they destroyed only twobuildings in Phnom Penh, one of which was the National Bank ofCambodia's (NBC) headquarters.

3In view of the shortcomings of their economic management,the Khmer Rouge considered reintroducing money in 1976, andwent as far as printing bank notes, but they stopped short of pro-ceeding. They later introduced a parallel currency, the Khmer riel,in March 1993 in the western border areas of the country undertheir control. This currency had only a limited circulation.

Box I. Key Dates in Cambodian Politicaland Financial History

November 9, 1953—Independence from France.March 18,1970—Lon Nol coup d'etat.April 17, 1975—Khmer Rouge take Phnom Penh

and empty the capital.January 7,1979—Vietnamese army enters Phnom

Penh and establishes a new government.1980—Reintroduction of the riel as domestic

currency.October 23, 1991—Paris Peace Agreements.May 1993—National elections sponsored by the

United Nations.June 1993—Restoration of monarchy.July 1993—Formation of a national government.May 6, 1994—IMF Enhanced Structural Adjust-

ment Facility (ESAF I) program.January 26,1996—Law on the Organization and

Conduct of the National Bank of Cambodia.July 5-6,1997—Factional fighting in Phnom Penh.August 22,1997—Law on Foreign Exchange.July 26, 1998—National elections.November 23,1998—Formation of a coalition

government.January 1,1999—Introduction of the value-added

tax.October 22, 1999—IMF ESAF II/Poverty Reduc-

tion Growth Facility (PRGF) program.November 18,1999—Law on Banking and Finan-

cial Institutions.May 31, 2000—Deadline for submitting applica-

tions for a new license for commercial banks.January 2, 2001—Launching of a clearinghouse

for dollar-denominated checks.December 31, 2001—Deadline for meeting the

new capital requirement for commercial banks.January 1, 2002—Acceptance of IMF Article VIII

status, which signifies general avoidance ofrestrictions on current payments.

March 19, 2002—Completion of relicensing ofcommercial banks.


Economic and Financial Developments

Dollars started to flow into Cambodia in the mid-1980s, as the United Nations (UN) dispatchedhumanitarian and emergency aid, international non-governmental organizations were allowed to return,and remittances from abroad resumed. During the1980s, the country achieved only limited monetiza-tion and most domestic transactions were based onbarter, with gold being the universal means of trans-acting and hoarding.4 During 1988-91, the Viet-namese disengaged from Cambodia, leaving anunsettled political situation in their wake. Massivecentral bank financing of recurrent budget deficitsduring that period resulted in high inflation, in therange of 90 to 177 percent a year (end-period), andin an erosion of public confidence in the nationalcurrency. In 1993, the official exchange rates wereunified de facto, and since then a free but managedfloating exchange rate regime has existed.5

During 1991-93, UNTAC took over the country'sadministration. UNTAC has represented the UN'scostliest peace restoration effort to date, and resultedin a virtually overnight dollarization of the urbaneconomy. This major nation-building operationinvolved the stationing of up to 22,000 UN militaryand civilian personnel throughout Cambodia, and itis estimated that the total cost of the two-year opera-tion was close to $2 billion. Although evidently notall of this cost occurred in Cambodia, the totalamount represented about 75 percent of the 1993Cambodian GDP.6 The UNTAC personnel arrivingwith dollars in cash and needing a wide array of ser-vices (local staff, housing, transportation, interpreta-tion, etc.) and goods in a largely barterized economyquickly introduced Cambodia to massive dollariza-tion. The inflow of cash dollars was compounded bythe return at about the same time of large numbers ofCambodian refugees and expatriates from abroad,also bringing dollars or Thai baht with them. As aresult, foreign currency deposits in the banking sys-tem started to rise notably. Since 1992, they haveconstituted an increasingly important component ofthe banking system's deposit base (Figure 1).

In 1994, the authorities embarked upon a reformprogram, supported by the IMF under the EnhancedStructural Adjustment Facility (ESAF), and madenotable progress in reining in inflation to single-digitlevels, strengthening the fiscal position, furtheringfinancial stabilization, and opening up the economy.

4The domleung, the unit for gold widely used in Cambodia,weighs 37.5 grams.

5During 1993-98, the National Bank of Cambodia (NBC) auc-tioned off a total of $177 million to strengthen the riel; however,since 1999, the NBC has refrained from doing so, except for thesale of a small amount of dollars in April-May 2001 to relievetemporary pressure on the riel's exchange rate.

6No GDP estimates exist for earlier years.

Figure I. Foreign Currency Deposits(in millions of U.S. dollars)

400-

350-

300-

250-

200-

150-

100-

50-

0

WA K>Ji-t. H ) >'••.••': •

1I

• ' • • * * n

1990 92 94 96 98 2000

Source: National Bank of Cambodia.

The government also achieved limited progress insome areas of structural reforms, such as state-owned enterprise reform and privatization, but policyimplementation in other areas (notably civil servicereform and forestry management) met with difficul-ties. However, owing to the continued unsettledpolitical situation and to a resurgence of governanceproblems, the ESAF program was suspended in Sep-tember 1995, and it expired in August 1997. Cumu-lative disbursements amounted to about $61 million.The political situation deteriorated to the point thatfactional fighting erupted in July 1997. Compoundedby the onset of the Asian crisis, the economic situa-tion worsened, growth slowed, the governmentresorted to budget financing from the National Bankof Cambodia in 1998 for the first time since 1993,and inflation picked up to double digits.

Not surprisingly, against the backdrop of contin-ued large inflows of foreign assistance and privatetransfers, and of political uncertainty, the dollariza-tion of the Cambodian economy continued during1994-96, as illustrated by the continued rise in for-eign currency deposits. However, the July 1997fighting and the regional financial crisis led to a dete-rioration in confidence. National elections took placeon July 26, 1998, but it took until November 1998 tocomplete the formation of a coalition governmentand to restore domestic stability and internationalconfidence. As a combined result of the expansion-ary policy and of the regional financial crisis, the rieldepreciated during mid-1997 and 1998, but its depre-ciation was smaller than those of other Asian curren-cies. The limited use of the riel in Cambodiaminimized the impact of this depreciation on the


II ECONOMIC AND FINANCIAL DEVELOPMENTS

economy. The effects of the political uncertaintyseem to have been stronger than those of the regionalcrisis. During 1997-98, there was a decline in for-eign currency deposits, but we show in the nextsection that "de-dollarization" during those twoyears, as evidenced by a decline in foreign currencydeposits, is not a correct economic assessment:dollarization (including cash holdings of dollars)during the period under review remained broadlyconstant—however, the composition of dollar-denominated assets held by agents did change.

During 1999, the authorities paved the way forresumed IMF lending by taking a number of funda-mental financial measures (e.g., introduction of thevalue-added tax on January 1, 1999) and structuralmeasures (curbing illegal logging and initiating civilservice and military reforms). In October 1999, theIMF approved a second three-year (1999-2002)ESAF program, soon replaced by a Poverty Reduc-tion and Growth Facility (PRGF) arrangement.Macroeconomic stabilization under the PRGF pro-gram has been largely successful to date: economicgrowth has resumed, fiscal balance has beenrestored, inflation has been reined in, revenue mobi-lization has improved, and a prudent monetarypolicy has been maintained. Accordingly, macroeco-nomic performance quickly rebounded from therelapse in 1997-98 and has improved notably, withestimated economic growth reaching on averagemore than 5 percent a year during 1999-2001,despite severe flooding in 2000 and 2001. Inflation,as measured by the consumer price index (CPI, end-of-period), turned negative during the same period(Figure 2) and the garment, tourism, and agriculturalsectors posted strong growth, resulting in additionalforeign currency inflows for the first two sectors.

Figure 2. Real GDP and Inflation(In percent)

Consumer price index(end-of-period;

right scale)

Sources: Cambodian authorities; and IMF staff estimates.Note: Old CPI, in use until the end of 2001.

The good inflation performance reflected the author-ities' cautious monetary and fiscal stance, as well ascheaper imports from neighboring countries. Theincrease in broad money essentially resulted from abuildup in foreign assets, which itself stemmed froma rise in foreign currency deposits, while net domes-tic assets of the banking system declined, owing to anet repayment of government debt to the NationalBank of Cambodia and a relative lack of financialintermediation.

4


Ill Dollarization in Cambodia

The notion of dollarization emerged as a noveleconomic phenomenon in the 1980s in Latin

America. Early discussions of this phenomenon inthe literature revolved around that region's experi-ence with the increasing use of the U.S. dollar alongwith national currencies.

A Synopsis of the Concepts ofDollarization and of CurrencySubstitution

Dollarization is described in the early literature asa situation where a foreign currency is used for thesame purposes as the national currency—that is, as amedium of exchange, a unit of account, and a storeof value. The loss of the domestic currency's exter-nal value and appeal as a store of value prompts dol-larization and the foreign currency assumes the threeclassic uses of the national currency. According toOrtiz (1983), dollarization is the degree to which realand financial transactions are performed in dollarsrelative to those performed in domestic currency.

Broader notions of dollarization also exist. Cud-dington (1989) and Calvo and Vegh (1992) definecurrency substitution as the use of cash foreign cur-rency and of foreign currency deposits only as amedium of exchange in the domestic economy.McKinnon (1996) suggests a more extensive defini-tion of currency substitution, distinguishing betweendirect currency substitution (several currencies com-pete as means of payment) and indirect currencysubstitution (several currencies serve as nonmone-tary financial assets for stores of value). This distinc-tion between the two motives for the demand forforeign-currency-denominated assets is also knownin the literature as currency substitution and assetsubstitution.1 Currency substitution occurs when for-eign-currency-denominated assets are used as meansof payment, while asset substitution occurs when

they are primarily used as a store of value. Calvo andVegh (1992) point out that asset substitution nor-mally characterizes the late stage of dollarization andthat it appears in a high-inflation environment wherea foreign currency becomes the unit of account or astore of value.8

Dollarization has elicited a large body of empiri-cal and theoretical studies. Despite difficultiesrelated to the measurement of the degree of dollar-ization, there is an extensive empirical literature onthis subject (Box 2). Among the dollarization factorsinvestigated by authors, the high degree of openness,low transactions costs for acquiring foreign currency,and lingering depreciation concerns are characteris-tic of Cambodia.

Theoretical models focus on the different purposesof demand for foreign currency (means of paymentor store of value—Box 3). They have been developedto describe money-demand behavior in countries likeArgentina (Kamin and Ericsson, 1993), Latvia (Sara-jevs, 2000), or in a group of developing countries(Agenor and Khan, 1996). However, neither of thetwo models described applies to Cambodia, as thereare no markets offering the possibility of arbitrageamong various financial assets.


According to Balino, Bennett, and Borensztein(1999), partial or unofficial dollarization "occurswhen residents of a country hold a large share oftheir financial wealth in assets denominated in for-eign currency, where foreign currency lacks the legaltender privileges that domestic currency enjoys."Bogetic (2000) defines full or official dollarization"as a complete monetary union with a foreign coun-try from which a country imports currency, bymaking the foreign currency full legal tender andreducing its own currency, if any, to a subsidiary

7However, there continue to be variations, divergences, andinconsistencies among authors on the notions of "dollarization,""currency substitution," and "asset substitution."

8For simplicity's sake, from this point on we use the term dol-larization for both currency and asset substitution, unless other-wise specified.

5


DOLLARIZATION IN CAMBODIA

Box 2. Empirical Models of Dollarization

The empirical studies can be roughly divided intotwo main groups: qualitative and quantitative analyses.The first group focuses on the costs and benefits of dol-larization in terms of economic performance and eco-nomic policies—Argentina, Ecuador, Mexico, andPanama are popular references—and so are sometransition economies (e.g., the Baltics and the RussianFederation). No consensus has emerged on the per-formance of dollarized versus nondollarized eco-nomies. On the one hand, those who reject dollarizationpoint out the loss of seigniorage, the inability of thecentral bank to act as a lender of last resort, and—moreimportantly—the loss of flexibility and independencein monetary and exchange rate policies, On the otherhand, those who favor dollarization emphasize thatmost dollarized countries exhibit low inflation, macro-economic stability, and enhanced fiscal discipline (e.g.,Balifio, Bennett, and Borensztein, 1999; Berg andBorensztein, 2000a; and Edwards, 2001). Covering theperiod 1970-98, and focusing on IMF member coun-tries with various exchange rate regimes (from peggedto floating, including currency boards), Ghosh, Guide,and Wolf (1998) provide empirical evidence that coun-tries with high levels of dollarization or currency boardarrangements may have sacrificed the flexibility oftheir monetary policy, but gained long-term benefits oflower inflation and a more stable exchange rate. Asnoted by Bogetic (2000), analysis and evaluation for asingle country are relatively rare, except for Panama,which seems to have become a benchmark for relatedworks. It is with regard to growth performance thatstudies differ most. Edwards (2001) provides evidencethat during 1970-98 some dollarized countries (Liberia,Panama, and ten micro-states) experienced: (1) lowerinflation; (2) similar fiscal deficits; and (3) lower GDP

per capita growth than nondollarized economies(emerging and industrial countries, with a variety ofexchange rate regimes, from floating through crawlingto pegged but adjustable exchange rates). To emphasizethese results, Edwards (2001) takes a closer look atPanama. He finds that Panama did not perform betterduring 1970-98 compared with nondollarized referenceeconomies, and external shocks generated higher costsin Panama—in terms of lower investment and growth.Conversely, Ghosh. Guide, and Wolf (1998) find thatsample economies exhibited faster economic growththan nondollarized ones.

The second group of empirical studies attempts toexamine the factors that may explain the emergence ofdollarization in a group of countries. Econometric stud-ies investigate the contribution of two sets of variablesto dollarization: institutional factors and economic fac-tors. Vetlov (2001) identifies the following institutionalfactors as possibly leading to dollarization: an openeconomy, great depth and large size of the domesticfinancial market, and relatively low transaction costs foracquiring foreign currency. He notes that key economicfactors arc the interest rate spread (between domesticcurrency and foreign currency deposits), the inflationrate differential (between domestic and foreign infla-tions), and devaluation expectations. Other variables,which can signal an expected devaluation, are also con-sidered: the real exchange rale, the current accountdeficit, international reserves, and other related factors.To explain dollarization, most studies arc money-demand models, in which the explanatory variables usu-ally include a combination of the aggregates notedabove. Agenor and Khan (1996) find that foreign inter-est rates and the expected depreciation of the exchangerate can be important factors leading to dollarization.

role." In 2001, 27 countries had officially dollarizedtheir economies, 7 of which used the currency ofanother country and 20 belonged to a currencyunion.

An alternative to full dollarization is the officialbimonetary system in which the domestic and a for-eign currency are both legal tenders. The foreign cur-rency usually tends to dominate in bank deposits, butthe domestic currency is often used for the executionof the budget, including the payment of taxes and ofthe salaries of civil servants. Transactions can bepaid for in one or both currencies, according to dif-ferent country experiences. Cambodia's legal tenderis the riel, but Cambodia's monetary system is char-acterized by a de facto dollarization, resulting in anunofficial multimonetary system. The dollar iswidely used by residents but it is not a legal tender.In urban areas the riel is chiefly used for small cashtransactions and as divisionary money. In those

areas, prices for goods and services are mostlyquoted in dollars and transactions are predominantlysettled in dollars. In rural areas, far from the borders,the riel remains the main means of payment, andserves also as a store of value, along with gold. Theriel is mainly kept in circulation through governmentpayments for goods and services (including civil ser-vants' salaries). In the border areas with Thailand,the Thai baht is widely circulating and is preferred tothe dollar. Along the border with Vietnam, the Viet-namese dong circulates to a limited extent amongtraders who engage in regular cross-border trading.9

Cambodia is essentially a cash-based economy,with checks used only by large institutions in urban

9It seems that because of its continued depreciation in recentyears, the Lao kip is not used/accepted in Cambodian border areaswith the Lao P.D.R.

6



Box 3. Theoretical Models of Dollarization

Giovannini and Turtelboom (1994) provide a usefulsurvey of the theoretical literature on dollarization,which takes money demand in a multicurrency environ-ment as its starting point. The bulk of the analysis canbe divided into two broad avenues: cash-in-advancemodels, and transactions costs models.

In the cash-in-advance models, as reviewed forinstance by Giovannini and Turtelboom (1994) andSarajevs (2000), two currencies exist in the economy,with or without legal restrictions. Both domestic andforeign currencies are used as a medium of exchange.Moreover, the following underlying assumptions areposed: there is perfect substitution between the two cur-rencies and financial assets (to avoid idle cash bal-ances), and the acquisition of foreign currency iscostless. Agents are maximizing their utility function(which is increasing in consumption and foreign cur-rency holdings), subject to budget and cash-in-advanceconstraints, by choosing between domestic and foreigncurrencies, and domestic and foreign bonds with posi-tive returns. It is assumed that prior to any consumptionpurchase, agents must acquire domestic or foreign cur-rencies. The return on bonds in foreign currencyexpressed in domestic currency depends on theexpected exchange rate depreciation and will positivelyaffect the demand for foreign bonds and negativelyaffect the demand for domestic assets. In fact, the dif-ferential of real return (nominal return minus inflationrate) from domestic and foreign currencies determinesthe demand for domestic and foreign currencies.Bogetic (2000) asserts that foreign currency holdingsare closely linked to domestic and foreign inflationrates and the credibility of legal restrictions. The higherthe domestic inflation rate compared with the foreignone, the higher the level of foreign currency holdings.Conversely, the stronger the legal restrictions, the lowerthe level of foreign currency holdings. If there are nolegal restrictions, there are two possible equilibria inwhich only one currency circulates: no currency substi-

tution (low inflation) or full substitution (high or hyper-inflation). Thus, in this model, the two currencies cir-culate simultaneously only if there are some credibleand effective legal restrictions.

In the transactions costs models, developed by Poloz(1986) and Marshall (1987), it is time consuming toacquire cash (both domestic and foreign) to purchasegoods. The inability of agents to acquire cash instanta-neously forces them to build up cash balances. Theassumptions in this model are as follows: two currenciescirculate simultaneously, there are available financialassets, and there are transactions costs, which reduce theagents' disposable income. The transactions costs func-tion takes into account the time and resources requiredto acquire cash (i.e., time consumed increasing cash bal-ances decreases the working time and consequentlyincome). Agents maximize their utility function (whichis increasing in consumption solely) subject to a budgetconstraint. This framework differs in two ways from theprevious one. First, foreign currency holdings do notappear in the utility function. Second, there are transac-tions costs that reduce the disposable income of agents.By assumption, the transactions costs function isincreasing in domestic and foreign goods consumptionand decreasing in real money balances (domestic andforeign). In other words, agents need money (domesticand foreign) to purchase goods; consequently, they willbear transactions costs in converting their assets intocash in both currencies. The "liquidity" of differentassets in the agents' portfolios will determine thedemand for different currencies to permit the purchaseof goods. Given the specific form of the transactionscosts function, this framework leads to different equilib-ria, depending on consumption levels and transactionscosts. If these levels and costs are high, agents will holdforeign assets for the purpose of payment and as a storeof value, and thus, currency substitution will occur.Conversely, if consumption levels and transactions costsare low, agents will prefer to use the domestic currency.

areas, and virtually no electronic payments. Financialintermediation is shallow and transactions in thebanking system are predominantly effected in dollars.Most banks (except the state-owned commercialbank) keep their books in dollars, using the daily mar-ket exchange rate to convert transactions in riels intodollars. During 1997-98, about 97 percent of bankassets and 95 percent of bank liabilities were denom-inated in dollars. However, since the 1998 nationalelections, assets and liabilities of commercial banksdenominated in riels have increased modestly, asmacroeconomic stabilization and growth led to anexpansion of banking operations in riels (Figure 3).

Beyond a qualitative description of dollarizationoccurring in an economy, the main challenge for char-

acterizing this phenomenon more precisely is themeasurement of the degree of dollarization. Usuallyforeign-currency-denominated assets held within acountry's financial system are reasonably well known,but the amount of foreign currency circulating in cashin the economy is unknown. Balino, Bennett, andBorensztein (1999) report that, based on U.S. Trea-sury data on net dollar inflows during 1989-96 inselected small open economies—comparable to Cam-bodia—such inflows represented three to four timesthe amounts of local currency in circulation.

Authors in this field of research acknowledge theproblem of measuring the true degree of dollariza-tion (i.e., foreign cash in circulation plus foreign-currency-denominated assets held in the banking

7



Figure 3. Assets and Liabilities of theBanking System in Foreign Currency(In percent of total assets and liabilities)

Figure

I 20 —

4. DoUarization RatiosWiiSSBB. •

2.5 1

1997 98 99 2000


system) and suggest circumventing the lack of dataon foreign cash circulation by comparing a monetaryaggregate denominated in foreign currency toanother monetary aggregate denominated in domes-tic currency (possibly including a subcomponentdenominated in foreign currency). These aggregatesare usually extracted from monetary statistics, whichare in general among the better statistics, especiallyin developing countries. For instance, Calvo andVegh (1992) propose to proxy the degree of doUar-ization by the ratio of foreign currency deposits heldin commercial banks to broad money (inclusive offoreign currency deposits). In most cases, the choiceof a doUarization index relies on data availability.Commonly, data on foreign banknotes circulating inthe economy are unavailable. The only series typi-cally available is residents' foreign currency depositsin the domestic banking system. Thus, habitual dol-larization indices tend to underestimate the truedegree of doUarization, as they do not include cashforeign currency in circulation.

Following Vetlov (2001), we start by using threeof the most common doUarization ratios (DR) foundin the literature (deposits defined as demand, sav-ings, and time deposits):

DR\\ residents' foreign currency deposits to domes-tic currency deposits;

DR2'. residents' foreign currency deposits to the sumof residents' domestic currency deposits anddomestic currency in circulation; and

1995 97 99 2001


DRy. residents' foreign currency deposits to broadmoney (M2).

In the case of Cambodia, the main differencebetween these ratios lies in their respective levels(Figure 4). With an appropriate re-scaling, thesethree time series follow each other rather closely.The cross-correlation coefficients between the ratiosare high: 0.99 between DR3 and DR2; 0.80 betweenDR3 and DR\\ and 0.76 between DR2 and DRh

Using ratio DR3, the degree of doUarization inCambodia rose from 50 percent at the end of 1994 to70 percent at the end of 2001. Between December1994 and July 1997, DR3 increased from 50 percentto 60 percent. It remained stable around that level forseven months, and then started to rise slightly, reach-ing 64 percent in January 1998. Between Januaryand December 1998, doUarization declined from64 percent to 54 percent. As noted in Section II,political uncertainty could explain this episode of"de-dollarization." In late 1998, after the formationof a coalition government, increased public confi-dence and heightened economic activity, in particu-lar in the tourism and retail trade sectors, resulted inthe return of foreign currency deposits to the domes-tic banking system. Since the beginning of 1999, thedegree of doUarization has constantly increased,reaching a 71 percent peak in November 2001.

Balino, Bennett, and Borensztein (1999) charac-terize an economy as highly dollarized when DR3exceeds 30 percent. According to this definition,Cambodia can be viewed as a "very highly" dollar-ized economy and finds itself in the uppermost group

III

8



10 20

Sources: IMF, International Financial Statistics', and IMF staff estimates.'2000 data.

of dollarized economies (Figure 5). At the end of2000 or 2001, dollarization in the world—excludingfull dollarization—as measured by DR$, ranged from7 percent in China to 84 percent in Bolivia; Cambo-dia, at 70 percent, ranked fourth after Bolivia, LaoP.D.R. and Uruguay.

Sources of Dollarization

By regional standards, the level of dollarization inCambodia in 2001, as measured by DR3, wasbetween the levels of its two comparable neighbors(Figure 6). In Vietnam, high inflation in the early1990s can be considered the main factor behind thesudden surge in dollarization, peaking at 41 percentin 1991. As a result of macroeconomic stabilization,dollarization receded somewhat in the followingyears, but rose again during the last half of thedecade, reaching more than 30 percent in 2001. Untilthe mid-1990s, dollarization in the Lao P.D.R. washigher than in Vietnam, but lower than in Cambodia.However, it has expanded remarkably since 1996 andhas now surpassed Cambodia. The main sources ofthe surge in dollarization in the Lao P.D.R. were theepisodes of high inflation (from December 1994 toDecember 1999) and a concomitant sharp deprecia-

tion of the domestic currency. However, since 1999,efforts at stabilizing the economy have resulted in adecrease in the inflation rate and dollarization hasleveled off, albeit at a high level (75 percent). Con-versely, the recent continued rise in dollarization inCambodia has clearly not resulted from high infla-tion, as inflation has all but disappeared since 1999.10

These findings are confirmed by a causality analysisbetween inflation and dollarization in Cambodia, LaoP.D.R. and Vietnam (Box 4).

Measuring Cash Dollars Circulating OutsideBanks in Cambodia

Balino, Bennett, and Borensztein (1999) contendthat "in general terms, dollarization is a response toeconomic instability and high inflation and to thedesire of domestic residents to diversify their asset

10The old CPI, used until the end of 2001, covered only the cap-ital city and suffered from a number of structural weaknesses, aswell as excessive sensitivity to seasonal fluctuations in foodprices. It is thus possible that it slightly underestimated the under-lying inflation. It was replaced in January 2002 by a new, updated,and expanded index.

9


Ill DOLLARIZATION IN CAMBODIA

| | ' , • • i I I . | I I I |J •

C a m b o d i a , Lao P.D.R., and V i e t n a m

. . - • ^ : H . : : ^ ^ ; ; ^ : r - - - : - r : ^

40-

30"

20"

Vietnam ••

10 I I I I I I I I

1994 96 98 2000

Source: IMF staff estimates.

portfolios." While in the early 1990s, economicinstability and high inflation certainly did prevailin Cambodia, these phenomena would likely nothave sufficed to cause such high doUarization, hadit not been for the sudden and massive inflow ofcash dollars in 1991-93. Given the sheer volumeof cash dollar inflows, we feel that the traditionaldoUarization ratios referred to earlier significantlyunderestimate the true degree of doUarization inCambodia, a characteristic that cannot be ignoredin economic analysis. We therefore endeavor toestimate cash dollars circulating in Cambodiaempirically.

At the outset, we would like to emphasize that ourempirical attempt at measuring cash dollars in circu-lation is hampered by severe data limitations. As inmany developing countries, some of the data usedare of mediocre quality, especially data on nationalaccounts. To our knowledge there are, however, noother statistical sources that would yield data of bet-ter quality. We recognize that our results rely heavilyon the data, and we are aware that a revised datasetcould result in changes in levels, but it is unlikelythat it would change dramatically the trend of thederived cash dollar time series. Accordingly, and asin related works, we emphasize that our resultsshould be considered only as rough estimates. Our

Figure 7. Dollars in Circulation Outside

(In billions of U.S. dollars)

3.5-

3.0

2.5

2.0

1.5

New coalition government

General elections .

Fighting ^^

1995 96 97 98 99 2000

Source: Authors' estimates.

intention is not to provide exact cash dollar time esti-mates, but rather a "baseline" for illustrative pur-poses that is helpful to assess the policy optionsavailable. The empirical results we find are probablyon the high end of the range of cash dollars in circu-lation, but we trust that they buttress the policy con-siderations and recommendations for Cambodia, laidout in Sections IV and V, in the context of very highdoUarization. The empirical model used to estimatecash dollar circulation in Cambodia is presented inthe Appendix.

According to our estimates, and as illustrated inFigure 7, cash dollars in circulation amounted toabout $1.2 billion in early 1995 and rose to $2.9 bil-lion at the beginning of 2001. The evolution of dol-lars in circulation can be divided into four phases: asteady increase from February 1995 until March1996; a relative stabilization from April 1996 toMarch 1998; a second overall, though erratic,increase until the end of 1998; and a relative stabi-lization from January 1999 to December 2000.Using dummy variables, the impact of major politi-cal events was tested but yielded no significantresults.

The estimated stock of cash dollars at the start of1995 seems broadly consistent with the infusion oflarge amounts of cash dollars since the mid-1980s,especially during the UNTAC period, for the pur-chase of local services and goods; remittancesand private transfers from abroad, which started in

0



Box 4. Causality Analysis Between Inflation and D ion in C L, Lao RD.R., and Vietnam

We examine the relationship between inflation anddollarization in three selected South-East Asian coun-tries (Cambodia, Lao P.D.R., and Vietnam) during thesecond half of the 1990s, using the Granger (1969)causality methodology. We use monthly data fromthe Cambodian authorities and from InternationalFinancial Statistics (IMF, 2001). We compute dollar-ization as the ratio of foreign currency deposits tobroad money, and inflation as changes in the monthlyconsumer price index. We use data from October1995 to August 2001 for Cambodia; from February1995 to September 2001 for Lao P.D.R.; and fromJanuary 1996 to July 2001 for Vietnam. All the vari-ables are specified as growth rates, that is, differencedonce.

To test the stationarity of the time series, integrationproperties are assessed using conventional unit roottests (augmented Dickey-Fuller statistics). The resultsof the unit root tests are summarized in the table below.All the time series are stationary and the order of inte-

Unit Root Tests

Granger Ca

Country

CambodiaLao RD.R.Vietnam

Null Hypothesis

Inflation does notGranger causedollarization

0.37382.5063'0.5097

Dollarizationdoes not Granger

cause inflation

1.01580.42000.8023

Country

CambodiaLao RD.R.Vietnam

Inflation

-1.6932"-2.54822

-2.31142

Dollarization

-3.670423

-I .97552

-2.97512 4

Note:The results were obtained using 12-month lags.

'Denotes significance at the 10 percent level.2Denotes significance at the 5 percent level,

includes deterministic elements (intercept and trend).4lncludes a deterministic element (intercept).

Note:The results were obtained using 12-month lags.1 Denotes significance at the 5 percent level.

gration is zero, so they arc 1(0) scries. For each timeseries, there is no unit root.

Having established the stationarity of all the variables,we perform Granger causality tests. The statisticsreported in the table above arc the conventional F-statistics of this method. The test shows that episodes ofhigh inflation in Lao RD.R. between early-1995 and theend of 1999 had an impact on the surge of dollarizationduring the same period. Conversely, in Cambodia and inVietnam, dollarization cannot be explained by inflationduring the period under review. In other words, inflationdid not predict dollarization in these two countries a yearlater. As we noted earlier, dollarization in Cambodiaseems to be the result, among other things, of a massiveexogenous shock during 1991-93. In Vietnam, high infla-tion in the early 1990s may explain the level of dollariza-tion then. Since 1997. other factors may have contributedto the increase in dollarization: spillovers from the Asiancrisis: real interest rate spreads in favor of foreign cur-rency deposits: and strong performance in exports.

1985; and the return of large numbers of refugeeswho brought cash with them. The steady increaseof cash dollars in 1995-96 can be explained by thestart of multilateral and bilateral aid disbursements($335 million in 1995 and $437 million in 1996),the return of foreign investment ($151 million in1995 and $294 million in 1996), and continuedlarge private transfers. According to the Council ofDevelopment for Cambodia, the country receivedabout $2.1 billion from bilateral donors alone duringthe period 1992-2000. However, as the politicalsituation, in the wider context of the Asian crisis,deteriorated in 1997 and the first half of 1998,cash dollars are estimated to have stabilized, butmassive capital flight seems not to have occurred, asone might have expected. During 1997 and early

1998, owing to political uncertainty, internationalaid inflows and foreign direct investment slowedand the amount of dollars circulating in the econ-omy stabilized at around $2.3 billion. After the gen-eral elections in July 1998, a new surge in cashdollars started, fueled by the spectacular increase inforeign direct investment in the garment industry.Between 1995 and 1998, some 165 garment facto-ries were opened and today they employ 160,000people who receive, in aggregate, an estimatedannual salary of $140 million in cash. Since early1999, the level of cash dollars has leveled off, as theunderlying factors described above seem to havestabilized.

If we measure currency substitution by the per-centage of the estimated foreign currency in circula-

11



Figure 8. Estimated Currency Substitution(In percent of total money circulating)

. ::mrFigure 9. Ratio of Dollars in Circulation toForeign Currency Deposits

96.5

96.0-

95.5

95.0-

94.5

94.0-

93.5L

1995 96 97 98 99 2000


. ' • • ' • • ; • • ; : • • ' : • • ; •

tion as a share of total currency in circulation(domestic and foreign currencies outside banks), wefind an average of 96 percent (Figure 8). Anecdotalevidence cited by Liang and others (2000) andMarciniak and Sa (2002) suggested that foreigncurrency cash holdings in Cambodia amounted to85-95 percent of total currency in circulation. Ourresults are consistent with those earlier estimates.

The high ratio of dollars in circulation to GDP andto foreign currency deposits can be explained by sev-eral reasons (Table 1). First, the lack of public confi-dence in the domestic banking system anduncertainty about the future lead people to hold highamounts of dollars in cash. Second, the lack of amodern payments system (electronic payments andcredit cards) and the limited use of checks promote

Table I. Ratio of Dollars Circulating inthe Economy

1995

Dollars/GDP 0.57

Dollars/Foreign

currency deposits 14.1

1996

0.72

12.6

1997

0.76

11.8

Source:Authors' estimates.

Note:Yearly averages of monthly estimates

1998

1.00

15.8

1999

1.02

13.9

2000

1.01

10.2

1995 96 97 98 99 2000


the use of cash dollars in the economy. Third, banksare overly liquid and a number of them decline tak-ing small deposits.11 Fourth, there are no financialinstitutions outside major cities, except some micro-finance institutions. Fifth, it is conceivable that largeamounts of cash dollars circulate in the economy asa result of smuggling and of illegal activities, whichtypically transact in cash using a major internationalcurrency.

Figure 9 illustrates the sensitivity of agents' pref-erence for holding cash dollars versus dollar depositsin the banking system in times of political uncer-tainty, in a general context of increasing dollariza-tion. During 1995:Q3-1997:Q1, as economic reformtook hold and financial intermediation deepened, theratio of cash dollars over dollar deposits declinedfrom about 14 to about 9Vi. During most of 1997,owing to political strife, and most of 1998, owing touncertainty linked to the upcoming general elections,the ratio surged and, at its peak in August 1998,reached 19. Starting from the effective operation ofthe coalition government in November 1998, theratio steadily decreased to reach about 9, its lowestvalue in the period under review.

We compute a new dollarization ratio, DR4, similarto DR3, but including in the numerator and in thedenominator our estimates for cash dollars in circula-tion. DR4 is thus defined as the ratio of residents' for-eign currency deposits plus cash dollars in circulation

uDuring 1998-2000, $530 million worth of banknotes weredeposited overseas by Cambodian commercial banks.

12


Measuring DoUarization in Cambodia

Figure 10. Estimated DoUarization(In percent of broad money)

96-

95-

94-

93-

92

91 •

9011995 96 97 98 99 2000


to total broad money (including cash dollars in circu-lation). The result of this computation, shown inFigure 10, is quite remarkable. We find that thedegree of doUarization in Cambodia, as measured byDR4, has been stable since early 1995, in the range of93 to 95 percent. This validates our earlier assertionthat Cambodia became very highly dollarized duringthe period 1991-95 and has stayed so ever since.While the economy has expanded and financial inter-mediation has deepened since 1995, and while therehave been large fluctuations in the composition ofagents' dollar assets, as discussed earlier, the overalldegree of doUarization has remained stable. Thus dol-larization has persisted despite macroeconomic stabi-lization, resumption of growth, and relatively stablepublic finances. This persistence—or hysteresis—hasalso been noted in other cases of high doUarization(e.g., Mueller, 1994). The continued lack of confi-dence in the national currency, owing to historicalreasons, could be a key factor in explaining this phe-nomenon. Another reason could stem from the highcosts of de-dollarization, in particular the cost associ-ated with the physical replacement of the large vol-ume of dollar banknotes.

13


IV Costs and Benefits of High Dollarizationin Cambodia

Ahigh degree of dollarization can yield both costsand benefits. This section examines how dollar-

ization can help or hurt an economy.

Drawbacks of High Dollarization

Loss of Seigniorage

A main cost of high dollarization in Cambodia isthe loss of seigniorage revenue for the government.Box 5 presents recent methods used in the literatureto measure seigniorage.

In Cambodia, commercial banks' required reservesat the National Bank of Cambodia are not remuner-ated, so iR is equal to zero and seigniorage is thusequal to iAA. Interest income series from the NationalBank of Cambodia overseas deposits are availablesince 1995 and were used to compute seigniorage.Estimates of seigniorage are shown in Table 2, inabsolute and relative terms.12 Both estimation meth-ods show a relatively low yearly seigniorage in termsof GDP, but yield a higher seigniorage in terms ofgovernment revenue, IVA percent a year on aver-age—however, that is chiefly a consequence of thevery low level of government revenue. Thus, with theexception of 1998, when there was a large expansionin domestic currency to finance the budget deficit,seigniorage associated with dollarization since 1994has been relatively moderate.

In case the authorities wanted to move to full dol-larization, such a move would involve additionalseigniorage losses. It would involve an initial "pur-chase cost" plus additional future annual seignioragelosses stemming from a larger money supply in dol-lars. To adopt the dollar as legal tender and withdrawthe domestic currency entirely from circulation,13

the monetary authorities would have to purchase the

12A similar table is presented by Liang and others (2000),although time coverage and results differ slightly for the reserve-money method. For the central-bank-profit method, we use actualinterest earnings data.

13Such a move would probably not prevent the Thai baht fromcontinuing to circulate in some border areas.

Box 5. Measuring Seigniorage

Berg and Borensztein (2000b) state that "the ancientconcept of seigniorage is a government's profit fromissuing coins for a cost less than face value. This con-cept is also relevant for paper currencies. Neglectingthe minor cost of printing money, seigniorage is sim-ply the increase in the volume of domestic currency:'In other words, seigniorage represents the profitaccruing to the monetary authorities from their rightto issue the legal tender. While there is a generalagreement on the concept of seigniorage, methods ofestimating it empirically vary among economists.

According to Berg and Borensztein (2000a),seigniorage can be measured by two alternativemethods: the reserve-money method and the central-bank-profit method. These two methods ignore thecosts of currency production and are also called the"flow cost" approach.

According to the reserve-money method, the mone-tary authorities earn seigniorage from the flow of newcurrency emitted into circulation, corresponding to thedemand of money over time to conduct transactions.The reserve-money method measures seigniorage asthe increase in reserve money in a given period:

S = AR/P = (AR/R) (R/P) = hm, (1)

where R is the nominal stock of reserve money, P isthe price level, h is the growth rate of reserve money,and m is the real stock of reserve money. This methodassumes that there is no interest paid on some com-ponents of reserve money, such as commercialbanks' reserves deposited at the central bank.

The central-bank-profit method measures seig-niorage as the difference between what the centralbank earns on its assets that back reserve money, andthe interest that it pays to the holders of reservemoney (typically commercial banks, which holdreserve deposits at the central bank). Seigniorage isthus equal to:

S' = iM - PR, (2)

where iA is the rate of interest earned by the centralbank on its assets, A is the nominal stock of centralbank assets, iR is the rate of interest that the centralbank pays on reserve money, and R is the nominalstock of reserve money.

14


Drawbacks of High Dollarization

Table 2. Estimates of Seigniorage

1994 1995 1996 1997 1998 1999 2000 2001

Reserve-Money MethodChange in reserve money (billions of riels)

In percent of GDPIn percent of government revenue

Central-Bank-Profit MethodInterest income (billions of riels)

In percent of GDPIn percent of government revenue

57.80.9

19.9

28.60.44.8

18.60.32.9

135.31.6

21.0

28.90.33.9

95.51.0

12.7

47.40.45.4

257.32.4

29.2

58.70.66.2

127.3I.I

13.5

67.60.65.1

231.11.9

17.3

114.70.98.0

198.61.5

13.0

88.40.75.8

Sources: National Bank of Cambodia; and authors' estimates.

stock of domestic currency held by the public (andbanks), effectively returning to the holders seig-niorage accumulated over time. Fischer (1982)measures this initial purchase cost of full dollariza-tion by expressing domestic currency in circulation.as percent of GDP. In the case of Cambodia, thiswould correspond to 4 percent of the estimated 2001GDP, and using the end-2001 exchange rate, thecost would be $139 million, or about one-third ofnet official reserves. The additional annual seignior-age losses associated with full dollarization couldbe computed by using one of the two methodsexplained earlier.

Lower Official International Reserves

Another potential drawback associated with highdollarization in Cambodia stems from lower officialinternational reserves. In the presence of an effectivenational currency, agents who hold foreign currency(e.g., exporters and foreign aid recipients) need toacquire national currency to do business in the coun-try. Part of the foreign currency sold by these agentsto purchase national currency is likely to end up in thecoffers of the central bank, thereby boostingthe country's international reserves. Conversely,those agents who would like to purchase foreigncurrency to do business abroad have to buy it from themarket. If the central bank refrains from supplyingforeign currency to the exchange market, foreign cur-rency outflows from official reserves are reduced.Accordingly, although Cambodia's net internationalreserves have steadily increased since the early 1990s(Figure 11) and are now equivalent to about threemonths of imports of goods and services, they couldbe still higher—including accrued interest earnings—were the country not so highly dollarized.

We would argue, however, that in a highly dollar-ized economy, official reserves do not play the same

role (i.e., increasing confidence in the national cur-rency and weathering temporary external shocks) asin a country with a sole national currency. In a highlydollarized economy, the external credibility of thenational currency is usually already largely compro-mised. And since the U.S. dollar is an international"reserves money," economic agents who hold a largeamount of dollars have the means to react to a tem-porary external shock. Thus, the main difference inthe status of "reserves" seems to be that these"reserves" are in the hands of agents, rather than inthose of the monetary authorities. We would suggestthat in an open economy like that of Cambodia thisis not necessarily undesirable—as far as minimizing

(In millions of U.S. dollars)

600-

500-

400

300

200

100

Gross official reserves

Net official reserves

ill ili

11995 96 97 98 99 2000 01


15


IV COSTS AND BENEFITS OF HIGH DOLLARIZATION IN CAMBODIA

output loss is concerned—because in such an econ-omy the markets are the economic driving force.14

We recognize, however, that if the current high lev-els of dollar inflows were to abate, and were nolonger available to finance the current accountdeficit, the importance of reserves would grow.

Loss of an Effective Monetary Policy

In a highly dollarized economy, the foreign cur-rency component of broad money cannot be directlyinfluenced by the monetary authorities. Money sup-ply in the economy is not determined by the mone-tary authorities but by the behavior of agents holdingboth foreign- and domestic-currency-denominatedassets, including cash. As money supply in the econ-omy becomes endogenous, the authorities may notbe in a position to fight inflation by tighteningdomestic money supply in an appropriate manner.Based on empirical evidence, Hoffmaister and Vegh(1995) assert that in Uruguay (a highly dollarizedeconomy), dollarization may have severely hinderedthe effectiveness of monetary policy. High dollariza-tion in Cambodia thus limits the effectiveness of theNational Bank's monetary policy in that its opera-tions in riels have almost no impact on overall mon-etary developments.15

While in a highly dollarized economy, the mone-tary authorities cannot influence money supplydirectly, they can control other related variables,such as base money and the reserve requirement rateof banks. In theory, these monetary policy toolsshould allow them to control domestic money supplyindirectly. In Cambodia, however, as financial inter-mediation is limited and conducted almost entirely inforeign currency, the National Bank's ability to con-trol base money is severely limited. The NationalBank retains changes in reserve requirements as apotential tool, but this regulatory instrument cannotbe used frequently for the sake of financial stability.In fact, since December 1993, the National Bank haschanged the reserve requirement only once, in Janu-ary 1998, raising it from 5 percent to 8 percent.16 Asanother potential monetary instrument, a refinancingfacility was introduced in Cambodia in June 1994.The only assets eligible for this facility are trade billsdenominated in riels. The refinancing facility allows

Table 3. Interest Rates on Deposits andLoans, February 2002(In percent p>~

Interest rate on depositsDemand depositsTime deposits1

Nominal interest rateon loans2

Real interest rate on loans3

Riels

3.28.6

18.415.7

U.S.Dollars

2.14.9

16.714.0

OtherForeign

Currencies

1.31.5

18.015.3

Source: National Bank of Cambodia.112-month deposits.2Lending rate to private enterprises, including small businesses.3The real interest rate is computed as the difference between

the nominal interest rate and the inflation rate—the latter is com-

puted as the average of the CPI during the three-month period

ending in February 2002 over the same period a year earlier.

the lender to redeem the bills before maturity at adiscount of 70 percent of face value. However, nocommercial bank has ever used this facility.

With regard to interest rate policy, commercialbanks are free to set their deposit and lending rates.Since the National Bank does not refinance banks, itdoes not influence interest rates and therefore cannotuse interest rate policy as an effective monetaryinstrument. The structure of deposit and lending ratesis shown in Table 3. High interest rate spreads arecharacteristic of lesser developed economies, but thepresence of dollarization usually enhances the credi-bility of the exchange rate and reduces inflation, andthus real interest rates tend to contract. In Cambodia,inflation has been very low and sometimes even neg-ative since 1999, yet the real interest rate on dollar-denominated loans remains at 14 percent a year.17

The sizable spread reflects the high country risk; thehigh costs of banking (legal uncertainty, litigationcosts, and default rates); and the lack of aggressivefinancial intermediation, owing to few secure lendingopportunities. In Cambodia, dollarization does notsuffice to offset developmental constraints.

14Current transactions in Cambodia are free of restrictions, andthe authorities adopted IMF Article VIII status on January 1, 2002.

15We do not consider the hypothetical issues related to a centralbank conducting monetary policy with dollar-denominated instru-ments or using its foreign reserves and correspondent accounts.

16Reserve requirements on riel and foreign currency deposits atcommercial banks are payable in riels and in foreign currency,respectively. Since most commercial banks are fully dollarized,they meet their reserve requirements in dollars, except the ForeignTrade Bank, which meets them in riels.

Loss of an Effective Exchange Rate Policy

High dollarization implies losing flexibility in theexchange rate policy, as it obviates exchange rate

17As the banking system in Cambodia is almost fully dollarized,interest rates for transactions in riels are largely irrelevant for ana-lytical purposes.

16


Benefits of High Dollarization

F *Figure 12. Spread Between Official andMarket Exchange Rates(In percent)

1993 95 97


99 2001

(End of period)

4500

2500

2000'1993 95 97 99 2001

Source: IMF, International Financial Statistics.

adjustments in response to external shocks. With"rigid" exchange rate arrangements, the monetaryauthorities cannot manipulate the exchange rate inorder to spur the real sector, because changes in theexchange rate of the domestic currency are largelyirrelevant in the face of the predominant role offoreign-currency-denominated assets in the econ-omy. The same holds true for market-inducedexchange rate depreciation in the case of "flexible"exchange rate arrangements. In the presence of anexternal shock, highly dollarized economies tend toadjust through the goods and factors market, with thehelp of the financial markets, if they are sufficientlydeveloped. However, as discussed earlier, somecountry experiences show that the lack of flexibilityin the exchange rate policy may be beneficial ratherthan detrimental.

Between 1994 and 1999, the National Bank ofCambodia pursued a flexible exchange rate policy,keeping the spread between the official and the mar-ket rates below 1 percent, except in a few exceptionalperiods (Figure 12). Most of the time since late 1999,the National Bank has further kept the spread at onlyVi of 1 percent, and it intends to eliminate the spreadentirely.

Since the National Bank of Cambodia cannotemploy most monetary and exchange rate instru-ments, it lacks the tools to conduct monetary andexchange rate policies effectively. Thus, in Cambo-dia, the brunt of macroeconomic adjustment falls onfiscal policy.


Isolation from the Effects of Exchange RateFluctuations

High dollarization provides some protectionagainst exchange rate risks, as a change in theexchange rate bears only on a small part of moneysupply (i.e., the domestic component thereof) andfinancial assets (denominated in domestic currency).Typically, in a highly dollarized economy the bulk oftrade-related and of large financial transactions aresettled in dollars, whereas the national currency ismainly used for dealing with small-scale nontrad-ables. As a result, in the case of an exchange ratedevaluation/depreciation, the pass-though effect ofhigher import prices on inflation is limited and pricesof nontradables settled in local currency are notdirectly affected. The Asian crisis illustrated dramat-ically this isolation effect. From July 1997 to Janu-ary 1998, the Thai baht depreciated by 71 percentagainst the dollar. During the same period, the rieldepreciated by 23 percent (Figure 13). The inflationperformance of the two countries shows a similardisparity in favor of Cambodia, and this in spite ofthe July 1997 political crisis in Cambodia, whichcertainly did contribute to the riel's depreciation andto the rise in inflation.18

18However, if international bank exposure to Cambodia hadbeen high relative to GDP, then withdrawal of such funding mighthave had more pronounced contagion-like effects.

17


IV COSTS AND BENEFITS OF HIGH DOLLARIZATION IN CAMBODIA

Figure 14. Riels per Thai Baht and100 Vietnamese Dong(End of period)

180-

160

140-

120

100

80-

60^

M

Thai baht 11(left scale) ]\

1

|\ Vietnamese dongl \ (right scale)

-40

-35

-30

-25

-20

^151993 95 97 99

Source: IMF, International Financial Statistics.

2001

Figure IS. Effective Exchange Rates

60 •-1993 95 97 99 2001

Source: IMF staff estimates.

Since 1999, the exchange rate of the riel againstthe dollar has been relatively stable, and both ofthem strengthened against the Thai baht and theVietnamese dong (Figure 14). The largest market forCambodian exports is the United States and the mainsources of Cambodian imports—excluding importsfor reexport—are Thailand and Vietnam. The recentappreciation of the riel has allowed Cambodia tobenefit from an improvement in its terms of trade.These evolutions could also explain the low level ofinflation in Cambodia since the end of 1999.

The real effective exchange rate has increasedregularly—except for the turbulence in 1998—bysome 15 percent since early 1995, but this increaseseems not to have hurt Cambodia's exports, presum-ably owing to the particular nature of the garmentindustry's external market (Figure 15).

Financial Re-Intermediation

Another benefit of dollarization in Cambodia hasbeen a gradual financial deepening of the bankingsystem. In economies that have experienced periodsof high inflation and an unstable macroeconomic sit-uation, residents tend to become reluctant to holddeposits in the domestic banking system. Dollariza-tion, through foreign currency deposits held indomestic banks, encourages agents to use bank ser-vices rather than to hold idle cash balances. Whenmacroeconomic stability is restored in a dollarizedenvironment, agents may have more confidence in

the banking system and may be more willing toreturn to domestic intermediaries if they can holddollar-denominated assets. More specifically, inCambodia the dollar deposit growth rate was verylow in 1998, at about 1 percent, but rose on averageby 25 percent during 1999-2001, while the averageriel deposit growth rate was about 32 percent duringthe latter period.

Economic and Financial Integration

Dollarization may contribute to greater economicand financial integration with the rest of the world.The use of a foreign currency, especially the dollar,which is the most widely used currency in interna-tional trade, reduces the transactions costs of pur-chasing international currency. The more a country'strade and financial flows are integrated with coun-tries using the dollar, the greater will be the gainsfrom reducing exchange risks. Nonetheless, exchangerisks with other currency zones remain. Mundell(1961) analyzes optimal currency areas and providesa good analysis of trade benefits associated with dol-larization. Countries highly integrated in terms oftrade and factor mobility can benefit from using thesame currency. More recently, Rose (2000) and Roseand Frankel (2000) note that a country's trade canincrease significantly if it belongs to a monetaryunion, or similarly, if its economy is dollarized.Edwards (2001) suggests that benefits from addedtrade integration can more than offset the costs ofdollarization, including the loss of seigniorage. Cam-

18



bodia, with its high level of dollarization, is likely tohave benefited from enhanced trade. Between 1995and 2001, the share of garment exports in Cambo-dia's total exports surged from nil to 83 percent, andsince 1998 the United States has become the largestmarket for Cambodian garment exports (71 percentof total garment exports). While other factors, suchas an export license system for the United States,have certainly played an essential role in that expan-sion, dollarization has also contributed to it, and islikely to be critical in diversifying exports further inthe future. However, dollarization hinders price/wage flexibility and therefore requires sufficient pro-ductivity gains in the export sector to keep abreast ofcompetitors.

Dollarization holds the promise of macro-economic stability for foreign investors and theelimination of the domestic exchange rate risk,especially with regard to the repatriation of profits.Dollarization also tends to limit the country's expo-sure to currency crises and to contagion episodes,as illustrated by the case of Cambodia during1997-98. In the medium term, dollarization can alsopromote the development of domestic financialmarkets, as the use of dollars facilitates the integra-tion of the domestic market into the rest of theworld, owing to lower costs of international finan-cial transactions.

ment's ability to fuel inflation through monetized fis-cal deficits. Most independent central banks willrefrain from unsustainable financing of budgetdeficits through money creation, since such financ-ing is inflationary and likely to lead to even largerdeficits. When monetization of the deficit is ham-pered, as in a highly dollarized economy where themonetary authorities cannot emit dollars, lest theydeplete their reserves, fiscal discipline will likelyresult (Sargent, 1986). Moreover, dollarization, byremoving the seigniorage as a source of easy rev-enue, leads also to tighter fiscal policy.

In the early 1990s, Cambodia experienced highinflation resulting from monetized budget deficits.Since then, high dollarization has limited the scopefor inflationary financing of fiscal deficits andassisted in building financial discipline.19 In addi-tion, Article 24 of the Law on the Organization andConduct of the National Bank of Cambodia (1996)forbids the National Bank to lend money to the gov-ernment. Since then the National Bank has done soonly once in 1998, at a time of shortfalls in revenuecollection linked to political upheaval. Henceforth,the government is committed to eschew central bankfinancing of the budget. This policy stance hasallowed Cambodia to minimize pressures on pricesand on the exchange rate.

Fiscal Discipline

Another benefit of dollarization is fiscal disci-pline, as dollarization greatly reduces the govern-

19Technically, the government could emit riels and monetize afiscal deficit, but given the narrow riel base, the government wouldrisk triggering inflation and a depreciation of the currency.

19


V Implications of High Dollarization forMacroeconomic Policy Design

The question arises as to what are the possiblemacroeconomic policy options for the Cambo-

dian authorities in the context of very high dollariza-tion. More specifically, the question is whether theauthorities would be better off continuing with thepresent policy stance, trying to de-dollarize the econ-omy, or choosing full dollarization. We firstendeavor to shed some light on these options byreviewing the role of commercial banks and of thecentral government budget. We then explore issueslinked to a possible currency board arrangement.Finally, we indicate our preferred course of action.

Dollarization and the Banking System

In spite of notable progress in the last few years(Box 6), the banking system remains undevelopedand financial intermediation is shallow. The progres-sive establishment of a modern payments systemwill help to foster financial intermediation. The con-templated financial laws and payments facilities willbe neutral with regard to the currency used (riels ordollars) and therefore will have no direct bearing onthe issue of dollarization.

The monetary authorities could in principle pro-mote de-dollarization through the banking system.The lack of a lender of last resort, in conjunctionwith the fact that Cambodia is still a largely cash-based economy, forces banks to hold excessive dol-lar balances and increases their vulnerability. Thefirst and most dramatic move for de-dollarizationwould be for the National Bank to provide bank refi-nancing in riels, for operations conducted in riels,while the monetary authorities would continue torefrain from refinancing banks in dollars. This couldentice banks into operations in riels. This would rep-resent an important move away from the NationalBank's present policy stance under which de facto norefinancing is available. This initiative could in the-ory help foster credit operations in riels, and hencede-dollarization, but we would argue that at the cur-rent juncture, such a move would be largely ineffec-tive, as other major conditions are yet not ripe for asignificant expansion in bank credit (for either riel-

or dollar-denominated loans). Banks are likely tocontinue to shy away from sizable loan operations,as long as they cannot secure adequate collateral.Even if they could, repossessing such collateralremains currently highly uncertain, considering theweaknesses in Cambodia's legislation and judicialsystem. Accordingly, as long as the inadequacy ofloan-guaranty legislation and enforcement has notbeen effectively addressed, offering refinancingwould be ineffective, while it would send out the"wrong" signal from the National Bank at a timewhen the restructuring of the banking system is stillunder way and the National Bank is building a trackrecord of encouraging banking discipline.

The second way of promoting de-dollarizationthrough the banking system could be accomplishedby rendering operations in riels more attractive secu-rity- and cost-wise than operations in dollars. How-ever, under the current situation of almost fulldollarization of the banking system, and given theprogress toward an efficient, largely electronic pay-ments system that is currency neutral, there seems tobe no immediate scope for achieving such a result.Conversely, were the authorities to opt for full dol-larization of the economy, the banking system wouldbe able to accommodate such a decision quickly andefficiently.

Another approach to fostering de-dollarizationcould be the development of financial instrumentsdenominated in riels (e.g., treasury bills). Suchinstruments would likely need to offer higher yieldsthan deposits with strengthened commercial banks(both in riels and in dollars), as the government willneed to build up public confidence. However, wenote that deposits in riels already carry higher inter-est rates than deposits in dollars, as shown in Table 3,yet deposits in riels remain modest compared withdeposits in dollars. For treasury bills to becomeattractive, they would thus need to offer very highremuneration to compensate for both sovereigndefault and exchange rate risks. This in turn couldjeopardize the fragile budget stance. Such instru-ments would also require the setting up of an activefinancial market for the sake of liquidity, but such adevelopment can be considered only in the medium

20


Dollarization and the Banking System

Box 6. Banking Reform in Cambodia

Following the adoption of the Banking and FinancialInstitutions Law in November 1999, the banking systemhas undergone extensive restructuring. At the start of theprocess, the banking system consisted of 31 banks:22 private commercial banks, 7 foreign bank branches,and 2 state-owned banks. As a result of the first phase ofa mandatory relicensing procedure in 2000, the NationalBank of Cambodia closed four banks in July 2000 and,after carefully evaluating the viability of the remainingbanks, it closed another seven banks in December 2000.In March 2001, one bank elected for voluntary liquida-tion, totaling 12 closed banks. Of these, eight were vol-untarily liquidated and four were compelled to go intoliquidation. Six banks were fully relicensed, including aspecialized state-owned bank, and 13 banks were reli-censed conditionally, including another state-ownedbank. The banks relicensed conditionally were requiredto take corrective actions, agreed in memoranda ofunderstanding signed with the National Bank of Cam-bodia. Typically, these banks had to fulfill a number ofprudential requirements, including the injection of sup-plementary capital to meet the $13 million minimumcapital requirement. At the end of 2001, eight additionalbanks fully paid up their required capital. By the end ofMarch 2002, two more banks met their capital require-ment, one bank was voluntarily wound up, one bank wascompelled to go into liquidation, and one bank was con-verted into a specialized bank, with a lower capitalrequirement. In addition, two specialized banks werecreated in October 2000 and in March 2001, respec-tively. Thus, currently 19 banks operate in Cambodia,down from the initial 31 banks.

The use of checks is currently limited to transactionscarried out by the government and large entities. Aclearinghouse for checks denominated in riel has beenin operation since 1995 (see the table). A new impetuswas given to check circulation with the opening of a

second clearinghouse for checks denominated in dol-lars on January 2, 2001. The clearing of such checkshas been active from the beginning. During 2001, thetotal value of checks in dollars cleared amounted to$643 million, versus only $285 million for checkscleared in riel during 1995-2001. These figures furtherillustrate the predominance of dollars in the bankingsystem. Both clearinghouses operate under strict sol-vency rules, as debit balances at the end of a clearingsession need to be paid up entirely by the bank con-cerned before the end of the day. The National Bank ofCambodia manages and supervises the clearinghousesfor a nominal membership fee, but does not assume anyrisk associated with their operations.

Before setting up the dollar clearinghouse under theaegis of the National Bank of Cambodia, the authoritiesconsidered whether this move would promote dollariza-tion further, or merely facilitate inter-bank operations.Since bilateral clearing of such checks between bankshad existed for several years, the authorities eventuallyconcluded that making the process more effective wouldprovide efficiency gains while being fairly neutral withregard to the level of existing dollarization. In their opin-ion, the new dollar clearinghouse buttresses the pay-ments system rather than promotes dollarization. In thenear future, another important step for the establishmentof a modern payments system will be the adoption of aset of new laws pertaining to the conduct of financialtransactions. The Law on Negotiable Instruments andPayment Transactions, prepared with technical assis-tance from the IMF. is in the process of being adoptedby the National Assembly. The enactment of this law,coupled with the introduction of enhanced paymentsfacilities, such as electronic payments, electronic corre-spondent accounts and credit transfers, and the use ofcredit cards, will contribute to the development of finan-cial services offered by the banking system.

Checks Cleared

Year CurrencyNumber of

Checks

Total Value of Checks

Millions ofBillions of riels U.S. dollars

Average Value per Check

Millions of riels U.S. dollars |

19951996199719981999200020012001

RielRielRielRielRielRielRiel

Dollar

1,473893692619723

1,2401,436

85,007

183.7148.688.660.688.5151.3177.2

Not applicable

75.056.630.116.223.239.244.9

642.9

124.7166.4128.098.0

122.4122.0140.9

Not applicable

50,88263,41443,45626,14632,09831,63535,6827,563


21


Figure 16. Co-Currency Circ

( Balance of payments Ainflows in dollars

(transfers, tourist receipts,foreign direct investments, etc.)J

Aid disbursementand fees/taxespaid in dollars

Taxes: Riel (80 percent); Dollar (20 percent)

Overseasexpendituresin dollars and

externallyfinanced capital

expenditure

Goods and ServicesPrivate sector

(individuals andRiel

(wages, salaries, paymentsfor goods and services)

Riel

State-ownedenterprises

Electricite du CambodgeRegie des eaux

Purchase/Sale of Riels

National Bankof Cambodia

(NBC)Purchase of Dollars

Net officialreserves

Deposit/Sale/Purchase ofDollars

Overseas dollardeposits, exports

of dollar bank notes(via the NBC)

Sources: IMF (1997); revised by Authors.

term, as the required legislation is only at an earlystage of preparation.

Execution of Budget

Co-currency circulation in Cambodia is fairlycomplex, as illustrated schematically in Figure 16.On the one hand, foreign currency enters the countrymainly through two broad channels: official transfersto government accounts (transiting through theNational Bank), and current and capital transactionsin the balance of payments, of which some examplesare tourism receipts, private transfers, and foreigninvestment. These inflows mainly fuel cash foreigncurrency circulation, foreign currency deposits inand outside Cambodia, and official reserves.

On the other hand, the central government budgetplays a crucial role in the circulation of riels. On therevenue side, the National Treasury collects taxes inriels (some 97 percent of tax revenue in 2001),including taxes linked to imports (Table 4). Con-versely, three-fourths of nontax revenue was paid indollars, essentially collected by the Foreign Cur-rency Management Unit at the Ministry of Economyand Finance. Official aid and budget support are paidin foreign currency, also collected by the same Unit.Thus, in 2001 about three-fourths of total revenuewas collected in riels. On the expenditure side, about86 percent of current expenditure was paid in riels(mainly in cash, but also in the form of "offsets"),while two-thirds of capital expenditure was dis-bursed in dollars, owing to the fact that 71 percent ofsuch expenditure was externally financed. In sum,

V IMPLECATION OFHIGH DOLLARZATION FOR MACROECONOMIC POLICY DESIGN

22


Consideration of a Possible Currency Board Arrangement

Table 4. Budget Execution in ForeignCurrency(In percent)

2000 2001

Revenue

Total revenue in foreign currency/total revenue

Tax revenue in foreign currency/tax revenue

Nontax revenue in foreign currency/

25.3 23.2

4.4 2.6

nontax revenue

Expenditure

Total expenditure in foreign currency/total expenditure

Current expenditure in foreigncurrency/current expenditure

Of which: Salaries paid in foreigncurrency/salaries

Capital expenditure in foreigncurrency/capital expenditure

65.0 73.5

22.6 25.9

15.4 14.3

4.0 7.0

43.9 67.4

Source: National Treasury of Cambodia.

about three-fourths of total expenditure was paid inriels in 2001. Riels injected into the economy by theTreasury are used by agents for their domesticneeds—chiefly tax payments, payments to twomajor utility companies in Phnom Penh, and smalldomestic transactions. Since there has been a currentbudget surplus and a negative or no domestic financ-ing for the last three years,20 the Treasury serves asthe main engine for circulating riels in the country.21

The National Bank of Cambodia also injects rielsinto circulation occasionally by buying dollars in theforeign exchange market, but the amounts concernedare relatively small.

The Treasury has an impact on the level of dollar-ization. The more it injects riels into the privatesector through its expenditure, the more it con-tributes to the stabilization of doUarization or even tode-dollarization. Conversely, the more the Treasuryspends in dollars, the more doUarization is boosted.In this regard, the budgets in 2000 and 2001 showthat doUarization of the national budget executionincreased notably on the expenditure side. If the gov-ernment wanted to promote de-dollarization, itshould reverse this trend. On the revenue side, thegovernment would need to initiate legislation requir-ing that all payments to the budget be made in riels.

Taxpayers would then have to purchase riels fromcommercial banks/money changers or from theNational Bank of Cambodia, which would result inan increase in riel circulation. Alternatively, the cen-tral bank law could be effectively enforced withregard to the relations between the National Bank ofCambodia and the Ministry of Economy andFinance. According to the law (Article 21), govern-ment receipts in foreign currency should be immedi-ately sold to the National Bank of Cambodia at theofficial exchange rate, with the understanding thatthe National Bank would provide the Treasury withforeign currency against riels on demand to meet for-eign expenditure requirements. However, in practice,budget revenue in foreign currency is deposited inthe government's accounts at the National Bank ofCambodia and the Foreign Trade Bank in dollar-denominated accounts.22 These accounts are used forexpenditure in foreign currency. On the expenditureside, full budget execution in riels would allow Cam-bodia to stabilize (or even to reduce) doUarization.However, we note that the size of the budget com-pared with the economy is relatively small (about 12percent of GDP), and since the country is highly dol-larized, the impact of the budget being entirely exe-cuted in domestic currency would, in all likelihood,have a limited impact on the level of doUarization.Yet the shift from three-fourths of a budget executedin riels to a budget entirely executed in riels wouldbe the right step for promoting de-dollarization andcould be done progressively but fairly rapidly, with-out upsetting the current state of economic affairs.

Consideration of a Possible CurrencyBoard Arrangement

A radical move toward de-dollarization andrestoration of confidence in the domestic currency inCambodia could be the adoption of a currency board.A currency board arrangement is the strongest formof a pegged exchange rate system. It is a monetaryinstitution that only issues currency that is fullybacked by foreign assets at a fixed exchange rate.The exchange rate is fixed not by policy but by law,and the foreign currency backing rule ensures thatthe currency board will not exhaust its reserves tomaintain the peg. In Cambodia, the natural choicefor the "anchor" currency would be the dollar. A cur-rency board arrangement and very high doUarizationpresent some similar features, but differencesbetween them are important. First, doUarizationimplies the loss of seigniorage for the government,

20Excluding exchange rate adjustments and outstanding operations.21 Riels in circulation have been relatively stable since end-1999.

22They are, however, included in the calculation of officialreserves.


Box 7. Currency Board Arrangement Versus National Bank of Cambodia Policies

Sonic of the differences and similarities between policies of the National Bank of Cambodia and of a possible cur-rency board arrangement are highlighted below.

"Orthodox" CurrencyBoard Arrangement NBC Policies

Institutional setup

Foreign exchange policy

Monetary policy

Relations with thegovernment

Relations withcommercial banks

1 Independence of the Currency Board1 Audited accounts publicly available1 Open economy

• Guaranteed peg vis-a-vis an anchorcurrency

• International reserves to backdomestic currency in circulation anddeposits at the pegged exchange rate

• Free convertibility of nationalcurrency into anchor currency

• Banks are free to set interest rates1 No issuance of Currency Board paper

• No monetary financing of the budget• No open market operations by theCurrency Board, neither in thegovernment bond issue market nor inthe secondary market

• No commercial bank refinancing1 No guarantee to the banking system• Strong banking supervision• Services provided for promotingfinancial intermediation but noguarantee

• NBC is not fully independent, in practice• NBC audited accounts published since 2001• IMF Article VIII adopted from

January 1, 2002; membership in ASEAN;WTO accession well underway

• Managed float aiming at a very stableexchange rate vis-a-vis the U.S. dollar

• NBC's net international reserves wereequal to about three times dels incirculation and riel deposits at end-2001

• Limited convertibility for selected state-owned enterprises only, in practice

• Banks are free to set interest rates; theNBC's refinance rate is only used forrepayments from the Ministry of Economyand Finance to the NBC

• No NBC paper issued (with one recenttemporary exception to recapitalize theForeign Trade Bank)

• No monetary financing of the budget(actually negative financing since 1999)

• NBC does not hold or trade any public orprivate securities or bonds

• No commercial bank refinancing• No NBC guarantee to the banking system,

as illustrated by closed banks• NBC banking supervision needs

substantial strengthening• Clearinghouse for checks in riels and

dollars sponsored by NBC but no riskassumed

whereas a currency board arrangement does not.Second, one of dollarization's key distinguishingfeatures is that it tends to be permanent. In additionto hysteresis, one of the largest benefits claimedfrom dollarization is its credibility, precisely becauseit is nearly irreversible. Some authors argue that it ismore difficult and costly to reverse dollarization thanto modify or abandon a currency board. In manycases, country experience with "orthodox" currencyboard arrangements has been positive (lower infla-tion, higher growth, and increased foreign invest-ment) but a number of currency boards had to be

abandoned under economic duress, notably in thecase of inappropriate fiscal and debt policies, remissof a pegged exchange rate.

Arguably, Cambodia's recent monetary andexchange policies could be characterized as akin to ade facto currency board arrangement. First, althoughthe National Bank of Cambodia does not ensure theparity of the exchange rate, its entire policy stancehas been geared toward maintaining the stability ofthe riel against the dollar. The National Bank ofCambodia has been largely successful in this attemptsince mid-1999, as illustrated by Figure 13. Second,

V IMPLECATION OFHIGH DOLLARZATION FOR MACROECONOMIC POLICY DESIGN

24


Consideration of a Possible Currency Board Arrangement

net official reserves presently cover three times theamount of riels in circulation at the current exchangerate.23 Third, the National Bank of Cambodia refrainsfrom refinancing commercial banks and focuses onenforcing stricter banking regulations—although itlacks a strong banking supervision capacity yet.Finally, the National Bank of Cambodia does not pro-vide advances or financing to the government, anddoes not trade any public securities (which actuallydo not exist yet). The National Bank of Cambodia'spolicies and those normally implemented under an"orthodox" currency board arrangement are comparedin Box 7.

Cambodia appears to have fulfilled a number ofthe necessary conditions for entering into a success-ful currency board arrangement. The country is asmall and open economy—the sum of exports andimports represents 90 percent of GDP—and is wellintegrated into world trade. Cambodia is a memberof the Association of South-East Asian Nations(ASEAN), and accession to the World Trade Orga-nization (WTO) is well under way. In addition, theauthorities have a liberal investment law and are inthe process of lowering tariff and trade barriers.Moving from very high dollarization toward a cur-rency board arrangement in Cambodia thus could, inprinciple, be conceivable. This would howeverrequire the effective independence of the NationalBank of Cambodia (e.g., removing the governmentrepresentatives from the National Bank of Cambo-dia's Board); the ability to resist pressure for budgetfinancing in times of uncertainties; and the ability todecide freely on the exchange rate policy. Two pos-sible currency board arrangements could be envis-aged: one with a single legal tender—the riel—andthe obligation to quote all prices and effectuate alltransactions in that currency; or one where therewould be two legal tenders, the riel and the dollar.

A successful riel-based currency board arrange-ment would allow Cambodia to revitalize its mar-ginalized domestic currency, while ensuring continued

23Some particularly strong currency board arrangement backingrules require foreign currency coverage of deposits in domesticcurrency in the banking system. Considering the low amount ofdeposits in riels in Cambodia, even if the coverage had been aug-mented to include such deposits, net official reserves would stillhave been equivalent to 2VA times all riel components of broadmoney at the end of 2001.

economic stability. However, adopting a currencyboard arrangement would represent a drastic shift ata time when it is not an absolute necessity andwould not yield any worthy benefit over the currentpolicy mix. Most countries that adopt a rigorouscurrency board arrangement do so in times of crisisand as a last attempt to bring their economy undercontrol. Cambodia, as described earlier, is far frombeing in a crisis and has enjoyed economic stabilitywith growth for some time. In addition, the bankingsystem's restructuring is under progress, the finan-cial system remains underdeveloped and fragile,bank supervision is inadequate, and the fiscal posi-tion is still weak. Also, the National Bank ofCambodia and the government may not have thecapacity to proceed smoothly with all the opera-tional aspects involved in reintroducing the riel on amassive scale and to enforce the exclusive use of thedomestic currency. Furthermore, the level of netinternational reserves, compared with the amount oftrade-related transactions and capital flows, is insuf-ficient to offer credible support for an unconditionalpeg. Finally, such a sweeping policy move couldhave a negative impact on foreign investors, whoappreciate the fact that they can operate entirely indollars without an exchange rate risk. All these rea-sons argue against considering a single-currencycurrency board arrangement in Cambodia, as thebenefits under such an arrangement would notexceed those currently enjoyed under the presentpolicy stance, and would entail a major risk in caseof rejection by the majority of economic entities.There is a strong possibility that an attempt toenforce a riel-based currency board arrangementadministratively would instead lead to an immediatede facto full dollarization.

Considering the current co-currency situation andthe liberal approach toward economic regulationsprevailing in Cambodia, the second option (i.e., acurrency board arrangement with two legal tenders)would be by far the most acceptable. However, wewould argue that little additional benefit would begained by "locking in" the riel into a pegged and irre-versible status vis-a-vis the dollar. Most likely, adual-currency board arrangement in Cambodiawould not lead to significant de-dollarization, butwould eliminate the possibility of the marginalexchange rate tool provided by the current flexibleexchange rate policy.

25


VI Conclusion

The involuntary form of dollarization, called par-tial or de facto dollarization, occurs when one or

several foreign currencies circulate alongside anational currency, with usually the banking systemalso operating in those currencies. These develop-ments largely eschew the direct control of the author-ities. Partial dollarization, stretching from low tovirtually full dollarization, typically stems fromdisorderly macroeconomic conditions. In most coun-tries high dollarization is a spontaneous response—often desperate—to persistent high or hyperinflation.In the case of high dollarization, full dollarization isnow considered an exchange rate regime choice andcan offer, under certain circumstances, an attractiveoption to countries that find themselves confrontedwith a protracted economic crisis. In Cambodia, dol-larization surged suddenly and significantly duringthe period 1991-93, primarily as a result of massivedollar inflows. Very high dollarization in Cambodiathus was principally an exogenous shock, and hasresulted in low inflation and economic stability. Yetdollarization has been persistent, despite increasedeconomic and political stability.

There is a sizable amount of cash dollars circulat-ing in the Cambodian economy, of which we providean econometric estimate since early 1995. Residentskeep a substantial amount of dollars in cash as a storeof value, probably owing to the limited public confi-dence in the banking system and shallow financialintermediation. These dollars circulate very slowly inthe economy, which remains relatively little mone-tized and operates predominantly on a cash basis, inparticular as regards the national budget. The lowestimated velocity provided by our empirical analysisis consistent with inflation that has remained subduedin spite of the large dollar money supply in the econ-omy. In this regard, the most important policy mea-sure for the National Bank of Cambodia was toeliminate its financing of the government budgetdeficit after mid-1998, which used to be the mainsource of inflation in Cambodia in the early 1990s.

In the face of very high dollarization, the NationalBank of Cambodia's monetary and exchange poli-cies and the Ministry of Economy and Finance'sbudget policy have served Cambodia well since1999. By refraining from budget financing and from

refinancing commercial banks on the one hand, andby producing a current budget surplus and an overallbudget deficit entirely financed by foreign financingon the other, both institutions have been conductingmutually reinforcing macroeconomic policies. Thismix has provided a stable riel supply, which in turnresulted in a relatively stable exchange rate, whilethe dollar supply has been absorbed by large idlecash balances held by residents and the export ofexcess dollar banknotes by commercial banks. Veryhigh dollarization, coupled with prudent macroeco-nomic policies, has thus largely sheltered the Cam-bodian economy from international economicturmoil, while providing a propitious environmentfor growth and poverty reduction, provided advan-tage is taken of this situation to push forcefullyahe.ad with wide-ranging structural reforms.

In our opinion, this policy stance is appropriate inthe short to medium term. We do not favor endeav-oring to de-dollarize the economy, beyond executingthe budget fully in national currency, as the currenteconomic context is not favorable for such an under-taking; nor do we recommend full dollarization.The current policy stance is not a panacea, however,and developments will need to be constantly moni-tored to ascertain that the domestic and internationalmacroeconomic environments remain auspiciousfor the continued implementation of the aforesaidpolicies. Downside risks of the current policy mixare linked to uncertainties of the growing economy.They could involve increased dollar inflows, stem-ming from renewed investment and higher tourismreceipts, coupled with improved financial inter-mediation and the launching of financial markets.Similarly, large amounts of idle cash dollar balancescould flow into the banking system, as confidenceimproves. Heightened bank capital requirements,coupled with excess liquidity, constitute a strongincentive for increased lending in the future.24 Such

24The recently approved Land Law, the prospect of a land reg-istry, forthcoming laws on corporate insolvency and secured trans-actions, and the ongoing reform of the judiciary system will makethe taking of collateral eventually easier, and thus may lead to newlong-term lending activities. In the meantime, short-term credit isthe most likely to develop.

D©International Monetary Fund. Not for Redistribution

developments could result in higher inflation, as theNational Bank of Cambodia currently has no meansto sterilize dollar inflows or dollars released fromcash balances, and enhanced financial inter-mediation could lead to increased velocity of moneycirculation.25 These likely developments call forimproved policy tools capable of handling a morecomplex economy that will come from higherdevelopment. Fiscal discipline will need to remaina cornerstone of the authorities' policy mix, andthe National Bank of Cambodia will need tostrengthen drastically its supervisory capacity andits ability to influence bank liquidity, presumablyvia the eventual development of some type of finan-cial instruments.

A number of studies have found that once an econ-omy is highly dollarized, it is very hard to reduce dol-larization significantly, even if economic and politicalstability has been restored. Only Liberia, with a longhistory of dollarization, reintroduced its own currencyin the 1980s. In Cambodia, dollarization could bemarginally reduced by implementing the budgetentirely in riels, and by issuing treasury bills in riels,26

but beyond that move, neither a single-currency boardarrangement, nor an administratively enforced de-dol-larization, appears desirable, as either would likelyresult instead in a de facto full dollarization.

25Commercial banks currently sterilize some of their dollardeposits by investing them abroad.

26For the first time, the Ministry of Economy and Financeissued government bonds in February 2002 to bring the capital ofthe Foreign Trade Bank to the statutory threshold. In early 2003,the Ministry of Economy and Finance issued treasury bills denom-inated in riels for a total amount of 32 billion riels. These billswere purchased only by commercial banks.

27


Appendix Cambodia—A Simple Model toEstimate Dollars in CirculationOutside Banks

% A #M,Vt=PtTt, (1)

where Mt, Vt, Pt, and Tt denote the money circulatingin the economy, the velocity of money, the pricelevel, and the number of transactions at time t,respectively. As a result of currency substitution, Mthas two components: riels in circulation (cash andchecks): RMRJ

21 and dollars in circulation (cash only)converted into riels: DMR>t

2S Then,

Mt = RMR>t + DMRJ = (1 + kt)

RMR>h with kt > 0. (2)

Replacing Mt in equation (1), we have:

(l+kt)*MR>tVt = PtTt. (3)

For the sake of simplicity, we assume that the veloc-ities of dollars and riels are the same.29 At eachperiod of time, PtTt is proxied by nominal GDP, andconsequently:

(4)(l+kt)RMR>tVt =

Taking logs and rearranging terms, we obtain:

,,) = \og{GDPt) - log(V,) - log(l + kt). (5)

Our goal is to give an evaluation of kt in order toderive an estimate for DMRt.

However, there are two unknown parameters inequation (5): the velocity of money, Vt, and the pro-portionality coefficient between riels and dollars incirculation, kt, as kt cannot be measured (or proxied)by direct accounting. Both variables are behavioral,as agents have to decide simultaneously betweencurrent expenditure, foreign and local money bal-ances for future spending, and financial assets (in theCambodian context chiefly foreign cash currency asstore of value). Agents optimize their decisions in

27Most checks clear immediately, hence no lags are used. Thereare virtually no electronic payments.

28This component includes all cash foreign currencies circulat-ing in Cambodia. There are no sufficient time series for checksdenominated in dollars.

29Relaxing this hypothesis leads to a nonlinear specification,which is hardly tractable, owing to the mathematical complexityof the resulting equations.

light of available economic information, includinginflation and exchange rate anticipations. A relatedcash-in-advance model is presented by Hromcova(1998) in which the velocity of money evolvesendogenously with time, because of uncertaintyabout the state of the economy. We develop the fol-lowing assumptions for the behavior of the twounobservable variables.

The velocity of money, Vt, evolves over timeaccording to three factors: changes in inflation, asmeasured by the CPI, including all items;30 changesin the level of the exchange rate, which expand orcontract the riel value of money; and stochasticshocks, which are unobservable.

The proportionality coefficient, ku evolves over timeaccording to two factors: the level of the exchangerate; and stochastic shocks, which are unobservable.

In addition, we assume that both variables dependon their levels during the previous period to take intoaccount persistence, in particular in the use of thedominant transactions technology—dollar papermoney.

We derive two more equations from these assump-tions:

= fli log(V,) + a2 dlog(CP/,+i)+ a3 d\og(EXRATEt+i) + u.,+i, and

b2 \og{EXRATEt+l)+ i),+

(6)

(7)

where dlog indicates the difference logt - log t_\; CPIdenotes the consumer price index; EXRATE is theexchange rate (riels per dollar); and [it and vt are twostochastic terms.

We thus have a system of three equations:

log(RMR>t) = log(GDPt) - log(Vi) - log(l + kt\ (5)

log(V,+1) = ax log(V,) + a2 dlog(CP/,+i)+ a3 d\og{EXRATEt+i)+ |l,+i, and

b2 \og(EXRATEt+l)

(6)

(7)

30We also tried using CPI all items less food, beverages, andtobacco, and found similar results.

e s t a r t f r o m t h e e q u a t i o n o f e x c h a n g e :

2 8


This system is known as a state—space representa-tion whose parameters can be estimated by theKalman filter (see Hamilton, 1994), which we use forsolving the model. This method is particularly usefulwhen two unobservable variables need to be esti-mated. The method estimates a system of equationscombining two equations describing the unobservedvariables to estimate (Vt and kt)\ and one equationlinking these two unobserved variables to an observedvariable. The Kalman filter has been used in recentresearch, for instance by Harvey and Pierse (1994),and Bernanke, Gertler, and Watson (1997), as thisdynamic procedure allows to update the first esti-mates as new information becomes available.

For the sake of simplicity we rewrite our system inmatrix form:

log I = Hog GDP, + (-1 -1) (J°* [' . A+ k,)l (8)

log Vtt+l

log V,

\log EXRATE(+1 /

Changing notations, we arrive at:

(9)

(10)

(11)

Equation (10) is called the observation equation (andis deterministic); equation (11) is the state equation(and is stochastic).

Let Y, = (yh..., y,, Xu ..., X,, x\,..., xf), be theinformation set at time t. And let us assume that:

(12)o t h e r w i s e .

Cuche and Hess (1999), using a similar framework,find that:

Correction update step:

U = U-i+CPIt/t-iH{E(yt-yt/t..i)(yt-yt/l-ly}-i

(yt-M-i), (13)

)(14)

CPIm =

(yt

where CPI^ = £(§ - U-M& " lt/t-i)'.

Prediction step:

(15)

(16)

Mean Square Errors (MSE) step:

MSE&+v,) = (PIM - FCPIt/, F + Q, and (17)

MSE(y,+u) = H'CPIt+x/tH. (14)

At each step, we need |^,_i and CPI^-i to calculate|,+vr and CPIl+\/t; therefore, to start the iterativeprocess, |M) and CPI\^ need to be specified. Whilewe do not have information for t = 0, usually £,\® isset to zero and CPIy§ to an arbitrarily large value. Fand C are matrices of unknown parameters whosevalues are derived by maximizing the log likelihood.

Log likelihood:

L =

' (18)\

Our goal is to infer ^ based on the full set of data col-lected; a smoothed estimate of %t based on the full setof data is noted ^t/j. For this, we assume that weknow the true value of %t+i; therefore we can reviseour estimates of t,u given Yt and ^t+\.

Then:

Yt) = It/t

(19)

According to our previous equation, we obtain:

E(&(Zt=h Yd = l/t + CPIt/tF'CPUlvi^x - lt+yt\ (20)

and consequently:

tF'CPIt+i/t(%t+i/T - E>t+\/t)' (21)

Then following Hamilton (1994), the Kalman fil-ter is calculated, and the sequences {|#}£i, {%t+i/t}J=u{CPIt/t}^, and {CPIt+ytftJ: are stored. \ m is the lastentry in {t;t/t}J=h we compute Pt/tF'CPI7+i/t and useequation (21) for t = T - 1 to calculate |r-i/r- Pro-ceeding backward, we derive the full set of smoothedobservations.

In order to estimate the parameters of the state-space representation, initial values are required. Tocircumvent the issue of uncertainty for those values,high or low initial values are usually chosen. As afirst estimate of the velocity of money, we dividenominal GDP by the average stock of broad money.While this method ignores cash dollars circulating inthe economy, we use this information as an a prioriupper limit for velocity. The computed velocities areprovided in Table 5.

A p p e n d i x


Table 5.Velocity of Broad

Nominal GDP (in billions of riels)Broad money (in billions of riels)Velocity (Vy

Money 1

1995

7,54364914.4

1996

8,32591210.7

1997

9,1491,063

9.7

1998

10,5431,230

9.5

1999

11,6461,443

8.5

2000

11,9231,831

6.9

2001

12,9322,209

6.4

111 Sources: Cambodian authorities; and National Bank of Cambodia.

Note:Yearly averages of quarterly data.

Happe (1995) assessed the possible degree of dol-larization at the end of 1994 in Cambodia, by calcu-lating velocity based on broad money (includingforeign currency deposits). She found that measuredvelocity amounted to 15.4. Our velocity in 1995 isbroadly consistent with this result.

Given our a priori bounds, we select an upperbound of 6 for Vo and 0 for ko. It is worth noting thatin general the choice of initial values does not greatlyaffect the final results. However, poor choices of ini-tial values may lead to convergence on local equilib-ria far from the true value of economic parameters(this is the major drawback of this methodology).They may also increase the time required for conver-gence on the estimation algorithm.31

To solve our system, we use data from theNational Bank of Cambodia and the National Insti-tute of Statistics covering the period January1995-January 2001. The NBC provides monthlydata for RMR t and for cleared checks denominated inriels. The National Institute of Statistics providesyearly GDP data for 1995-2000. Monthly data are

31 We experimented with different initial values and found thatthey led to similar results.

derived from the yearly series using cubic interpola-tion. Using the XI2 procedure, all data are season-ally adjusted. Computations are done using theEviews Ver. 4.0 software. Using the data specifiedabove and the Kalman filter methodology, we findthe maximum likelihood estimates for the parame-ters, as shown in Table 6.

We expect all parameters except a?> to be positive.Some of the parameters are not significantly differ-ent from zero; nevertheless, we do not drop themfrom the equations in view of their economic rele-vance. Under the hypothesis that a\ is equal to unity,equation (6) can be reorganized so that the leftmember becomes log(V,) - log(Vr_i). The equationwould then describe the growth rate of Vt. Ourempirical results indicate that a\ is significantly dif-ferent from unity. As expected, inflation raises thevelocity of money (although this coefficient is notsignificantly different from zero), while an increasein the exchange rate decreases velocity. Other thingsbeing equal, if the dollar appreciates, the totalamount of money circulating in the economyexpands and velocity falls. Concurrently, the dol-lar's appreciation leads agents to hold cash in dol-lars rather than in local currency, increasing theproportionality coefficient.

Parameter Coefficient Std. Error z-Statistic Prob.

aia2

a3

bib2Log likelihood

0.9610500.174978

-0.6444820.4120670.229147

143.6093

0.0198280.3800140.1886820.2967850.1 17801

48.468630.460451

-3.4157021.3884341.94521 1

Akaike crit.Schwarz crit.Hannan-Quinn

0.00000.64520.00060.16500.0517

-3.794702-3.573359-3.706585

Source: Authors' calculations.

A p p e n d i x

3 0


fable 7. Velocity of Broad Money Kl

l

Average

Year 1995 1996 1997 1998 1999 2000 1995-2000

1.57 1.26 1.23 0.95 0.92 0.93 1.14


Note: Yearly averages of monthly estimates.

Table 8. Value of the ProportionalityCoefficient

\ Year

Kt

1995

19.5

1996

20.4

1997

21.1

1998

23.5

1999

23.8

2000

23.9

Average I1995-2000 !

22.1


Note: Yearly averages of monthly estimates.

We solve the system to compute the parameters ofinterest, Vt and kt. According to our estimates, theaverage velocity of money is 1.14 (Table 7).

We find significantly lower velocities then thosepresented in Table 5, owing to the estimated largermoney supply. A velocity close to, or below, unityreflects limited financial intermediation and theabsence of financial assets, as we assume that cashbalances are held by households as unproductivesavings, and dollars are used largely as a store ofvalue. In theory, residents should be able to switchbetween money and nonmonetary liquid assets; how-

ever, this is not possible in Cambodia. Therefore, thevelocity of the noncirculating money is zero or closeto zero. In other words, it is likely that a large part ofdollars outside the banking system would beexchanged for nonmonetary assets with positive realreturns, if this were possible.

Solving the system for the proportionality coeffi-cient, kt, provides an average value of 22.1 (Table 8).Using the earlier findings, our monthly estimates ofdollars circulating in the economy are shown inFigure 7 of the main text.

3 1


Bibliography

Agenor, Pierre-Richard, and Mohsin S. Khan, 1996, "For-eign Currency Deposits and the Demand for Money inDeveloping Countries," Journal of Development Eco-nomics, Vol. 50, pp. 101-18.

Allison, Theodore E., and Rosanna S. Pianalto, 1997, "TheIssuance of Series—1996 $100 Federal ReserveNotes: Goals, Strategy, and Likely Results," FederalReserve Bulletin, July.

Baliiio, Tomas J.T., and Charles Enoch, 1997, CurrencyBoard Arrangements: Issues and Experiences, IMFOccasional Paper No. 151 (Washington: InternationalMonetary Fund).

Balino, Tomas J.T., Adam Bennett, and Eduardo Boren-sztein, 1999, Monetary Policy in DollarizedEconomies, IMF Occasional Paper No. 171 (Wash-ington: International Monetary Fund).

Berg, Andrew, and Eduardo Borensztein, 2000a, "The Prosand the Cons of Full Dollarization," IMF WorkingPaper No. 00/50 (Washington: International MonetaryFund).

, 2000b, Full Dollarization: The Pros and Cons,IMF Economic Issues Series No. 24 (Washington:International Monetary Fund).

Bernanke, Ben S., Mark Gertler, and Mark Watson, 1997,"Systemic Monetary Policy and the Effects of OilPrice Shocks," Brookings Papers on Economic Activ-ity: lz Brookings Institution.

Bogetic, Zeljko, 2000, "Official Dollarization: CurrentExperiences and Issues," Cato Journal, Vol. 20 (Fall),pp. 179-213.

Bolbol, Ali A., 1999, "Seigniorage, Dollarization and Pub-lic Debt: The Lebanese Civil War and Recovery Expe-rience, 1982-97," World Development, Vol. 27, No.10, pp. 1861-73.

Brodsky, Boris, 1997, "Dollarization and Monetary Policyin Russia," Review of Economies in Transition(Helsinki: Bank of Finland, Unit for Eastern Euro-pean Economies, June).

Calvo, Guillermo A., and Carmen M. Reinhart, 1999,"Capital Flow Reversals, the Exchange Rate Debate,and Dollarization," Finance and Development, Vol. 36(September).

, 2000, "Reflections on Dollarization," in CurrencyUnions, ed. by Alberto Alesina and Robert J. Barro(Stanford: Hoover Institute Press).

Calvo, Guillermo A., and Carlos A. Vegh, 1992, "CurrencySubstitution in Developing Countries: An Introduc-tion," IMF Working Paper No. 92/40 (Washington:International Monetary Fund).

Cuche, Nicolas A., and Martin K. Hess, 1999, "EstimatingMonthly GDP in a General Kalman Filter Frame-work: Evidence from Switzerland," Working PaperNo. 9902 (Gerzensee, Switzerland: StudienzentrumGerzensee).

Cuddington, John T., 1989, Review of Currency Substitu-tion Theory and Evidence from Latin America, by Vic-tor A. Canto and Gerald Nickelsburg, in Journal ofMoney, Credit, and Banking, Vol. 21 (May), pp.267-71.

Edwards, Sebastian, 2001, "Dollarization: Myths andRealities," Journal of Policy Modeling, Vol. 23(April), pp. 249-65.

Eichengreen, Barry, 2001, "What Problems Can Dollariza-tion Solve?" Journal of Policy Modeling, Vol 23(April), pp. 267-77.

Fischer, Stanley, 1982, "Seigniorage and the Case for aNational Money,"Journal of Political Economy,Vol.90 (April), pp. 295-313.

Fontaine, Juan Andres, 2000, "Official Versus Sponta-neous Dollarization," Cato Journal, Vol. 20(Spring/Summer).

Ghosh, Atish R., Anne-Marie Guide, and Holger C. Wolf,1998, "Currency Boards: The Ultimate Fix?" IMFWorking Paper No. 98/8 (Washington: InternationalMonetary Fund).

Giovannini, Alberto, and Bart Turtelboom, 1994, "Cur-rency Substitution," in The Handbook of InternationalMacroeconomics, ed. by F. Vander Ploeg (Cambridge,Mass.: Blackwell), pp. 390-436.

Granger, C.W.J., 1969, "Investigating Causal Relations byEconometric Models and Cross-Spectral Methods,"Econometrica, Vol. 37, pp. 24-36.

Hamilton, James D., 1994, Time Series Analysis (Prince-ton: Princeton University Press).

Hanke, Steve H., 1999, "Some Thoughts on CurrencyBoards," in Balance of Payments, Exchange Rates,and Competitiveness in Transition Economies, ed. byMario I. Blejer and Marko Skreb (Boston: KluwerAcademic Publishers).

Happe, Nancy, 1995, "Cambodia—Financial Policy Impli-cations of Dollarization," (unpublished; Washington:International Monetary Fund).

Harvey, Andrew C, and Richard G. Pierse, 1994, Estimat-ing Missing Observations in Economic Time Series,"Journal of the American Statistical Association, Vol.79 (March), pp. 125-31.

Hoffmaister, Alexander, and Carlos A. Vegh, 1995, "Dis-inflation and the Recession-Now-Versus-Recession-

3 2


Later Hypothesis: Evidence from Uruguay," IMFWorking Paper No. 95/99 (Washington: InternationalMonetary Fund).

Horvath, Balazs, 1991, Are Policy Variables Exogenous?Lecture Notes in Economics and Mathematical Sys-tems, Vol. 364 (New York: Springer-Verlag).

Hromcova, Jana, 1998, "A Note on Income Velocity ofMoney in a Cash-In-Advance Economy with Capital,"Economics Letters, Vol. 60 (July), pp.91-96.

International Monetary Fund, 1997, "Cambodia—RecentEconomic Developments," IMF Staff Country ReportNo. 97/9 (Washington: IMF).

Kalman, Rudolf E., 1960, 'A New Approach to Linear Fil-tering and Prediction Problems," Journal of BasicEngineering, Transactions of the ASME Series D,Vol. 82, pp. 35-^5.

Kamin, Steven B., and Neil R. Ericsson, 1993, "Dollariza-tion in Argentina," International Finance DiscussionPaper No. 460 (Washington: Board of Governors ofthe U.S. Federal Reserve System).

Keat, Chhon, and Moniroth Aun Porn, 1999, Cambodia'sEconomic Development: Policies, Strategies, andImplementation (London: ASEAN Academic Press).

Koopman, Siem Jan, Neil Shephard, and Jurgen A.Doornik, 1999, "Statistical Algorithms for Models inState Space Using SsfPack 2.2," Econometrics Jour-nal, Vol. 2, No. 1, pp. 107-60.

Liang, Hong, Thomas Rumbaugh, Kotaro Ishi, and AtsushiMasuda, 2000, "Cambodia: Selected Issues," IMFStaff Country Report No. 00/135 (Washington: Inter-national Monetary Fund).

Marciniak, Philippe, and Sopanha Sa, 2002, "Cambodia:Integration and Competitiveness Study; Part B,Macroeconomic Assessment" (Ministry of Com-merce, Royal Government of Cambodia), pp. 1-24.

Marshall, David A., 1987, "Inflation and Asset Returns ina Monetary Economy with Transactions Costs,"(Ph.D. diss., Carnegie-Mellon University).

May, Tola, 1995a, "Auction System of the National Bankof Cambodia," Indochina Monitor (Standard Char-tered Bank, January 25).

, 1995b, "De-Dollarization," National Bank ofCambodia Bulletin (Phnom Penh: National Bank ofCambodia, December 1).

McKinnon, Ronald I., 1985, "Two Concepts of Interna-tional Currency Substitution" in The Economics ofthe Caribbean Basin, ed. by Michael B. Connolly andJohn McDermott (New York: Praeger).

, 1996, "Direct and Indirect Concepts of Interna-tional Currency Substitution," in The Macroeconom-ics of International Currencies: Theory, Policy, andEvidence, ed. by Paul Mizen and Eric J. Pentecost(U.K.: Edward Elgar), pp. 44-59.

Mendoza, Enrique G., 2001, "The Benefits of Dollariza-tion When Stabilization Policy Lacks Credibility andFinancial Markets Are Imperfect," Journal of Money,Credit, and Banking, Vol. 33, pp. 440-75.

Menon, Jay ant, 1998a, "De-Dollarizing Cambodia," Cam-bodian Development Review, Vol. 2, No. 2 (Cambo-dian Development Resource Institute).

, 1998b, "Cambodia: ASEAN Membership andMacroeconomic Policy Issues," ADB T.A. No.5713—CAM/LAO: Technical Assistance to Cambo-dia and Lao P.D.R. to Prepare for Membership inASEAN (Manila: Asian Development Bank).

Milner, Chris, Paul Mizen, and Eric Pentecost, 1998, "ACross-Country Panel Analysis of Currency Substitu-tion and Trade," LSE Discussion Paper No. 99/14(London School of Economics).

Mongardini, Joannes, and Johannes Mueller, 2000,"Ratchet Effects in Currency Substitution: An Appli-cation to the Kyrgyz Republic,"/MF Staff Papers,International Monetary Fund, Vol. 47 (December), pp.218-37

Mueller, Johannes, 1994, "Dollarization in Lebanon," IMFWorking Paper No. 94/129 (Washington: Interna-tional Monetary Fund).

Mundell, Robert A., 1961, "A Theory of Optimal CurrencyAreas," American Economic Review, Vol. 51 (Sep-tember), pp. 657—65.

Niskanen, William A., 2000, "Dollarization for LatinAmerica?" Cato Journal, Vol. 20 (Spring/Summer).

Ortiz, Guillermo, 1983, "Currency Substitution in Mexico:the Dollarization Problem," Journal of Money, Credit,and Banking, Vol. 15 (May), pp. 174-85.

Poloz, Stephen S., 1986, "Currency Substitution and thePrecautionary Demand for Money," Journal of Inter-national Money and Finance, Vol. 5 (March), pp.115-24.

Rose, Andrew, 2000, "One Money, One Market: the Effectof Common Currencies on Trade," Economic Policy:A European Forum, Vol. 0 (April), pp. 7-33.

, and Jeffrey Frankel, 2000, "Estimating the Effectof Currency Unions on Trade and Output," NBERWorking Paper No. 7857 (Cambridge, Mass.:National Bureau of Economic Research).

Sahay, Ratna, and Carlos A. Vegh, 1995, "Dollarization inTransition Economies: Evidence and Policy Implica-tions," IMF Working Paper No. 95/96 (Washington:International Monetary Fund).

Sarajevs, Vadims, 2000, "Econometric Analysis of Cur-rency Substitution: The Case of Latvia," DiscussionPaper No. 4 (Helsinki: Bank of Finland, Institute forEconomies in Transition).

Sargent, Thomas, 1986, Rational Expectations and Infla-tion (New York: Harper & Row).

Sturzeneggar, Federico, 1997, "Understanding the WelfareImplications of Currency Substitution," Journal ofEconomic Dynamics and Control, Vol. 21 (Feb.-March), pp. 391-416.

Vetlov, Igor, 2001, "Dollarization in Lithuania: AnEconometric Approach," Discussion Paper No. 1(Helsinki: Bank of Finland, Institute for Economies inTransition).

3 3


OCCASIONAL PAPERS

Recent Occasional Papers of the International Monetary Fund

219. Economic Policy in a Highly Dollarized Economy: The Case of Cambodia, by Mario de Zamaroczy andSopanha Sa. 2003.

218. Fiscal Vulnerability and Financial Crises in Emerging Market Economies, by Richard Hemming, MichaelKell, and Axel Schimmelpfennig. 2003.

217. Managing Financial Crises: Recent Experience and Lessons for Latin America, edited by Charles Collynsand G. Russell Kincaid. 2003.

216. Is the PRGF Living Up to Expectations? An Assessment of Program Design, by Sanjeev Gupta, MarkPlant, Benedict Clements, Thomas Dorsey, Emanuele Baldacci, Gabriela Inchauste, Shamsuddin Tareq,and Nita Thacker. 2002.

215. Improving Large Taxpayers' Compliance: A Review of Country Experience, by Katherine Baer. 2002.214. Advanced Country Experiences with Capital Account Liberalization, by Age Bakker and Bryan Chappie.

2002.213. The Baltic Countries: Medium-Term Fiscal Issues Related to EU and NATO Accession, by Johannes

Mueller, Christian Beddies, Robert Burgess, Vitali Kramarenko, and Joannes Mongardini. 2002.212. Financial Soundness Indicators: Analytical Aspects and Country Practices, by V. Sundararajan, Charles

Enoch, Armida San Jose, Paul Hilbers, Russell Krueger, Marina Moretti, and Graham Slack. 2002.211. Capital Account Liberalization and Financial Sector Stability, by a staff team led by Shogo Ishii and Karl

Habermeier. 2002.210. IMF-Supported Programs in Capital Account Crises, by Atish Ghosh, Timothy Lane, Marianne Schulze-

Ghattas, AleS Bulfr, Javier Hamann, and Alex Mourmouras. 2002.209. Methodology for Current Account and Exchange Rate Assessments, by Peter Isard, Hamid Faruqee,

G. Russell Kincaid, and Martin Fetherston. 2001.208. Yemen in the 1990s: From Unification to Economic Reform, by Klaus Enders, Sherwyn Williams, Nada

Choueiri, Yuri Sobolev, and Jan Walliser. 2001.207. Malaysia: From Crisis to Recovery, by Kanitta Meesook, II Houng Lee, Olin Liu, Yougesh Khatri, Natalia

Tamirisa, Michael Moore, and Mark H. Krysl. 2001.206. The Dominican Republic: Stabilization, Structural Reform, and Economic Growth, by Alessandro

Giustiniani, Werner C. Keller, and Randa E. Sab. 2001.205. Stabilization and Savings Funds for Nonrenewable Resources, by Jeffrey Davis, Rolando Ossowski,

James Daniel, and Steven Barnett. 2001.204. Monetary Union in West Africa (ECOWAS): Is It Desirable and How Could It Be Achieved? by Paul Mas-

son and Catherine Pattillo. 2001.203. Modern Banking and OTC Derivatives Markets: The Transformation of Global Finance and Its Implications

for Systemic Risk, by Garry J. Schinasi, R. Sean Craig, Burkhard Drees, and Charles Kramer. 2000.202. Adopting Inflation Targeting: Practical Issues for Emerging Market Countries, by Andrea Schaechter,

Mark R. Stone, and Mark Zelmer. 2000.201. Developments and Challenges in the Caribbean Region, by Samuel Itam, Simon Cueva, Erik Lundback,

Janet Stotsky, and Stephen Tokarick. 2000.200. Pension Reform in the Baltics: Issues and Prospects, by Jerald Schiff, Niko Hobdari, Axel Schimmel-

pfennig, and Roman Zytek. 2000.199. Ghana: Economic Development in a Democratic Environment, by Sergio Pereira Leite, Anthony Pel-

lechio, Luisa Zanforlin, Girma Begashaw, Stefania Fabrizio, and Joachim Harnack. 2000.198. Setting Up Treasuries in the Baltics, Russia, and Other Countries of the Former Soviet Union: An Assess-

ment of IMF Technical Assistance, by Barry H. Potter and Jack Diamond. 2000.197. Deposit Insurance: Actual and Good Practices, by Gillian G.H. Garcia. 2000.196. Trade and Trade Policies in Eastern and Southern Africa, by a staff team led by Arvind Subramanian, with

Enrique Gelbard, Richard Harmsen, Katrin Elborgh-Woytek, and Piroska Nagy. 2000.195. The Eastern Caribbean Currency Union: Institutions, Performance, and Policy Issues, by Frits van Beek,

Jose Roberto Rosales, Mayra Zermeno, Ruby Randall, and Jorge Shepherd. 2000.

3 4


Occasional Papers

194. Fiscal and Macroeconomic Impact of Privatization, by Jeffrey Davis, Rolando Ossowski, ThomasRichardson, and Steven Barnett. 2000.

193. Exchange Rate Regimes in an Increasingly Integrated World Economy, by Michael Mussa, Paul Masson,Alexander Swoboda, Esteban Jadresic, Paolo Mauro, and Andy Berg. 2000.

192. Macroprudential Indicators of Financial System Soundness, by a staff team led by Owen Evans, AlfredoM. Leone, Mahinder Gill, and Paul Hilbers. 2000.

191. Social Issues in IMF-Supported Programs, by Sanjeev Gupta, Louis Dicks-Mireaux, Ritha Khemani,Calvin McDonald, and Marijn Verhoeven. 2000.

190. Capital Controls: Country Experiences with Their Use and Liberalization, by Akira Ariyoshi, Karl Haber-meier, Bernard Laurens, Inci Otker-Robe, Jorge Ivan Canales Kriljenko, and Andrei Kirilenko. 2000.

189. Current Account and External Sustainability in the Baltics, Russia, and Other Countries of the FormerSoviet Union, by Donal McGettigan. 2000.

188. Financial Sector Crisis and Restructuring: Lessons from Asia, by Carl-Johan Lindgren, Tomas J.T.Balifio, Charles Enoch, Anne-Marie Guide, Marc Quintyn, and Leslie Teo. 1999.

187. Philippines: Toward Sustainable and Rapid Growth, Recent Developments and the Agenda Ahead, byMarkus Rodlauer, Prakash Loungani, Vivek Arora, Charalambos Christofides, Enrique G. De la Piedra,Piyabha Kongsamut, Kristina Kostial, Victoria Summers, and Athanasios Vamvakidis. 2000.

186. Anticipating Balance of Payments Crises: The Role of Early Warning Systems, by Andrew Berg,Eduardo Borensztein, Gian Maria Milesi-Ferretti, and Catherine Pattillo. 1999.

185. Oman Beyond the Oil Horizon: Policies Toward Sustainable Growth, edited by Ahsan Mansur and VolkerTreichel. 1999.

184. Growth Experience in Transition Countries, 1990-98, by Oleh Havrylyshyn, Thomas Wolf, Julian Beren-gaut, Marta Castello-Branco, Ron van Rooden, and Valerie Mercer-Blackman. 1999.

183. Economic Reforms in Kazakhstan, Kyrgyz Republic, Tajikistan, Turkmenistan, and Uzbekistan, by EmineGiirgen, Harry Snoek, Jon Craig, Jimmy McHugh, Ivailo Izvorski, and Ron van Rooden. 1999.

182. Tax Reform in the Baltics, Russia, and Other Countries of the Former Soviet Union, by a staff team ledby Liam Ebrill and Oleh Havrylyshyn. 1999.

181. The Netherlands: Transforming a Market Economy, by C. Maxwell Watson, Bas B. Bakker, Jan Kees Mar-tijn, and Ioannis Halikias. 1999.

180. Revenue Implications of Trade Liberalization, by Liam Ebrill, Janet Stotsky, and Reint Gropp. 1999.179. Disinflation in Transition, 1993-97, by Carlo Cottarelli and Peter Doyle. 1999.178. IMF-Supported Programs in Indonesia, Korea, and Thailand: A Preliminary Assessment, by Timothy Lane,

Atish Ghosh, Javier Hamann, Steven Phillips, Marianne Schulze-Ghattas, and Tsidi Tsikata. 1999.177. Perspectives on Regional Unemployment in Europe, by Paolo Mauro, Eswar Prasad, and Antonio Spilim-

bergo. 1999.176. Back to the Future: Postwar Reconstruction and Stabilization in Lebanon, edited by Sena Eken and

Thomas Helbling. 1999.175. Macroeconomic Developments in the Baltics, Russia, and Other Countries of the Former Soviet Union,

1992-97, by Luis M. Valdivieso. 1998.174. Impact of EMU on Selected Non-European Union Countries, by R. Feldman, K. Nashashibi, R. Nord,

P. Allum, D. Desruelle, K. Enders, R. Kahn, and H. Temprano-Arroyo. 1998.173. The Baltic Countries: From Economic Stabilization to EU Accession, by Julian Berengaut, Augusto

Lopez-Claros, Francoise Le Gall, Dennis Jones, Richard Stern, Ann-Margret Westin, Effie Psalida, andPietro Garibaldi. 1998.

172. Capital Account Liberalization: Theoretical and Practical Aspects, by a staff team led by Barry Eichen-green and Michael Mussa, with Giovanni Dell'Ariccia, Enrica Detragiache, Gian Maria Milesi-Ferretti,and Andrew Tweedie. 1998.

Note: For information on the titles and availability of Occasional Papers not listed, please consult the IMF's Publications Catalog or contact IMFPublication Services.

3 5


Documents

Mario de Zamaroczy and Sopanha Sa · 2. Empirical Models of Dollarization 3. Theoretical Models of Dollarization 4. Causality Analysis Between Inflation and Dollarization in Cambodia,