Dedollarization
  • 1 https://isni.org/isni/0000000404811396, International Monetary Fund

Contributor Notes

Author’s E-Mail Address: rveyrune@imf.org; akokeny@imf.org

This paper provides a summary of the key policies that encourage dedollarization. It focuses on cases in which the authorities’ intention is to gain greater control of monetary policy and draws on the experiences of countries that have successfully dedollarized. Unlike previous work on the subject, this paper examines both macroeconomic stabilization policies and microeconomic measures, such as prudential regulation of the financial system. This study is also the first attempt to make extensive use of the foreign exchange regulation data reported in the IMF’s Annual Report on Exchange Arrangements and Exchange Restrictions. The main conclusion is that durable dedollarization depends on a credible disinflation plan and specific microeconomic measures.

Abstract

This paper provides a summary of the key policies that encourage dedollarization. It focuses on cases in which the authorities’ intention is to gain greater control of monetary policy and draws on the experiences of countries that have successfully dedollarized. Unlike previous work on the subject, this paper examines both macroeconomic stabilization policies and microeconomic measures, such as prudential regulation of the financial system. This study is also the first attempt to make extensive use of the foreign exchange regulation data reported in the IMF’s Annual Report on Exchange Arrangements and Exchange Restrictions. The main conclusion is that durable dedollarization depends on a credible disinflation plan and specific microeconomic measures.

I. Introduction

1. While dollarization is a rational response of economic agents to political or economic uncertainties, its adverse effects often motivate countries to reduce its level.2 Dollarization is also a rational reaction to interest rate arbitrage opportunities. It may have some benefits, and in extreme cases may be the only viable option available to a country. In the latter case, dollarization can be the choice of the authorities or a result of private agents’ decision to stop using the local currency. However, most countries seek to limit the extent of dollarization, owing to its potential adverse effects on macroeconomic policies and financial stability. These include a reduction or loss of control of monetary and exchange rate policy, a loss of seigniorage, and increased foreign exchange risk in the financial system and other sectors.

2. A successful dedollarization policy makes the local currency more attractive to residents than foreign currency. Dedollarization entails a mix of macroeconomic and microeconomic policies to enhance the attractiveness of the local currency in economic transactions and to raise awareness of the exchange-risk related costs of dollarization, thus providing incentives to economic agents to dedollarize voluntarily. It may also include measures to force the use of the domestic currency in tandem with macroeconomic stabilization policies.

3. Experience shows that, even in successful cases, dedollarization is protracted. Dollarization remains persistent even when macroeconomic stability and the credibility of government policies have been (re)established, i.e., when the original causes of dollarization have been eliminated.3 Once dollarization takes hold, economic agents are reluctant to switch back to using the local currency, because they lack confidence and the cost of redenominating transactions is high until consensus is reached among market participants on the use of the local currency. Indexation or a fixed exchange rate facilitates switching from foreign currency to local currency, but indexation may become persistent and the implicit guarantee of a fixed exchange rate can reinforce dollarization.

4. This paper provides a summary of the key policies that encourage dedollarization. Few studies of dedollarization have been undertaken, even though it is an important policy issue in many emerging and developing economies. This paper draws on the experiences of countries that successfully dedollarized and of those that have had less success. While acknowledging that economic agents target a mix of foreign and domestic assets and liabilities to minimize the volatility of their portfolio, it does not address the optimal level of dollarization or policies to reduce the risks related to a given dollarization level.4

5. The paper concludes that dedollarization requires credible macroeconomic stabilization complemented by microeconomic measures. In addition to macroeconomic stabilization, two—way exchange rate volatility and stable and low inflation are key ingredients of dedollarization. Additional policies and measures are often necessary to break persistent dollarization and encourage the use of the local currency.

6. The paper is organized as follows. Following a brief overview of the causes and effects of dollarization in Section II, Section III discusses the macroeconomic and microeconomic policies and measures that contribute to dedollarization. Section IV analyzes the effect of dedollarization policies, in particular exchange rate volatility and prudential measures. An extensive review of the financial and prudential measures that can facilitate dedollarization are provided in Appendix I, Tables IV.

II. Dollarization

7. Dollarization refers to the use by the residents of one country of assets (or liabilities) denominated in another country’s currency (Baliño, 2003) and can take many forms. Dollarization significantly differs according to (i) the type of assets (or liabilities) dollarized; (ii) whether the dollar has the status of legal tender5 (official dollarization) or whether there is de facto dollarization; and (iii) the extent of dollarization (full or partial).6

8. There are several different types of dollarization. Financial dollarization involves the substitution of local currency assets or liabilities for foreign currency assets or liabilities. Real dollarization involves the indexation of domestic transactions to the exchange rate. Transaction dollarization or currency substitution means that the dollar (or another foreign currency) is used as a means of payment in domestic transactions (Armas, Ize, Levy, and Yeyati, 2006). More than one type of dollarization can exist in a country. Obviously, measures for dedollarization have to reflect the prevailing type of dollarization. Table 1 shows the various assets and liabilities which can be subject to dollarization.

Table 1.

Liabilities and Assets Typically Dollarized

article image
Source: Reinhart, Rogoff, and Savastano (2003).

9. Dollarization is commonly measured as financial institutions’ holding of foreign exchange assets and liabilities. These measures thus refer to financial dollarization. Financial sector asset dollarization is usually assessed by quantifying the loans granted by the resident financial sector to resident households and nonfinancial corporations. Financial sector liability dollarization is determined as the share in the total liabilities of foreign currency denominated deposits and debt issued by financial institutions. Due to data limitations, non financial sector dollarization is rarely measured, and foreign exchange in circulation is often omitted.7

10. Dollarization generally develops when the local currency performs its basic functions poorly compared to other accessible currencies. The basic functions of a currency are reserve value, means of payment, and unit of account. The poor performance of the local currency in these functions encourages the development of various forms of dollarization, e.g. currency substitution develops if the local currency does not function properly as a means of payment. Real exchange rate instability, due to a high and volatile inflation rate, would motivate residents to denominate contracts in foreign currency when it ensures more stable purchasing power in terms of domestic consumption, fostering financial dollarization (Ize, Levy, and Yeyati, 2005).

11. An asymmetric exchange rate policy that resists nominal appreciation while allowing for depreciation can facilitate dollarization by providing a one-way bet for preserving purchasing power by holding foreign currency deposits (FCDs) (Rennhack and Nozaki, 2006). Conversely, resisting nominal appreciation can encourage borrowing in foreign exchange.

12. Dollarization can have some benefits and may be a sign of increasing integration in the world economy. Asset substitution may be a natural consequence of opening the economy and financial market liberalization (Baliño, Bennett, and Borensztein, 1999). Countries integrating into the world economy are increasingly exposed to shocks, which may require hedging and lead to some level of dollarization, which in turn may enhance closer integration with international markets and the development of domestic financial markets. Dollarization may mitigate exchange rate risk for foreign investors, increasing their confidence, thus boosting investments, and provides for portfolio diversification of residents. Allowing foreign currency deposits in domestic banks may help to remonetize the economy by encouraging a move from currency substitution to foreign currency deposits, provide funding to local banks, and reverse capital flight following episodes of unstable macroeconomic conditions. Lending in foreign currency may have a positive impact on domestic consumption and investment, and thus economic growth, through a reduction in the cost of credit. Therefore, there is an optimal level of dollarization that depends mainly on structural factors, such as the economy’s size, openness, the degree of financial integration, and market development.

13. Dollarization has also significant disadvantages that could offset its benefits and thus motivate country authorities to adopt dedollarization as a policy objective:

  • Dollarization limits the effectiveness of monetary policy (Ize, Levy, and Yeyati, 2005).

  • Dollarized countries lose part or all of their seigniorage. The loss in revenues depends on the degree and type of dollarization and could be large for economies with high currency substitution. Moreover, the use of foreign currency also means that the country implicitly pays some seigniorage to the issuers of the other currencies.

  • Dollarization reduces the efficiency of payments. Foreign banknotes are not always adapted to local business needs (small transactions), and the monetary authorities of dollarized countries cannot control the quality of the banknotes in circulation.

  • Partial dollarization increases balance sheet risks. Direct exchange rate risks resulting from currency mismatches in banks’ balance sheets may render the banking system more vulnerable. Further, indirect credit risks related to a devaluation arise if there is substantial dollar lending to nonhedged borrowers. A devaluation may offset the interest rate benefit from foreign currency borrowing, limit borrowers’ ability to repay dollar debts, and potentially trigger or aggravate a banking and macroeconomic crisis.

  • Dollarization increases the likelihood of a liquidity crisis. While lender-of-last-resort facilities can provide funding in domestic currency in the event of bank runs, they usually cannot provide unlimited funding in foreign currency. This may render foreign currency holders more prone to panic.

14. Reducing dollarization requires comprehensive actions by the authorities. While dollarization has pros and cons, it is likely that in many cases some degree of partial dollarization would be optimal. Estimating the optimal level of dollarization is beyond the scope of this paper. However, we argue that dollarization can exceed its optimum level and that it will often persist without official action.

III. Policies for Dedollarization

15. Dedollarization usually requires a combination of macroeconomic policies and microeconomic measures to enhance the attractiveness of the local currency versus the foreign currency. Against the backdrop of macroeconomic stabilization, several measures can foster dedollarization. These range from market-based measures that provide incentives to reverse currency substitution to measures that prohibit or strictly limit the use of foreign currency (forced dedollarization).

16. Dedollarization is facilitated by proper sequencing of the policies and microeconomic measures. Generally, the credibility of monetary policy needs to be reestablished to give full effect to measures to reverse dollarization. Since establishing monetary credibility may take a long time, certain policies and measures promoting voluntary dedollarization can be adopted during this period. In addition, measures to force dedollarization can also be taken in parallel with comprehensive stabilization policies. In either case, policy makers need to take account of risks, including capital flight, disintermediation, and banking sector instability.

17. This section reviews the effects of dedollarization policies. It shows that building on a credible macroeconomic stabilization, exchange rate variability, public debt management that shifts away from foreign currency denomination, and prudential frameworks internalizing the risks of balance sheet dollarization are key ingredients of a durable dedollarization.

A. Macroeconomic Stabilization

18. The first step toward dedollarization is macroeconomic stabilization, focusing on the credible reduction and stabilization of inflation. Stabilization policies include fiscal consolidation and appropriately tight monetary policy to reduce the inflation rate. Fiscal consolidation lessens the need for government borrowing from the central bank, and a tighter monetary policy reduces credit growth. Both policies restrain aggregate demand, resulting in a drop in inflation and, eventually, the appreciation of the real and/or nominal exchange rate. Credible policies curb inflationary expectations and lower the cost of stabilization. Against the backdrop of a durable disinflation, the need for hedging against inflation via holding foreign currency is significantly reduced, and demand for assets denominated in local currency can expand.

19. Dedollarization policies need to be set up differently depending on the exchange rate arrangement.

  • Under a genuinely flexible exchange rate regime, the country is typically seeking to restore monetary policy autonomy.8 The appreciation of the exchange rate following the contraction of money supply during the stabilization process can jumpstart dedollarization. Accordingly, intervention in the foreign exchange market should signal that the central bank is willing to accept nominal exchange rate appreciation (Brazil and Hungary).9 Moreover, monetary authorities can envisage targeting inflation directly and thus enhance the stability of the inflation rate, which can be forecast more accurately, consolidating the benefits of the macroeconomic stabilization.10

  • Under less flexible exchange rate regimes, a credible commitment to a fixed exchange rate would reduce the cost of macroeconomic stabilization because the authorities do not have to pay the cost of building reputation. However, the outcome depends critically on the credibility of the peg. Expectations of devaluation would increase FCDs, while expectations of intervention to help borrowers to pay their foreign exchange debt after a devaluation would continue to encourage borrowing in foreign currency (moral hazard).11 12 In addition, the continued linkage to the foreign currency does not allow for a (fully) autonomous monetary policy. Through dedollarization, the authorities’ may aim to recover seigniorage, adapt the currency in circulation to domestic needs, introduce a more flexible exchange rate regime ultimately, or mitigate risks to financial stability.

20. Dollarization may continue even if successful stabilization has increased the attractiveness of the local currency (Reinhart, Rogoff, and Savastano, 2003). Although stabilization reduces the risks of using the local currency, it does not raise the private sector’s consciousness about the risks of financial dollarization. Also, once the public has become accustomed to using foreign currency in domestic transactions, additional measures may be required to change this entrenched behavior, in particular, in the context of financial dollarization.13

21. As a side effect of macroeconomic stabilization, strengthened confidence in the local currency could also support certain form of financial dollarization. Public expectations of the local currency’s appreciation provide a one-way bet for borrowing in foreign currency, supporting demand for foreign currency loans. As long as banks benefit from easy access to foreign financing (capital inflow), they would also benefit from the appreciation of the local currency by maintaining a short open position, encouraging lending in foreign currency. Although covered interest rate parity may hold in the medium to long term, generally there is arbitrage in the short-run since interest rate differentials often do not cover exchange rate fluctuations, encouraging economic agents to dollarize their balance sheet.

22. The countries of Central and Eastern Europe (CEE) experienced a period of financial dollarization before October 2008 despite a relatively stable macroeconomic framework. Given the prospects of European Union (EU) accession, privatization receipts, and successful macroeconomic stabilization, the expectation of a continuous appreciation of these countries’ currencies vis-à-vis the euro became firmly entrenched. Associated with a positive interest rate differential for the local currency vis-à-vis the euro in order to control inflation, euro borrowing increased steadily, in particular for nontradable goods production. It is important to note that the key variable influencing financial sector asset dollarization is the expectation of real exchange rate appreciation. Therefore, asset dollarization can occur under a peg as long as economic agents expect inflation to remain higher than foreign inflation, as was the case in Croatia.

23. Additional policies and measures can help to reduce dollarization once a credible macroeconomic stabilization plan is in place. They may be market based or forced depending inter alia on the extent dollarization has become entrenched. They include a wide gamut of policies and measures ranging from exchange rate policies to administrative measures to discourage the use of foreign currencies (see also Appendix I, Table I-V).

B. Market-Based Dedollarization Policies

Exchange rate, monetary, and fiscal policies

  • Exchange rate flexibility—An exchange rate that can move in either direction would render the foreign exchange risk more apparent, thus introducing a disincentive to financial dollarization.14 According to Rennhack and Nozaki (2006), a flexible exchange rate arrangement with less bias toward currency depreciation discourages financial dollarization. Hardy and Pazarbasioglu (2006) show that greater two-way exchange rate flexibility may deter foreign currency deposits, as they increase the risk of holding foreign currency assets. Confronted by foreign exchange risk on their assets and liabilities, banks and nonbanks develop hedging facilities over time but, due to the cost of hedging, they would also increase the share of their assets and liabilities denominated in local currency. However, a trend in the exchange rate could entrench the expectation of continuous appreciation/depreciation that could foster dollarization. The impact would nevertheless be different depending on the direction of the trend: the expectation of devaluation would increase liability dollarization (deposit) and currency substitution, while the expectation of appreciation would support asset dollarization.

  • Efficient liquidity management—Strengthening day-to-day liquidity management by the central bank would make local currency more attractive as short-term interest rates become less volatile. The introduction of reserve requirements, standing deposit and lending facilities, and open market operations may help stabilize the domestic interbank rate. Furthermore, issuing medium-term paper as a benchmark for interest rates can improve monetary policy signaling and develop a yield curve (Poland). Similarly, the development of a well-functioning foreign exchange market and an adequate level of official reserves would ensure easy access to foreign exchange, diminishing the need to hold foreign exchange for precautionary reasons.

  • Fiscal consolidation—Fiscal restraint can help reduce the need for government borrowing in foreign currency, thus directly reducing dollarization of government liabilities. It also lessens the need for central bank financing of government debt and contributes to decreasing differential between domestic and foreign interest rates.

  • Unbiased taxation—A tax system that does not treat income from foreign currency more beneficially than income from local currency would not create a bias toward holding foreign currency assets. For example, interest earned on FCDs or bonds should not be exempted from taxation if taxes are levied on similar income from domestic-currency-denominated sources (Pakistan). Financial transaction taxes, if any, should be levied at least equally on foreign and domestic currency transactions (Bolivia).

Public debt management and financial market development

  • Public debt management—Active public debt management that aims at issuing local currency-denominated bonds (if necessary, inflation indexed) would dedollarize the government’s balance sheet, foster the market for domestic paper, and allow for more exchange rate flexibility (Bolivia, Brazil, Israel, Mexico, and Turkey). However, this may result in higher debt service due to the higher interest rates on the government debt to compensate for exchange rate risk and may not be an option for countries where investors are unwilling to take exposure in local currency. It may however, reduce the consequences of devaluation or depreciation of the exchange rate on the official debt service. It may also foster some loan dollarization recurrence as the high local currency interest rate required to switch from foreign currency to local currency for the public debt would discourage private sector borrowing in local currency and encourage private loans in foreign currency.

  • Development of a domestic financial market—A deep and liquid bond market provides flexible alternative investment opportunities to dollar deposits. Increasing the choice of local currency-denominated securities traded on the domestic capital and money markets may contribute to the decrease in dollar-denominated assets. Encouraging the development of the domestic investor base, such as pension funds, would likely support demand for longer-term local currency instruments and markets.

  • Alternatives to dollar-denominated assets—In the absence of confidence in local currency-denominated assets, a credible indexation system can enhance investments in such assets. Ideally, indexation should be to local prices (for example, inflation indexed bonds) because this avoids the reference to foreign currencies and the likely co-movement between government revenues and debt servicing costs (Chile, Colombia, Israel, and Mexico).15 However, country experiences show that indexation may continue even if it is no longer necessary. In these circumstances, widespread indexation can also complicate macroeconomic management by introducing rigidities in the monetary transmission mechanism. The idiosyncrasies of indexation, e.g., inflation is not correctly measured, may also render foreign investors more reluctant to participate in the local market. The timing for eliminating indexation should be determined with great care.

  • Instruments to hedge currency risk—Where exchange controls restrict hedging instruments, residents may have an incentive to build up foreign exchange holdings as an alternate hedge facility (Israel).

Financial policy and prudential regulation

  • Financial liberalization—Freeing banks from administrative controls on the determination of interest rates makes it more likely that domestic real interest rates will be positive, thus helping to promote the use of the local currency (Estonia, Haiti, and Hungary). A more competitive domestic financial system will also enhance the attractiveness of the local currency (Egypt and Poland).

  • Withdrawal of the legal tender status from foreign currency. Dedollarization is unlikely to be achieved if the foreign currency remains the legal tender of the country, since it entrenches its legitimate use in local transactions.16

  • Increased usability of the local currency—To decrease currency substitution, a domestic currency that is attractive for use needs to be provided. This means the continuous availability of domestic currency in the denominations best adapted to the needs of market participants. For example, larger denominations of riel banknotes increased the demand for local currency in Cambodia. Introducing a new currency may enhance the use of the local currency by providing banknote denominations more suitable for local transaction needs than foreign currency banknote denomination.

  • Government operations in local currency—The government should operate in local currency to the extent possible. Raising taxes in local currency can support an increase in the demand for local currency, as can public payments for wages, goods, and services in local currency (Angola). In Peru, the government switched its public lending program to local currency.

  • Use of foreign aid in local currency—Foreign aid if used in foreign currency in the recipient country in may increase dollarization. In small or post-conflict countries, which often have a high degree of dollarization, foreign aid can play a large role in the economy. When possible, the in-country use of the aid should be denominated in the local currency to promote dedollarization.

  • Reserve requirements—A regulatory bias of the reserve requirement framework for FCDs needs to be avoided. Local currency deposits (LCDs) should be subject to conditions which are at least as favorable as those applied on FCDs, while avoiding undue distortions, including those stemming from excessively high reserve requirements. Measures such as requiring banks to denominate reserve requirements on FCDs in local currency,17 remunerating the reserve requirement on LCDs at a higher rate than the FCD reserve requirement (Bolivia, Honduras, Israel, and Nicaragua), or imposing higher reserve requirements on FCDs, would encourage banks to attract LCDs, thereby increasing the deposit interest rate differential.18 Extending the reserve requirement base to nonbank financial institutions involved in dollar intermediation, such as leasing companies, or to unhedged creditors has also been observed in country practices (Croatia).

  • Payments system—The domestic payments system should ensure local currency payments at terms which are at least as favorable as those for foreign currency payments. The central bank should offer convenient and low-cost payment services for domestic currency payments and should not favor payments in foreign currency (Angola and Lao P.D.R.). Peru imposed a 2 percent tax on checks denominated in foreign currency to discourage the use of foreign currency in payments.

  • Prudential regulations—Measures aimed at ensuring proper management of foreign exchange risk and internalizing the true cost of doing business in foreign currency can help to create a level playing field for the domestic currency and eventually encourage dedollarization.19 These include (i) narrow open foreign currency position limits20 (Croatia and Turkey); (ii) higher liquidity requirements on FCDs (Angola and Uruguay) or foreign exchange loans (Croatia); (iii) charging a higher risk premium on dollar deposits participating in deposit guarantee schemes (Uruguay); (iv) higher capital and provisioning requirements on foreign currency loans especially to unhedged borrowers (Angola, Honduras, and Vietnam); (v) lower loan-to-value ratio to secure foreign exchange denominated loans (Hungary), and (iv) stronger collateral and valuation rules for foreign exchange denominated loans.

  • Effective supervision—This can significantly contribute to the internalization of the risks resulting from balance sheet dollarization. Substantial efforts have been made in economies where foreign credit has been growing fast to monitor the risk taken by banking and nonbanking institutions. In particular, stricter internal controls have been imposed on banks to address the risk related to foreign exchange loans, and improved risk disclosure to borrowers has been required.

  • Foreign exchange regulations—No preference should be given to holders of foreign currency over those who only have access to domestic currency funding. For example, allowing residents to make outward capital transactions from their own foreign exchange resources while prohibiting the same transactions for domestic currency holdings can provide incentives to residents to accumulate foreign exchange for future transactions (Philippines).

C. Forced Dedollarization

24. Measures to force dedollarization are not recommended in isolation from market-based measures. Many of the measures listed below involve interference with private contracts, often retroactively, and can diminish the confidence of market participants in the protection of property rights and contracts, with deleterious effects on the credibility of economic policies more generally. Nevertheless, there may be instances when the use of the dollar is so entrenched that market-based dedollarization measures by themselves are insufficient, and more forceful government intervention becomes necessary. However, these measures are likely to prove ineffective or even counterproductive unless accompanied by a strong macroeconomic stabilization plan.

Measures that frequently failed

  • Mandatory conversion of FCDs into domestic currency—These measures frequently resulted in increased capital flight and disintermediation (Bolivia and Mexico in 1982 and Peru in 1985). In some instances, the authorities were obliged to reverse them (Bolivia and Peru).

  • Suspending access to FCDs—The measure prevents depositors from withdrawing their dollar deposits from banks for a certain period (the length is not always determined in advance). Frozen FCDs could be indexed for inflation (Argentina) or redeemable in local currency (Pakistan). These measures jeopardize the trust in the domestic banking system and often lead to disintermediation and capital flight.

Measures that more frequently achieved the expected results

  • Mandatory holding period for FCDs—This measure may be coupled with the introduction of indexed domestic currency instruments (Israel). One likely disadvantage of this measure is that economic agents may be encouraged to keep foreign exchange in cash instead of depositing it on banks’ accounts.

  • Mandatory use of local currency in domestic transactions and for listing the prices of goods and services—The requirement to list domestic transactions in local currency is sometimes coupled with the obligation to make payments in domestic currency (Appendix I, Table IV). Disallowing the use of the foreign currency in internal transactions is a commonly used measure (Angola, Israel, Lao P.D.R., Peru, and many others). However, even if payments continue to be made in foreign currency, displaying prices in local currency can provide an additional impetus to dedollarization (Lao P.D.R).

  • Regulations that discriminate against the use of foreign currency—Measures may include imposing limits on foreign currency borrowing or lending (Angola, Argentina, Israel, Lebanon, Turkey, and Vietnam), and excluding dollar deposits from the deposit insurance scheme (Venezuela).21 Unless the public is well informed about the exclusion of the FCDs from the deposit guarantee scheme, it may encourage bank to accept more FCD.

  • Interest rate control on FCDs—Controls, which cap the interest local banks may offer on foreign currency deposits, could also be used to encourage liability dedollarization (Bangladesh and India).

  • Capital controls—The temporary imposition of surrender requirements on the foreign exchange proceeds of residents may lead to a rapid decline in FCDs.22 Access to FCDs can be limited to certain economic agents (Mexico), or banks’ cross border credit transactions can be prohibited or require prior approval (Appendix I, Table IV).

IV. Exchange Rate Volatility, Prudential Measures, and Dollarization: Cross-Country Evidence

A. Case Studies

25. Argentina and Pakistan have carried out dedollarization in the context of a currency and banking crisis. Although the measures forced dedollarization, they were necessary to mitigate a banking crisis. In 1998, a political shock and the ensuing capital flight, declining reserves, and the drying up of foreign liquidity forced the Pakistani authorities to withdraw the preferential treatment previously provided to FCDs. In parallel with macroeconomic stabilization, the authorities introduced a temporary freeze on FCDs and a number of exchange controls. Later, frozen deposits could be withdrawn in domestic currency or converted into medium-term dollar-denominated bonds. Although the opening of new FCDs was allowed, the preferential tax treatment of FCDs and the convenient forward cover scheme for banks that had encouraged dollarization were discontinued. Following macroeconomic stabilization, conditions normalized and the stock of FCDs represented only a relatively small part of deposits. Similar measures were introduced in Argentina with limited access to deposits by investors. In addition to limiting access to LCDs (corralito), in 2001, the authorities introduced a freeze on FCDs to limit capital flight and prevent bank runs. Although the freeze was abandoned in 2002 after the conversion of existing FCDs into pesos, the share of FCDs in private deposits remained significantly below the pre-crisis level, reaching about 10 percent by 2005.23 In both Pakistan and Argentina, the temporary freeze on deposits eroded the credibility of the banking system, as reflected in the slow pace of private deposit growth in the aftermath of the crises.

26. The success of dedollarization in noncrisis contexts typically cannot be attributed to microeconomic measures.24 In Chile, Israel, and Poland, the process began with a successful disinflation program, leading to a more flexible exchange rate and a monetary policy aimed at lower inflation. Dedollarization was a by-product of macroeconomic stabilization, which created an interest rate wedge in favor of the local currency. However, steps were taken in some or all of these cases to ensure that the prudential framework did not disadvantage the local currency and to shift the composition of public debt away from foreign-currency-denominated bonds.

27. Successful macroeconomic stabilization supported by additional policies helped Israel to dedollarize its economy. The gradual stabilization program in Israel brought down inflation from about 400 percent in 1984 to single digits in the late 1990s. In addition to macroeconomic stabilization, the program aimed at decreasing the foreign-currency-denominated part of the public debt and supported the introduction of hedging instruments to manage foreign exchange risks. These policies were reflected in the lengthening of the maturity of the public debt and a larger share denominated in domestic currency. Israel implemented higher reserve rates on FCDs, and offered alternative investment possibilities by introducing dollar indexed deposits and inflation indexed bonds. In addition, prudential rules were introduced on foreign-currency-denominated loans to ensure adequate hedging of risks. Dollar-denominated deposits declined from 39 percent of total deposits in 1984 to 17 percent in 2002 (Galindo and Leiderman, 2005). Dollarization exhibited considerable inertia even after a decade of low inflation, possibly reflecting lingering doubts about the credibility of monetary policy. Banking supervision played a key role in ensuring that banks met the prudential requirements such as open position limits.

28. Chile’s experience shows that indexation introduced to avoid dollarization may persist even after macroeconomic stabilization has been achieved. Despite the presence of indexation since the 1960s, by the 1990s Chile had successfully stabilized the economy, liberalized the financial sector, and gradually dismantled controls on international capital movements.25 The government also undertook a debt conversion scheme to change foreign-currency-denominated debt into indexed debt. Although Chile adopted inflation targeting and allowed the peso to float freely in 1999, indexation continued to be prevalent. It was ultimately reduced by targeting a fixed nominal interest rate instead of the inflation indexed “real” interest rate. Subsequently, peso deposits increased and reached 90 percent of total deposits in 2010.

29. In Poland, dedollarization was encouraged by a successful reform program and tight monetary policy. In the early 1990s, following episodes of high inflation and frequent step devaluations, Poland embarked on a macroeconomic stabilization program, coupled with financial sector liberalization and the gradual opening of the capital account. In addition, domestic interest rates were raised well above foreign currency interest rates. The sharp reduction in inflation together with interest rate deregulation, which established positive real interest rates, resulted in an increase of zloty deposits to 50 percent of total deposits, while FCDs declined to 30 percent of broad money by end-1993 from 72 percent in 1989.

30. Egypt dedollarized its economy in the context of macroeconomic stabilization and bank reform. The share of FCDs to total deposits declined from 56 percent in 1991 to 22 percent in 1999 and further decreased to 18 percent by 2004. In 1991, the authorities launched a set of fiscal and monetary reforms to reduce inflation and liberalize the financial system, which was heavily controlled (credit ceilings, interest rate controls, and differential reserve requirements). The liberalization of the banking system led to a significant decline in inflation, positive real interest rates, and ultimately to the decline in FCDs.

B. Measuring the Effect of the Nominal Exchange Rate and Real Exchange Rate Volatility on Dollarization

31. We assume that increasing exchange rate volatility reduces both foreign exchange deposits and loans. However, beyond a certain threshold, exchange rate volatility reflects a trend appreciation or depreciation of the exchange rate, which can lead to financial dollarization. Therefore, we estimate a nonlinear (quadratic) equation between financial dollarization and exchange rate volatility to determine the level of volatility beyond which the relation between exchange rate volatility and financial dollarization changes.

32. The data sample is from the Monetary and Banking Time Series database of the IMF. This database includes data on foreign currency deposits for 32 countries and on foreign currency loans for 21 countries during the period November 2001 to March 2009. The sample includes slightly more flexible and managed exchange rates than fixed exchange rates (Appendix IV).

33. Explanation of the equation. In the equation below FDi,t represents the financial dollarization for the country i the time t (the sample is from January 1999 to March 2009), voli,t represents the past 125 days standard deviation of the exchange rateas a percentage of its average for the same period, and ci is the country specific effect of unobserved country specificities. In order to identify different relations between foreign currency denominated liabilities and assets and exchange rate volatility, FDi,t would be alternatively the share of foreign currency loans in total loans and the share of foreign currency deposits in total deposits.26

d(FDi,t)=ci+α1voli,t+α2voli,t2+α3d(REERi,t)+et(1)

The number of variables in equation 1 has been extended to control for the macroeconomic context and autocorrelation. M2 as a percentage of M1 was introduced to control for financial development. Inflation (πi,t) reflects the macroeconomic stability. The lag of the financial dollarization was also introduced to deal with the inertia of the process and autocorrelation.

d(FDi,t)=ci+β0d(FDi,t1)+α1voli,t+α2voli,t2+α3d(REERi,t)+β1M1+β2πi,t+et(2)

34. Table 2 presents two estimates of the determinants of financial dollarization.

  • Exchange rate volatility has a significant impact and the expected sign on liability dollarization. The results are comparable for asset dollarization except that the relation is less significant. The exchange rate thresholds for asset and liability dollarization appear close.27

  • Changes in the real exchange rate are significant in both regressions. They tend to reduce deposit dollarization as expected, but they also seem to reduce asset dollarization.

  • Inflation has a mostly positive influence on liability dollarization with a lag. This supports the assumption that the stability of the macroeconomic environment is conducive to dedollarization. The estimate shows a positive but not significant relation with asset dollarization.

  • Financial dollarization, especially asset dollarization, appears to have a significant inertia as shown by the number of significant autoregressive terms.

Table 2.

Regression Explaining Financial Dollarization

article image

The estimates include all significant consecutive lags.

C. The Effect of Measures to Mitigate Financial Dollarization in Croatia

35. The Croatian authorities relied on a combination of prudential measures and capital controls to mitigate financial dollarization pressures in the context of a stable exchange rate. Since 1994, the authorities have considered the stability of the Kuna as the basis of macroeconomic stability. Although dollarization stabilized, the level of dollarization remained high. To keep the financial sectors’ balance sheet dollarization under control and deal with the risks of dollarization, the Croatian authorities have relied on marginal reserve requirements (MRRs), which targeted specifically liability dollarization, and foreign exchange liquid asset requirements (FX LAR), which targeted the asset side of financial institutions’ balance sheet.28 These measures aimed at encouraging the banks to reconsider their pricing of the foreign exchange risk (increase foreign exchange loan interest rate) and thus internalize the risk of financial dollarization.

  • FX LAR29 obliged the banks to maintain a minimum level of their foreign currency assets in liquid assets on a daily basis.

  • MRRs were introduced in stages between June 2004 and June 2006. The marginal requirement is based on the additional foreign currency funding attracted by banks. They were calculated as the positive difference between the average daily balance of sources of funds received from nonresidents and legal persons in a special relationship with the bank in a particular calculation period and in the initial calculation period. The different MRRs varied in terms of the initial calculation period and rates. All MRRs were maintained in foreign currency accounts with the Croatia National Bank (CNB) and were not remunerated.

Figure 1.
Figure 1.

Asset Dollarization and Prudential Measures in Croatia

Citation: IMF Working Papers 2010, 188; 10.5089/9781455202225.001.A001

Source: Central Bank of Croatia
Figure 2.
Figure 2.

Liability Dollarization and Prudential Measures in Croatia

Citation: IMF Working Papers 2010, 188; 10.5089/9781455202225.001.A001

Source: Central Bank of Croatia

36. Since market participants often find ways to circumvent the regulation, there is a constant need for adjustments to maintain the efficiency of the prudential provisions. For instance, in 2005 the MRR was extended to the mushrooming financial leasing companies, whose sole purpose was to avoid the prudential regulation. Similarly, the MRRs were applied on foreign exchange indexed loans when they developed as an alternative to foreign exchange loans since they were subject to less regulatory constraints.

37. The mix of the prudential measures and capital controls in Croatia appears to have succeeded in shifting risks to the nonfinancial sector at the expense of increased disintermediation.30 The growth in the external debt differential between banks and the nonfinancial sector, indicate a significant shift to nonfinancial corporation borrowing after the introduction of MRR in June 2006. The average monthly growth rate of the external debt of banks fell from 1.81 percent, for the January 1999 to May 2006 period to a negative 0.90 percent for the June 2006 to September 2008 period. In contrast, the rest of the private sector increased its external debt by 2.95 percent a month during this second period, compared with a 1.11 percent monthly average growth rate for the first period. Further evidence of this policy related shift is the move back to bank borrowing since October 2008, after the abolition of all MRRs. Between October 2008 and February 2009, external debt of the banking sector grew by a monthly average of 4.07 percent, while that of the nonfinancial private sector rose by a monthly 0.72 percent.

38. To quantify the impact of prudential measures on dollarization, we added to equation two (section IV C) a set of dummy variables for the dates when prudential measures were introduced. They indicate the average change in asset or liability dollarization in the month of the introduction of the prudential measures compared to the rest of the period. We also introduce the dummy lagged by a period to control for the lagging effect of the measures. We do not consider further lags significant.

39. Financial dollarization shows a moderate response to prudential measures in Croatia (Table 3). Asset dollarization has been more sensitive to the MMR than to the LAR. On the other hand, liability dollarization seems to have reacted to the LAR more than to the introduction of the MRR. However, the lag indicates that the effect of the measures may have been short-lived as liability dollarization increased again rapidly after the introduction of the measures. Interestingly, the increase in the REER is positively correlated with asset dollarization for Croatia.

Table 3.

Regression on the Effect of Prudential Measures on Dollarization in Croatia

article image
Source: Staff calculation.

V. Conclusion

40. A durable dedollarization requires credible macroeconomic stabilization and the implementation of additional policies. Dedollarization entails restoring the proper functioning of a currency. Stable and low inflation increases the confidence in the local currency and renders hedging to preserve its purchasing power through foreign exchange holdings unnecessary. While a certain level of dedollarization can be achieved in the context of a fixed exchange rate regime, a flexible exchange rate arrangement appears to be more conducive to it. An exchange rate policy that allows two-way movements of the exchange rate introduces a disincentive to financial dollarization by rendering the foreign exchange risk more apparent.

41. Specific policies and measures may be necessary to end persistent dollarization. These can end the entrenched behavior of market participants by further increasing the attractiveness of the local currency. Adequate liquidity management decreases the volatility of the domestic interest rates, while public debt management can facilitate savings in local currency denominate instruments. Financial policies, prudential regulations and effective financial sector supervision can contribute to the internalization of the costs of foreign exchange transactions, and administrative measures can ensure that local currency is available in convenient denominations and quality for local payments. Regulations and practices that favor foreign currency transactions could be changed to avoid a bias in favor of foreign currency. In certain cases, it may be necessary to implement measures to force the use of the domestic currency in tandem with macroeconomic stabilization policies.

42. The data appear to provide broad support for the arguments advanced in this paper. While measuring the effectiveness of the selected policies on dedollarization is challenging, as there is no single case of policies targeting solely dedollarization, the analysis provides some convincing results. It provides evidence that exchange rate volatility and inflation have significant effect on dollarization. It also shows that prudential regulation (in conjunction with capital controls) has an effect on both asset and liability dollarization; however, these are relatively short-lived as market participants find ways to circumvent them requiring further adjustments in the regulations.