1 A Primer on Dollarization – Understanding Dollarization

1A Primer on Dollarization


Traditionally, each country and economic territory has issued and circulated its own currency. By its most common definition, a currency is a form of fiat money issued by a government and used within a certain economic region. Money can be described by its functions as a medium of exchange (to buy and sell goods and services), as a unit of account (to keep track of revenues, costs and profits), and as a store of value (to save, and smooth consumption over time).

However, it is important to note that in an increasingly globalized marketplace, where not only goods and services but also people and capital move across borders, governments have lost their monopoly over the currency used among their citizens. This is particularly true for emerging economies.1

Currency substitution, or Dollarization, refers to the use of another country’s currency in exchange of or in addition to the local currency. This phenomenon has become widespread in today’s global economy.

This definition can be expanded to include the degree by which “real and financial transactions are actually performed in dollars relative to those performed in local currency” (Ortiz, 1983).

The term “dollarization” was coined after the US Dollar, since it has traditionally been the most preferred replacement currency of choice. Nowadays, the term “dollarization” generically refers to the use of another country’s currency (not necessarily the US Dollar) in addition to or in exchange of a local currency.

According to FED’s estimates as of 2011, almost $538 billion of US currency in print was held outside the US, and nearly two-thirds of all $100 bills were in use outside US borders (Judson, 2012). Considering that 6.7 billion people live outside the US, this translates into an average of 80.3 US dollars in every non-American’s pocket worldwide.

Another way of examining the use of foreign currency is to observe the currencies used for international trade transactions (trade dollarization). Table 1.1 shows the percentage use of foreign currencies in global trade finance, and provides an illustration of the relative importance of these currencies on an international scale.

In what follows in this chapter, we explain the concept of dollarization in detail and consider several important aspects of dollarization. The remainder of the chapter proceeds as follows. Section 1.2 defines different types of dollarization. Section 1.3 provides an overview of the causes of dollarization, and Section 1.4 discusses the peculiarities of banking in a dollarized economy. Section 1.5 provides a brief historic description of dollarization around the globe, and Section 1.6 discusses the impact of different exchange rate regimes and their relationship with dollarization.

Table 1.1: Percentage participation of several currencies in trade finance.

Currency %
EUR 42.09
USD 31.12
GBP 8.48
JPY 2.41
AUD 2.18
CAD 1.89
CHF 1.83
SGD 1.04
SEK 0.98
HKD 0.96
NOK 0.87
THB 0.76
RUB 0.61
CNY 0.51
DKK 0.51

Data as of September 2012.

Source: SWIFT.

1.2Asset vs Currency Substitution

Before we proceed with a more detailed analysis of dollarization and its causes, it is important to distinguish between two causes that determine the demand for foreign currencies and assets: currency and asset substitution.

Asset substitution is centered on portfolio allocations where investments in foreign currency denominated assets are determined based on their positive and larger expected returns and their marginal contribution to the risk of the overall portfolio.2 This phenomenon is particularly notable in high inflation environments, where local economic agents look for ways of obtaining positive real interest rates.3 In order to achieve real returns, local investors typically select assets that are denominated in hard currencies—for example, the US dollar or Euro—that provide a greater degree of protection against inflation and isolate their portfolios from local currency risk.

On the other hand, currency substitution refers to the use of foreign currency to replace the roles of local currency with regard to its use as a unit of exchange and as a unit of account. We explain some of the reasons behind this in the following chapters.

1.3Official and Partial Dollarization

The dollarization phenomenon is observable in different forms. Some countries may choose to give up their local currency entirely and adopt another country’s currency as legal tender. This type of dollarization is known as “official” or “de jure” dollarization and is observable in more than ten countries (Table 1.2).

Countries decide to embark on a path of full dollarization for a number of valid reasons. Amongst these motives are the desire to eliminate sudden and strong currency depreciations, to reduce the interest rates on their (public and private) international loans by reducing depreciation risk premiums, to attract more foreign direct investment and portfolio investment and to cut the costs of servicing debt. Dollarization can be judged as a commitment towards low inflation, fiscal discipline and transparency (Berg and Borensztein, 2000).

Heavily dollarized economies carefully consider depreciation risk. Economies with high ratios of dollarized loans (public and private) and large capital inflows are particularly vulnerable to sudden depreciation pressures. As we describe in more detail later, a sudden and large depreciation of local currency automatically raises the value (in local terms) of the outstanding balance on a dollar-denominated loan, which negatively affects not only public finances, but also the finances of citizens and firms who have foreign currency denominated loan obligations. Consequently, such currency depreciation can cause banking crises and social unrest as citizens see their wealth eroded.

Table 1.2: A selection of countries and territories that are officially dollarized.

Countries that have adopted the US dollar as legal tender Countries that have adopted the euro as legal tender Countries using a third currency as legal tender
British Virgin Islands Andorra Cook Islands (New Zealand Dollar)
Caribbean Netherlands Kosovo Nauru (Australian Dollar)
East Timora Mónaco Lesotho (South African Rand)
Ecuadora Montenegro Macau (Hong Kong Dollar)
El Salvador San Marino Palestinian Territories (Israeli
Marshall Islands Vatican City Turkish republic of Northern
Cyprus (Turkish lira)
Turks and Caicos

a Uses its own coins.

Source: Edwards and Magendzo (2003).

However, even though currency risk is reduced by full dollarization, it is possible that the interest rates charged on public debt will not decrease due to the existence of other risk factors. For instance, a country’s default or sovereign risk (measured as the spread between yields on local currency bonds over US Treasuries) may implicitly reflect other fundamental information about the health of the economy. This information could include the existence of political instability, a lack of national security, growing income inequality, poverty and social unrest or other such socio-economic variables. Such risks may not be properly assessed by analyzing currency risk alone.

Another perceived benefit of full dollarization is that it allows greater integration of the country’s financial system with the financial markets of developed economies, thus providing a platform for sustainable economic growth in international trade as importers and exporters are able to access currency and settle trades with greater efficiency.

The benefits of full dollarization are not free. The most obvious cost is that fully dollarized countries lose their monetary and exchange rate policy tools. These policies are relevant in certain extreme circumstances; for instance, when the international price of a certain export commodity that is integral to maintaining the country’s fiscal stability falls dramatically. We demonstrate the costs of the loss of control over monetary policy control through a simplified example that looks at the evolution of prices for a ton of iron ore.

Assume that the total cost per ton equals US$ 45. 4 Now assume that we have two countries, one fully dollarized (country A) and another non or partially dollarized (country B). Let us assume the exchange rate in country B, in January 2011, was 2.5 local currency units per US$. This implies that in that month, the production cost was equal to 112.5 local currencies per ton. Finally, assume that the production cost is set in local currency, that it is constant in time and that the components of the costs (basically labor) are kept at January 2011 values.

Figure 1.1: Iron ore price evolution.

As observed in Figure 1.1, in January 2011 the international price of iron ore was at US$ 187 per ton and dramatically fell to US$ 60 by the beginning of 2015. In this circumstance, the ratio of local production cost to international price rises in country A from 24% (45/187) to 75% (45/60). What happens in country B clearly depends on its exchange rate. Let us assume two scenarios, one in which the exchange rate in January 2015 is the same as the one in 2011 (2.5 local currency per USD) and a second scenario where the local currency depreciates to 3.5 per US$.

If the exchange rate of country B remains the same, the ratios of production costs to the international price are the same as those observed in country A: 25% ((45 × 2.5) / (187 × 2.5)) and 75% ((45 × 2.5) / (60 × 2.5)). However, in the second scenario, country B is in a better position: 25% ((45 × 2.5) / (187 × 2.5)) in January 2011 and 54% ((45 × 2.5) / (60 × 3.5)) in January 2015. 5, 6

Many researchers and politicians alike argue that a government’s ability to adjust the exchange rate to manage the negative effects of adverse price movements is important to maintain socio-economic stability, and that the loss of these tools by fully dollarizing an economy can have serious negative repercussions that reverberate throughout the economy and even stunt future growth prospects.

Another well-known cost of full dollarization is the loss of seigniorage for these countries. In general, seigniorage refers to a country’s ability to issue local currency (non-interest paying debt). However, seigniorage is also referred to all the monies the central banks keep from private institutions, for which they do not pay any type of interest (private banks’ reserve requirements, for example). It is for these reasons that seigniorage is normally measured by the increases in base money (currency plus reserves). Seigniorage is complemented by the profits governments make when they issue non-interest paying debt (currency) and buy interest-paying assets (such as government securities, loans to private banks, etc.). This power is lost when full dollarization is adopted. As the central bank’s ability to use monetary policy disappears, so does its ability to issue currency.

Once full dollarization is adopted, central banks effectively need to buy back the stock of domestic currency held by depository institutions. If central banks do not have enough reserves to do this, they would be required to borrow reserves, which would consequently increase transition costs.

Finally, an additional problem with full dollarization is the loss of the central banks’ capacity to act as the lenders of last resort in situations of extreme need (e.g. in the circumstance of systemic bank runs). The ability of central banks to act as lenders of last resort depends on their ability to issue more currency in order to ensure that all deposits in local currency will be fully protected.7 In a fully dollarized economy, the central bank’s response is limited by the level of international reserves it has at the moment of the crisis, and by its ability to obtain dollar-denominated loans from international sources while suffering from liquidity problems.

As we have seen in Table 1.2, fully dollarized economies are not common. For most economies, dollarization is a de facto process during which local currency slowly loses its roles as a medium of exchange, a unit of account and a store of value in favor of other “hard” currencies. During this process, big ticket items such as cars or houses start being denominated in foreign currency; banks begin to offer saving accounts in foreign currency or offer accounts that are indexed to a foreign currency. This type of dollarization is known as “de facto” or “partial” dollarization, and this de facto dollarization is main type of dollarization analyzed in this book.

1.4Measuring the Degree of Partial Dollarization

Research on dollarization (especially partial dollarization) is often hindered by the lack of availability of precise data. One main reason for this paucity of data is that there is no feasible way of accounting for foreign currency circulating in an economy, especially outside the banking system—foreign currency banknotes in the hands of agents in an economy cannot be accurately counted until they are deposited into the banking system. However, not all foreign currency in an economy makes its way to depository institutions.

In partially dollarized economies, economic agents might hold a significant amount of their savings “under the mattresses” for various reasons (i.e. tax avoidance, lack of trust in the banking system, or the fear of being questioned on the origins of their income, particularly if this income was gained through fraudulent means). This makes dollarization measurement a challenge, forcing us to rely on proxies that provide estimates on the level of dollarization in a particular economy.

The most commonly used measures for dollarization in a partially dollarized economy are based on the amount of foreign currency deposits, total liabilities or loans in the banking system.

The ratio of these variables to total deposits (also referred to as deposit dollarization) or to total liabilities (liability dollarization) or to the amount of loans (loan dollarization) in the banking system is used as the measure of the level of dollarization as a whole in an economy. While these ratios do not provide us with an exhaustive picture of the degree of currency substitution in an economy, they are still important proxy measurements that allow researchers to study dollarization from an objective standpoint.

Tables 1.31.5 provide dollarization data on different measures of dollarization, compiled from various sources.

1.5Causes of Dollarization

In this section, we first describe the most likely causes of official (full) dollarization, and then present the causes of partial dollarization.

Edwards and Magendzo (2003) divide officially dollarized countries into two groups based on the reasons that led them to fully dollarize. The first group includes independent nations that have chosen to use a third country’s currency as legal tender; these countries include Ecuador, El Salvador, Andorra, Micronesia, Vatican City and Monaco. The second group encompasses territories or colonies within a national entity, for instance, Puerto Rico, US Virgin Islands, Guam, Greenland and American Samoa.

For small independent nations, official dollarization is adopted due to these countries’ proximity to and political and economic ties with a hard-currency partner.

Table 1.3: Liability dollarization—foreign currency liabilities to total liabilities ratio.

Source: IMF—Financial Soundness Indicators.

Table 1.4: Loan dollarization—foreign currency loans to total loans.

Source: IMF—Financial Soundness Indicators.

Table 1.5: Deposit dollarization—foreign currency deposits to M2 money supply.

Source: Kutan, Ozsoz, and Rengifo (2012).

In the case of Andorra, Monaco or San Marino, it made perfect sense to adopt the Euro as legal tender. The relatively small size of these economies and each country’s economic and political ties with surrounding EU states made using the Euro a necessity.8

There are other economies that are larger than the aforementioned group, yet also adopted a foreign currency for reasons other than economic or political ties. For instance, Ecuador and El Salvador have chosen this option to control their runaway inflationary problems. This appears to be one common factor in almost all episodes of dollarization, official and partial; dollarization follows periods of high and persistent inflation. Central and South American countries are fertile ground for the examination of this phenomenon. A number of Central and South American economies—for example, Peru and Nicaragua—experienced bouts of high inflation in the 80s and early 90s, and to this day, the government and citizens of these countries remain acutely aware of the difficulties of daily life in a period of high inflation. Consequently, it is no coincidence that nowadays, Latin American economies are some of the most heavily dollarized economies in the world. To illustrate, as of 2013, almost 70% of all liabilities in the Uruguayan banking system are denominated in foreign currency.

In the case of Ecuador, the inflation rate fell from an annual rate of 96% to an average of 6% following the adoption of the US Dollar in 2000, as can be observed in Figure 1.2. 9

The same relationship is observable in other partially dollarized economies. Turkey experienced chronic high inflation throughout the 80s and 90s. The average inflation rate in the 80s was 51%, and during the 90s, average inflation climbed to over 75%, hitting 106% in 1994 (Source: IMF-IFS). As a corollary, the dollarization of bank accounts also reached its peak level in that year. The average ratio of foreign currency deposits in the banking system to overall bank deposits rose to 50% in 1994 (see Figure 1.3).

Countries transitioning from a centrally controlled economy to a market economy, such as Russia and Bulgaria, experienced similar bouts of high inflation. Both economies experienced hyperinflationary periods in the 90s, with inflation reaching 1,058% in Bulgaria in 1997 and 874% in Russia in 1993. Both countries also witnessed their dollarization rates reach their peaks in these years. For Russia, the dollarization ratio (measured as a percentage of foreign currency account balances in the banking system to overall bank deposits) was over 40% in 1993; in Bulgaria, this same ratio climbed to over 53% in 1993.

Figure 1.2: Inflation dynamics in Ecuador.

Source: (Until 2005) Index Mundi (http://www.indexmundi.com/g/g.aspx?c=ec&v=71). (From 2006) Ecuador en Números (http://www.ecuadorencifras.gob.ec/indice-de-precios-al-comunidor-2013/).

The link between inflation and dollarization can be understood by realizing that in countries with high inflationary processes, foreign currency provides a shelter from domestic inflation and enables agents to retain the value of their financial assets. In this regard, dollarization emerged not only as a natural hedge, but also as an incentive to save in high inflation economies. Some studies have shown that by providing this incentive, dollarization also helped stem capital flight from those economies (Feige, 2003). Figure 1.4 shows this relationship for the case of Peru.

1.6Dollarization and Banking Dollarization

As mentioned previously, a major problem that academics and industry professionals face when working on the dollarization phenomenon is the limited availability of data, compounded by the issue of inadequate datasets that do not fully capture the true extent of countries’ dollarization. As we have already mentioned, one common metric used to measure dollarization is the ratio of foreign currency deposits in the banking system to the size of overall deposits. Another commonly used metric is theratio of foreign currency deposits in the banking system with respect to a measure of the money supply, either the M1 or the M2. 10 Note that in high inflation environments where the M2 figure might be unnaturally inflated, it may be more appropriate to employ the M1 measure of the money supply rather than M2.

Figure 1.3: Dollarization and inflation in selected economies.

The figures show dollarization ratios and inflation rates in selected dollarized economies. Dollarization is measured as the ratio of foreign currency deposits in the banking system to total bank deposits.

Source: IMIF-IFS, Yeyati (2006).

Figure 1.4: Dollarization and inflation in Peru (1959–2013).

Source: Central Reserve Bank of Peru (https://estadisticas.bcrp.gob.pe/estadisticas/series/anuales).

When data regarding actual foreign currency holdings is unavailable, using these metrics to approximate the extent of the dollarization of an economy is necessary, but sometimes yields inaccurate results. The quality of the approximation depends on many factors, including the ability to open bank accounts in foreign currencies (financial services availability).

Countries where opening a foreign currency account is allowed show high dollarization ratios compared to countries where such accounts are not permitted or are not available.11 This partly explains why Africa, which has low banking penetration and generally low quality financial institutions, shows lower dollarization ratios compared to other regions of the world. Normally, we would expect Africa to show higher dollarization rates than the currently observed levels, due to the aforementioned inflationary factors that often lead to high levels of dollarization. The average number of commercial bank branches per 100,000 adults in Sub-Saharan African countries is 5.6, while the same ratio in the US is 35.3 and in Europe 33.2. Under these circumstances, and given limited access to banks, savers in Africa may choose to keep their foreign currency holdings under their mattresses simply because they cannot gain access to a bank that will accept their deposit (Table 1.6).

Even in countries where foreign currency bank accounts are allowed, experiences of forced conversions by governments or restrictions placed on withdrawals in times of economic crises may hinder depositors’ willingness to keep their foreign currency savings in the banking system.

Argentina’s currency crisis of 2001 offers a pertinent example. During this period, the Argentinean government passed a series of austerity measures known as “corralito;” at the height of the financial crisis, the government froze bank accounts, restricted withdrawals of the local currency (pesos) and ultimately forced foreign currency account holders to convert their foreign currency deposits into pesos. As expected, the amount of foreign currency held within the banking system in Argentina fell dramatically following the crisis (Figure 1.5).

Table 1.6: Banking penetration in sub-Saharan African economies.

Table 1.6 shows the number of commercial bank branches per 100,000 people in selected sub-Saharan countries.

Source: World Bank.

Thus, it is important to note that deposit dollarization in several countries banking systems can hide the real magnitude of dollarization. In aggregate, people understand their political environment and the strength (or lack thereof) of their institutions, and accordingly decide where keep their dollar deposits (and deposits in general) based on their faith in the banking system or government.

Another way to see the existence of hidden dollarization is to observe the constant growth and use of “black” or “shadow” markets. A black or shadow market is an unofficial market where individuals and even corporations change their foreign currency holdings in order to obtain local currencies (sell dollars) or to obtain foreign currency (buy dollars).

Black markets are common mechanisms for investors in particular and individuals in general to exchange foreign currency in countries where government restrictions prohibit such transactions. The percentage difference between the official exchange rate and the prevailing rate in the black market is usually referred to as the black market premium. During times of uncertainty and restricted foreign currency supply through official channels (such as central bank auctions), the premium usually rises. For instance, following the escalation of tension between Ukraine and Russia in the winter of 2015, the black market exchange rate for the US dollar in Ukraine reached 21.5 Hryvnia while the official exchange rate was 16.2 Hryvnia.

Figure 1.5: Ratio of foreign currency deposits to total deposits in Argentina.

Figure 1.5 shows the ratio of foreign currency deposits in the Argentinian Banking system to total deposits.

Source: Yeyati (2006).

In Argentina, capital controls instituted by the government have made it difficult and costly for agents to buy and sell foreign currencies. The government has issued restrictions on the percentage of individuals’ salaries that can be converted into USD. Moreover, as a way to reduce the use of USD, dollar purchases are taxed up to 20%.12 This government attempt to prevent capital flight has been largely unsuccessful, as its citizens continue to dump Pesos to look for more stable currencies to protect their savings and to realize their everyday transactions. Argentinean citizens have developed a well-organized and dynamic black market for the purchase and sale of foreign currencies, especially US dollars. These unofficial markets are informally arranged and organized in certain streets in Buenos Aires (as well as in several other cities in Argentina). One can observe several buyers/sellers openly advertising their exchange services. Moreover, almost all Argentinean newspapers publish daily black market exchange rates, known as “Dólar blue” or “Mercado Paralelo”. Figure 1.6 is a screenshot from the website of La Nación, the largest and one of the most respected newspapers in Argentina. As we can see, this newspaper lists both the official and the unofficial (or Dólar blue) exchange rates.13

In Figure 1.6 one can also observe the spread between buy and sell prices and also the spread between official and parallel market exchange rates. The buy / sell spread is 9.690 and 9.720 in the official market and 14.660 and 14.730 in the black market.

Another example of the black market premium can be seen in Nigeria, where the government has pegged its exchange rate to the US Dollar. As of November 2015, the difference between the official and black market exchange rates was close to 20 Naira, or in other terms, 10% of the official spot rate of 198 Naira per dollar (Source: bra research, Nigeria) (Figure 1.7).

Following the above discussion, it is obvious that in dollarized economies where foreign currency accounts are permitted (common practice in most dollarized economies), banks face particular challenges. In such economies, the problem of a mismatch in banks’ assets and liabilities arises. Even though banks may hedge against this risk by using derivative instruments such as foreign currency swaps, in most dollarized economies, the financial system may not be sophisticated enough for such hedging options to be available. As a result, banks may simply counter their foreign currency liabilities by creating foreign currency denominated assets such as foreign currency denominated loans. However, this practice does not necessarily remove banks’ foreign currency risk but transforms it into another form of risk known as “default risk”. Studies (such as Kutan, Ozsoz and Rengifo, 2012) have shown that because of this transformation, the dollarization of the banking system has a pervasive effect on bank profitability and system stability. We will investigate this dimension of dollarization in more detail in upcoming chapters.

Figure 1.6: Argentinean black market (“Mercado Paralelo”).

This figure shows a snapshot from the Argentinean newspaper “La Nación”.

Source: “Dólar Hoy,” La Nación, December 5, 2015, http://www.lanacion.com.ar/dolar-hoy-t1369.

Figure 1.7: Nigerian Naira official spot rate vs black market exchange rate (“Bureaux de Change Rate”).

Figure 1.7 shows the official Naira-USD rate vs the black market exchange rate (Bureaux de Change Rate) in three Nigerian states (Lagos, Anambra and Plateau).

Source: bra Research, Lagos, Nigeria.

1.7A Brief History of Dollarization Around the Globe

In this section, we briefly describe the circumstances under which dollarization developed in several regions around the world. This section provides the historical framework of the material presented later in the book. We start by looking toward Latin America, and follow with Transition Economies, Asian and African economies.

1.7.1Latin America

The first cases of currency substitution started in Latin America, where local currencies were (partially) substituted with the US dollar, hence the term “dollarization”.

Due to the proximity of Central American countries to the US and close economic ties, Central and South America have naturally been the first regions outside the US where the US dollar was used as a mean of exchange, unit of account and store of value.

For instance, Panama adopted the US dollar as its legal tender as early as 1904. Although the country maintains its own national currency, the “Balboa”, its usage is mostly symbolic and monetary authorities cannot issue Balboa-denominated debt. As a result, the country has enjoyed relatively low levels of inflation, even during the 80s when most economies in Latin America were dealing with hyperinflationary episodes. During this period, Panama enjoyed inflation rates even lower than those in the US did; the average inflation in Panama during the 1981–1990 period was 1.8% while the same rate in the US was 4.7% (Calvo and Veigh, 1992) (Figure 1.8).

However, Panama was an exception in Latin America. Most countries continued using their own currencies as legal tender, while their citizens turned to US Dollars as their primary currency of choice for their savings. For instance, as of 1935, almost 35% of all demand deposits in the Mexican banking system were in US dollar denomination (Ortiz, 1983). Table 1.7 provides the dollarization ratios of selected Latin American economies and their average inflation rates.

Among the dollarized Latin American economies, two stand out from the crowd: Bolivia and Uruguay. Starting in the 80s, these countries have seen dollarization ratios rise to and exceed 50%. As of 2003 (Levy-Yeyati, 2006), the ratio of foreign currency deposits to the overall deposits in the Bolivian banking system was close to 93%. For Uruguay, the same ratio reached as high as 88% for the same year, declining to 69% by 2013 (IMF-FSI.)

In the case of Bolivia, the shift toward dollarization has long historical roots, tracing back to the 1960s when the country started to rely on foreign currency loans from international banks to finance its development programs (Morales, 2003). These loans required payments in the same currency, locking Bolivia in a foreign currency loan trap and further increasing its reliance on foreign currency financing.

The hyperinflationary episodes experienced during the 80s were also another instigating factor in the country’s dollarization story. Inflation in Bolivia reached 11,750% in 1985 (See Figure 1.9). The average inflation rate between 1982 and 1986 was 2,741%. The Central bank in Bolivia conducted monetary policy in local and foreign currency using “the standard instruments and the standard IMF type of monetary programming” (Morales, 2003). However, this policy was not successful. In the second half of the 90s, dollarization ratios increased even further, reaching their peaks in 2003. The effects of the East Asian Financial Crisis and the devaluation of Latin American currencies were major factors in the rise of dollarization ratios at this time.

Figure 1.8: Inflation rates in Panama and in the US (1980–2013).

Figure 1.8 shows the annual inflation rates in the US and in Panama between 1980 and 2013.

Source: World Bank.

In the case of Uruguay, inflationary episodes were not as severe as in the case of Bolivia. Uruguay’s highest inflation rate between 1980 and 2000 was experienced in 1992, when inflation reached 112%. However, inflation was still a persistent phenomenon in Uruguay during the 80s, as illustrated in Figure 1.10. Inflation rates did not fall below 18% (1982) and during the 90s, inflation remained consistently high (with the average around 48%). Inflation rates eventually fell to single digits at the end of the 90s, reaching 5.7% in 1999 (IMF, IFS).

In the case of Peru, the ratio of foreign currency deposits in the banking system remained consistently above 70% until 2001, after which it started a steady decline. The monetary authorities were able to control inflation by 1995. However, even though inflation considerably decreased to single digit levels after 1996, foreign currencydeposits as a ratio to total deposits remained consistently high, settling at a level of 32% in 2014. This level of dollarization is of concern to the Peruvian government, which will need to tackle the negative effect of falling prices and bolster the economy in case low prices remain an enduring problem.

Table 1.7: Dollarization ratios for selected Latin American economies.

Table 1.7 shows the average dollarization ratios and inflation rates for selected Latin American economies subject to data availability.

a Foreign currency liabilities to total liabilities in the banking system.

b Foreign currency deposits to the M2 money supply.

c Ratio of foreign currency deposits to total deposits in the banking system.

d Ecuador fully dollarized in 2000.

e El Salvador fully dollarized in 2001.

Source: IMF-IFS, Bank of Chile, Central Reserve Bank of Peru, Banco Central de Costa Rica, IMF-FSI, Yeyati (2006).

Figure 1.9: Dollarization ratios and inflation rates in Bolivia (1990–2014).

Figure 1.9 shows the inflation rate in Bolivia between 1980 and 2014.

Source: IMF-IFS and IMF-FSI.

Figure 1.10: Inflation rates and dollarization in Uruguay (1981–2013).

The figure shows the inflation rate in Uruguay between 1981 and 2013. Dollarization measure used is the ratio of foreign currency deposits in the banking system to total deposits.

Source: IMF-IFS, IMF-FSI and Yeyati (2006).

Figure 1.11: Inflation rates and dollarization in Peru (1992–2014).

Figure 1.11 shows the inflation rate and dollarization ratios in Peru between 1992 and 2014. The dollarization measure used is the ratio of foreign currency deposits in the banking system to total deposits.

Source: IMF-IFS, Banco Central De Reserva Del Perú.

Figure 1.12: Inflation rates and dollarization in Chile (2004–2014).

Figure 1.12 shows the inflation rate and dollarization ratios in Chile between 2004 and 2014. The dollarization measure used is the ratio of foreign currency deposits in the banking system to total deposits.

Source: IMF-IFS, Banco Central de Chile.

As the first South American country to join the OECD, Chile currently has the second highest per capita income in the South Americas ($14,528 in 2014 compared to $16,806 in Uruguay) and one of its lowest inflation rates (the average inflation rate in Chile between 2004 and 2014 has been 3.15%.) (Source: World Bank and IMF-IFS). The country also enjoys some of the lowest dollarization ratios among other South American countries. Between 2004 and 2014, the average ratio of foreign currency deposits to total deposits in the banking system was around 20%.

1.7.2Transition Economies

The dollarization episodes seen in Latin America in the 70s and 80s were repeated in Transition14 Economies during the 90s. The fall of the Berlin Wall and the end of the communist regimes in Eastern Europe and in former Soviet Bloc economies marked the start of increasingly high rates of dollarization in these countries.

The runaway inflation experienced in these economies was a major cause of the currency substitution processes, much as it was in the case of Latin America during the 80s. As shown in Table 1.8, high dollarization ratios are correlated with high levels of inflation. For instance, among the fifteen countries presented in the table, Croatia has experienced the highest levels of dollarization. As illustrated in Figure 1.13, between 1994 and 2014, the dollarization ratio in the Croatian economy remained persistently high. The average ratio of foreign currency deposits in the banking system to total deposits for this period was 60.9%. It is also no coincidence that this country has experienced one of the highest inflation rates in Europe in recent years. The annual inflation rate in Croatia in 1993 was 1494%. Mainly due to the civil turmoil experienced in former Yugoslav countries at the time, the inflation rate fell to 107.15 in 1994 and sharply dropped to 4.45 in 1995. Although the inflation rate has stayed in single digits from 1995 until today, dollarization rates have not decreased in this country. On the other hand, the Czech Republic, which has had the lowest dollarization ratios in the group (around 9%), also had one of the lowest inflation levels in the sample (less than 6%) and did not experience an inflationary period like Croatia did during its transition process to a market economy.

Table 1.8: Dollarization ratios for selected transition economies.

Table 1.8 shows the average and median dollarization ratios and inflation numbers in selected transition economies during the 1990s and early 2000s.

a Foreign currency deposits to total deposits in the banking system.

b Foreign currency deposits to the M2 money supply.

Source: Bank of Albania, National Bank of republic of Belarus, Bulgarian National Bank, Croatian National Bank, Czech National Bank, Bank of Estonia, Bank of Latvia, Bank of Lithuania, Nat. Bank of the Rep. of Macedonia, National Bank of Moldova, National Bank of Poland, National Bank of Romania, National Bank of Slovakia, Bank of Slovenia, National Bank of Ukraine.

Figure 1.13: Dollarization and inflation in Croatia.

Figure 1.14 shows the evolution of dollarization and inflation in Albania between 1994 and 2008. Despite a drop in the inflation rate from a record high of 33% in 1997, the dollarization ratio in the country kept rising throughout the 2000s. The persistently high ratio of dollarization can be partially attributed to the large flow of remittances from Albania’s neighboring countries, reaching a level as high as 15% of the country’s GDP at the beginning of the 2000s. Remittances make up a large portion of savings in the Albanian economy and are usually deposited in the original currency (Manjani, 2015). This illustrates that although inflation has been a major source of dollarization in Albania (at least during the last years of the 90s), it was nevertheless not the only determinant of dollarization, as can be the case for other regions and countries around the world.

Figure 1.14: Dollarization and inflation in Albania.

Source: National Bank of Albania, IMF-IFS.


Among Asian economies, observed dollarization ratios were not uniform. Developed Asian economies such as Singapore show relatively low rates of deposit dollarization (the ratio of foreign currency deposits to the M2 money supply in Singapore averaged 15% between 2000 and 2004) while exhibiting high liability dollarization figures—an average of 45% between 2009 and 2014.

Some of the less-developed Asian economies such as Cambodia have the highest dollarization rates on the continent. The average ratio of foreign currency deposits to the M2 money supply in Cambodia is 67% between 2000 and 2004 (see Table 1.9).

1.7.4Sub-Saharan Africa

As indicated earlier, measured dollarization rates are relatively lower in Africa; one of the reasons for this situation is the low banking penetration rates in Africa and the use of banking system statistics for dollarization measurements. Table 1.10 shows some available statistics.

1.8Exchange Rate Regimes and Dollarization

The exchange rate regime followed by a country’s monetary authorities has a crucial impact on the country’s dollarization experience. In fact, official dollarization is usually referred to as an extreme example of a fixed exchange rate regime.

In a fixed exchange rate system, the monetary authorities usually peg the value of the local currency to an anchor currency or to a basket of currencies and stand ready to intervene in the forex markets to defend their peg. The dollar and the Euro are usually the most widely used anchor currencies.15 The monetary authority that is authorized to maintain the fixed exchange rate system in a country is also called a currency board. Separate from the monetary authority of the country, a currency board cannot determine monetary policy and does not lend to the country’s government. Currencyboards usually have no restrictions on current and capital account transactions; they buy and sell foreign currency in an attempt to maintain the pegged exchange rate to the US dollar. Examples of currency boards include the Hong Kong Monetary Authority—which also acts as the country’s de facto central bank—and the Bermuda Monetary Authority, which issues the country’s national currency and maintains its peg to the US dollar.

Table 1.9: Dollarization ratios in selected Asian economies.

a Foreign currency deposits as a ratio of M2 money supply (Source: Kutan, Ozsoz and Rengifo, 2012)

b Foreign currency denominated loans to total loans ratio (Source: IMF—financial soundness indicators (FSI)).

Table 1.10: Dollarization ratios in sub-Saharan Africa.

a Foreign currency deposits as a ratio of total deposits in the banking system (Source: IMF—financial soundness indicators [FSI]).

b Foreign currency deposits as a ratio of M2 money supply (Source: Kutan, Ozsoz & Rengifo, 2012).

c Foreign currency liabilities as a ratio of total liabilities as a ratio of total liabilities (Source: IMF—financial soundness indicators [FSI]).

Figure 1.15: Inflation rates in Ecuador, El Salvador and the US.

The figure above shows the annual inflation rates in Ecuador, El Salvador and the US prior to and after the adoption of the US dollar as legal tender. El Salvador adopted the US currency in 2000 and Ecuador in 2001.

Source: World Bank.

In the case of a free float or a flexible exchange rate system, market forces determine the value of the local currency. The monetary authorities do not provide an explicit or implicit guarantee as to the value of the local currency vis-à-vis other currencies.

There are several variations within each system. A free-float is a flexible exchange rate regime where the monetary authorities adopt an observatory approach to the forex markets while in the case of a dirty-float regime, the monetary authorities may intervene in currency markets to control the value of the local currency or the inflation rate.

Variations among fixed exchange rate regimes are also common. The best example is official dollarization. Official or full dollarization can be considered as an extreme case of fixed exchange rate regime where the monetary authority adopts the anchor currency as the legal tender. This can also be regarded as an example of a currency union with the adopted country’s economy, but on a unilateral basis. The monetary authority of the adopted currency usually does not provide any guarantees (such as providing loans in the event of liquidity problems) to the adopting nation’s monetary authorities, thus the country that fully dollarizes gives up its monetary power in exchange for stability in financial and real markets.

In general, over time, the interest rates and the inflation rate in the adopting country converge to those in the adopted currency’s economy. Ecuador and El Salvador both enjoyed inflation rates similar to the US inflation rate upon their adoption of the US dollar as legal tender in 2000 and 2001 respectively. In the case of El Salvador, the inflation rate converged to that of the US almost immediately, while in the case of Ecuador, it took three years for the inflation rate to converge fully to that of the US (see Figure 1.15).