This Selected Issues paper for Cambodia explores the implications for long-term sustainable growth from cross-country analysis of the sources of growth. Cambodia’s low labor productivity, inadequate and expensive infrastructure, and a cumbersome regulatory environment do not bode well for future sustainable growth. Favorable external conditions, including foreign aid flows and trade agreements, have helped propel growth. As with other transition economies, Cambodia lags in institutional and market development. Cambodia has also benefited from large aid inflows, which have boosted economic activity.


This Selected Issues paper for Cambodia explores the implications for long-term sustainable growth from cross-country analysis of the sources of growth. Cambodia’s low labor productivity, inadequate and expensive infrastructure, and a cumbersome regulatory environment do not bode well for future sustainable growth. Favorable external conditions, including foreign aid flows and trade agreements, have helped propel growth. As with other transition economies, Cambodia lags in institutional and market development. Cambodia has also benefited from large aid inflows, which have boosted economic activity.

Chapter 5. International Experience of De-Dollarization40

64. This chapter summarizes recent research by Fund staff on country experiences oi dollarization. The first section describes recent global trends of increasing dollarization; the experience with price stability and exchange rate pass through in dollarized economi Section B outlines various de-dollarization approaches countries have pursued. Section C reports that there are only few cases of successful dollarization. The final section discus; options for Cambodia and the Annex presents an adaptation of the portfolio choice model to the case of Cambodia showing that greater exchange rate volatility could lead to further dollarization.

A. Dollarization Trends and Implications

65. Dollarization has been on the rise in the past two decades. A recent IMF stud covering some 117 countries found that financial dollarization, as measured by the share foreign currency deposits (FCD) in broad money, doubled in the last decade.41 This tren partly reflects the return of deposits previously held abroad following the easing of FCD restrictions in most countries. Financial dollarization (the use of foreign currency as a store of value) is easy to measure as data are readily available in financial statistics. However, the true extent of dollarization, which should encompass payment dollarization (the use of foreign currency for transactions purposes) and real dollarization (the use of foreign currency for denominating prices and wages), is more difficult to assess as information is not readily available.

Increasing Trend of Dollarization

(Percent of foreign currency deposits to broad money)

article image
Source: Macroeconomic Policies in Dollarized Economies.

66. Some countries, however, were able to avoid or contain dollarization. The IMF study found that those countries in Asia that did not experience periods of high inflation or severe macroeconomic instability, including India, Sri Lanka, and Bangladesh (as well as more advanced economies such as Singapore, Malaysia, and Taiwan, Province of China) retained domestic-currency denominated private savings. Some countries, including Chile and Colombia, that did experience large macroeconomic imbalances barely escaped dollarization by introducing financial indexation that helped contain the erosion of financial savings. Yet another group of countries, such as Venezuela, Nigeria, and many countries in sub-Saharan Africa, relied on financial repression and capital controls. However, these measures led to waves of capital flight and financial disintermediation.

B. Approaches to De-dollarization

67. The approaches taken by countries to reverse dollarization can be broken down into three types:

  • Macroeconomic policies: Pursuit of policies aimed at maintaining exchange rate and price stability (to avoid erosion of local currency value), including through inflation targeting to arrest inflation expectations. Financial liberalization that resulted in higher domestic interest rates also contributed to limiting dollarization.


    • Following price stabilization, the FCD ratio declined substantially in Poland and Israel. Armenia, Estonia, and Lithuania are countries where financial liberalization led to an increase in domestic interest rates, thereby helping to contain dollarization.

  • Regulatory/legal reforms: Changing the regulatory incentive structure through the setting of differential reserve requirements or remuneration rates, or adjusting provisioning and liquidity requirements, introducing alternative financial instruments, and requiring all or certain payments or contracts to be conducted in local currency.


    • In Nicaragua, a premium was paid on dollar-indexed deposits over dollar deposits. Price and interest-rate indexation were broadly used in Brazil to contain dollarization.

    • Peru, in the late 1980s, imposed a 2 percent transaction tax on check payments in foreign currency. Lao P.D.R. introduced a decree requiring all domestic transactions to be carried out in local currency, supplemented by improvements in the payments, clearing, and settlement system and the issuance of large denomination bank notes.

    • Bolivia, Honduras, Nicaragua, and Peru had higher reserve requirements for FCD than on local currency deposits, although Uruguay refrained from imposing the higher reserve requirement to avoid driving dollar deposits offshore. Israel used differential remuneration rates and imposed a one-year holding period for all FCD to encourage the use of dollar-indexed deposits over dollar deposits.

  • Administrative enforcements: Direct administrative measures such as prohibition of FCD for residents, restrictions on residents holding accounts abroad, and forced conversions of dollar to local currency deposits.


    • Israel limited payments in foreign currency by imposing a ban on direct transfers of FCD among residents.

    • Lebanon limited foreign currency lending to 60 percent of FCD, forcing banks to keep the remainder offshore. Vietnam, Malaysia, and the Philippines restricted foreign currency loans to particular uses or borrowers.

    • Peru, Bolivia, and Mexico had forced conversions of FCD in the eighties. In many African countries, FCD are still not allowed or severely restricted. But in most cases there are indications of extensive (unmeasured) use of dollars as cash in circulation.

C. Successful De-dollarization Experiences

68. There were, however, only four out of 85 countries surveyed covering 1980–2001 that succeeded in de-dollarization.42 Of those, only two countries, Poland and Israel, appear to have had lasting reversals with minimal side effects. For Mexico and Pakistan, it is too early to tell if de-dollarization will be sustained. Moreover, Mexico experienced doubling of capital flight and a drastic reduction of bank credit to the private sector.

69. In the case of Poland and Israel, both countries embarked on a successful disinflation program initially built around a strong exchange rate anchor. In Israel, the domestic financial system offered alternative forms of indexed assets, including dollar-indexed deposits (Patzams) with higher reserve remuneration rates. The Patzams proved an effective substitute for dollar deposits. In Poland, interest rates on domestic currency assets were raised to maintain a differential in favor of local currency deposits. But it is not at all clear that the conditions in Israel and Poland can be replicated by other countries, especially since the initial level of dollarization was not high in the first place.

Use of Foreign Currency in Selected Countries

(Percent of foreign currency bank deposits to broad money)

article image
Sources: Rogoff, Reinhardt, Savastano, Addicted to Dollars, NBER Working Paper 10015, October 2003, p. 46; Balino et al. Monetary Policy in Dollarized Economies, IMF Occasional Paper 171, 1990, pp. 4-6; Leung and Kompas, Dollarization and Macroeconomic Policy in Vietnam, manuscript, September 2003; pp. 7-8.

70. In contrast to the few success cases, there have been many more countries with experiences of unsuccessful attempts at de-dollarization. Often, these attempts involved administrative enforcements without fully restoring confidence in the local currency or eliminating the underlying instability that led to dollarization in the first place. In both Peru and Bolivia, foreign currency deposits accounted for about 30 percent of total deposits in the early 1980s. Both countries attempted forced conversion that led to an increase in cross-border deposits, capital flight, and reduced financial intermediation. Eventually, by the end of the 1980s, foreign currency deposit in both countries increased further to 70-80 percent.

71. There are also several countries that have intentionally opted to maintain a high level of foreign currency as part of their broad money. In Asia, Bhutan allows free use of the Indian rupee and Brunei the Singapore dollar to facilitate trade and economic cooperation with its larger neighbor, and to benefit from the stable macroeconomic conditions. For similar reasons, Lesotho and Namibia allow the use of the South African rand, Bosnia-Herzegovina the euro, and Haiti and the Bahamas maintain use of the U.S. dollar alongside their own currency. A few other countries, including Panama, El Salvador, and East Timor, opted for full dollarization.

D. Steps Toward De-dollarization in Cambodia

72. De-dollarization is a long-term objective for Cambodia. Country experience has shown that de-dollarization is a long-term process that requires foremost restoring confidence in the local currency. Confidence is restored when the private sector is sure that it will not be financially penalized for holding the local currency. Accordingly, only when continued macroeconomic stability and exchange rate stability are maintained, will financial deepening be brought about by an increase in the use of domestic currency.43

73. Exchange rate stability is important in maintaining price stability in highly dollarized countries. The 2003 Board paper found systematic differences in the pass through from exchange rate to prices. The impact of exchange rate changes on inflation was found to be largest for countries with a high degree of dollarization and where there was little private liability dollarization (low share of private sector debt in total external debt). However, the impact was the lowest in countries where overall dollarization is low and domestic dollarization was negligible.44 These results were consistent with the reluctance of central banks to tolerate large exchange rate changes, and also supports Cambodia’s pursuit of stable exchange rate.

74. Policies could be pursued that encourage the use of local cash without recourse to administrative controls that might result in capital flight. Such policies would comprise measures to conduct all public sector transactions in local currency, including an increase in the use of riel in the collection of tax and non-tax revenue and in payments for capital expenditure. In 2001, 74 percent of nontax revenues and 67 percent of capital expenditure were collected in foreign currency. Creating a wedge in reserve requirements between FCD and LCD could be another potential instrument. Introducing riel-denominated treasury bills could he useful in the future but is not an immediate measure to facilitate de-dollarization since banks have little need to manage riel liquidity given the low demand for riel. Finally, introducing a larger denomination of riel currency would enable payment of larger transactions in riel.

Annex: A Model for Determination of Foreign Currency Deposit Ratios45

We modify the model in Ize and Levy-Yeyati (2003) to analyze the dollarization issue in Cambodia. The key modification is that depositors evaluate their portfolios in U.S. dollar terms instead of in local currency terms. There are two types of deposits, one in local currency (riel), the other in U.S. dollars. Let rUSDh denote the real return for the local currency deposits, rUSDf the real return for foreign currency deposits, both measured in U.S. dollar terms. Assume that depositors’ preferences are given as:


where r is the averaged real return of the deposit portfolio, λ is the share of deposits in the foreign currency account, TD is the amount of total deposits, and c is the risk aversion measure. Given interest rates, depositors maximize the return on their deposits by choosing what share of their deposits to hold in the foreign currency account. The optimal share of foreign currency deposits in total deposits, λ*, can be written in the following way:


We use the following approximations:


where eUSD/Riel denotes the rate of change in the nominal exchange rate (U.S. dollar per Riel); λUS denotes the inflation rate in the United States. Substitute equations (4) and (5) into equation (3), and the optimal share of foreign currency deposits is derived as a function of nominal exchange rate changes and the inflation in the United States.


An increase in exchange rate volatility leads to a higher share of foreign currency deposits. This result is intuitive. If depositors measure everything in U.S. dollar terms, everything else being equal, higher exchange rate volatility only adversely affects the value of the riel account, but not the US dollar account.


  • International Monetary Fund, 1999, Monetary Policy in Dollarized Economies, Occasional Paper No. 171, September,(Washington DC, International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Ize Alain and Eduardo Levy-Yeyati, 1998, “Dollarization and Financial Intermediation: Causes and Policy Implications,” IMF Working Paper 98/22, March, (Washington DC, International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Reinhardt, Carmen M. Kenneth Rogoff and Miguel A. Savastano, 2003, “Addicted to Dollars,” National Bureau of Economic Research Working Paper 10015, October, (Cambridge, MA: NBER).

    • Search Google Scholar
    • Export Citation

Prepared by Wafa Fahmi Abdelati (APD).


This note draws on Rogoff Savastano and Reinhardt (2003), “Addicted to Dollars.”


The criteria used to identify successful reversal include reducing foreign currency deposit to broad money by 20 percent and remaining below that level until the end of the sample period.


In a recent paper, Ize and Levy-Yeyati (2003) argued that greater exchange rate flexibility would reduce incentives for dollarization. They advocated a floating exchange rate combined with an inflation targeting approach to foster the use of the local currency, based on a theoretical model that derives depositors’ optimal portfolio. In the Annex to this note, we show that this relation between exchange rate flexibility, inflation volatility and de-dollarization does not hold in the case of Cambodia when the model is adjusted to allow depositors to evaluate their portfolio in US dollar terms instead of in local currency terms.


Domestic dollarization in this case refers to the ratio of foreign currency deposits in broad money and the share of government debt that is foreign-currency-denominated.


Prepared by Zhiwei Zhang (APD).