Beckerman P20041011 Ecuador Dollarization

EXTERNAL DEBT, OIL DEPENDENCE AND A NATION’S CURRENCY:
WHY AND HOW ECUADOR DOLLARIZED
Paul Beckerman* October 11, 2004
Abstract. Ecuador’s government fully dollarized its economy during 2000. It had little choice:
full dollarization was the only way it could stop the hyperinflation that had apparently
commenced at the end of 1999. The chain of events that led to the hyperinflation began in 1998,
when plunging oil-export prices and damage to export crops from the 1997-1998 “El Niño”
episode forced sharp exchange-rate depreciation and thrust the banking system into crisis. These
in turn induced monetary issue, inflation, and further depreciation. Exchange-rate depreciation
aggravated the banking crisis by sharply increasing commercial banks’ non-performing dollar
loans, since many were to entities that lacked dollar earnings. Banks had found it necessary to
lend in dollars during the 1990s because their deposit base had become increasingly dollarized,
because repeated exchange-rate depreciation over the 1980s and 1990s had discouraged holding
wealth in sucres. Exchange-rate depreciation had proven necessary to limit the external-
borrowing requirement arising from Ecuador’s overwhelming external debt-service burden. In
this sense, Ecuador’s hyperinflation and forced move to dollarization were consequences of its
exposure to volatile world oil markets and excessive external debt.
The transition to dollarization during 2000 went relatively smoothly, despite some hurdles.
Because dollarization took place at a deeply depreciated exchange rate, the domestic price level
had to rise significantly to adjust to parity with world prices. Relative-price shifts engendered by
the price-level increase caused a wide range of problems, particularly when adjustment of prices
of energy, supplied by public entities, lagged behind production costs. Moreover, by 2003 the
price-level rise had gone so far that it seriously affected Ecuador’s external competitiveness.
Many Ecuadorians are now persuaded that a new national currency should be introduced to deal
with this problem. Given recent experience, however, it is doubtful that a new currency could be
successfully introduced, especially if it were likely again to depreciate. Introduction of a new
currency would likely return Ecuador to the semi-dollarized system that proved so destabilizing
in the late 1990s. Ecuador’s basic competitiveness problems derive, in large measure, from its
oversized public sector, its financial-system weaknesses, and its external-debt burden. Its best
hope of improving competitiveness is to focus on structural reform and careful debt management
in both the public and private sectors. Reintroduction of a national currency cannot realistically
be expected to make a significant difference for longer-term competitiveness.
Data are available from the writer on request ([email protected]).
*Independent consultant. The writer is grateful to many readers of this and earlier versions for
valuable comments, including Alfredo Arízaga; Paulo Nogueira Batista, Jr.; Hernán Cortés
Douglas; James Hanson; Mariana Naranjo; Martha Rodríguez; Jurgen Schuldt; and Andrés
Solimano. The writer alone is responsible for errors of fact and judgment. Views expressed here
do not necessarily reflect those of any institutions with which this writer is now or has been
associated. This paper is an updated and revised version of “Dollarization and Semi-
Dollarization in Ecuador” (World Bank Policy Research Working Paper No 2643, July 2001).
Updating apart, the subject matter is treated in greater detail in Beckerman, Paul and Andrés
Solimano, eds. (2001). Ecuador: Crisis and Dollarization. (Washington, DC: World Bank).
TABLE OF CONTENTS
TABLE OF CONTENTS_____________________________________________________ ii
LIST OF TABLES __________________________________________________________ ii
LIST OF FIGURES ________________________________________________________ iii
1. Introduction _______________________________________________________ 1
2. Ecuador’s pre-dollarization crisis, 1998-1999 ____________________________ 2
3. Oil and external debt: historical origins of Ecuador’s pre-dollarization crisis__ 10
4. Causes and implications of Ecuador’s semi-dollarization __________________ 16
5. Ecuador’s approach to dollarization ___________________________________ 19
6. Ecuador’s transition to dollarization in 2000 and 2001 ____________________ 23
7. Ecuador’s macroeconomic performance since 2000 ______________________ 28
8. Ecuador’s prospects under dollarization: the tradeoff of stability and
competitiveness ________________________________________________________ 33
9. Conclusions ______________________________________________________ 35
REFERENCES _______________________________________________________ 38
LIST OF TABLES
Table Page
Table 1. Selected Latin American economies: Public and publicly 3

guaranteed external debt (per cent of GDP, 1997-2002
Table 2. Ecuador: Semi-dollarization 18
Table 3. Ecuador: The Central Bank’s four balance sheets, March 10 – 21
June 30, 2003
Table 4. Ecuador: Recent macroeconomic performance, 1997-2003 28
iii
LIST OF FIGURES
Figure 1. Ecuador: Consumer prices, exchange rate, and per-capita real GDP, 1989-2002
..................................................................................................................................... 1
Figure 2. Ecuador: Indicators of macroeconomic balance, 1998-2002 (per cent of GDP) 3
Figure 3. Ecuador: Monthly average crude oil-export price (US$/bbl.) ............................. 4
Figure 4. Ecuador: On-shore commercial-bank deposits, December 1996 – February 2000
(US$ million) .............................................................................................................. 5
Figure 5. Ecuador: On-shore commercial-bank credit performing normally and in arrears,
December 1996 – February 2000 (US$ million) ........................................................ 6
Figure 6. Ecuador: Non-financial public deficit, 1991-2002 (per cent of GDP) ................ 8
Figure 7. Ecuador: Per-capita real GDP, real private consumption, and year-end per-
capita public external debt in 1999 U.S. dollars and prices,* 1972-2000 ................ 11
Figure 8. Ecuador: Public and publicly guaranteed external debt, 1971-2000 (US$
million)...................................................................................................................... 13
Figure 9. Ecuador: Nominal- and real-effective exchange rates, December 1970 –
August 2003 .............................................................................................................. 14
Figure 10. Ecuador: Real GDP growth, capital formation, and net imports of goods and non-factor
services, 1976-2002 ....................................................................................................... 15
Figure 11. Ecuador: Evolution of domestic motor-fuel prices (deflated by consumer
prices; December 1999 = 100) ................................................................................. 26
Figure 12. Ecuador: Total remuneration received by a typical private-sector worker
earning the minimum wage ....................................................................................... 27
Figure 13. Ecuador: Monthly real economic activity, January 1992 – November 2003 .. 29
Figure 14. Ecuador: Open unemployment in the three largest cities, March 1998 –
December 2003 ........................................................................................................ 29
Figure 15. Ecuador: Quarterly current-account performance, 1998 Q1 – 2003 Q3 ......... 30
Figure 16. Ecuador: Quarterly fiscal performance, 1998 Q2 – 2003 Q2 .......................... 31
Figure 17. Ecuador, United States: Monthly inflation (percentage change over previous
month, December 1997 – August 2003 .................................................................... 31
Figure 18. Ecuador: Banking-system “reference” interest rates in U.S. dollars,
December 1999 - December 2003 ............................................................................ 32
1. Introduction
1. On January 9, 2000, Ecuador’s government announced that it would fix the exchange rate
and submit legislation to the Congress eliminating the national currency and making the
U.S. dollar the sole legal currency. It had little choice, because this was the only way it could
deal with what had become a massive macroeconomic crisis. The monetary unit (the sucre),
which had been floating since February 1999, had lost nearly two thirds of its U.S.-dollar value
over 1999, and had then lost a fourth of its value over the first week of 2000. Consumer prices
had risen at an annualized 88-per-cent rate in the last quarter of 1999 and were accelerating.
Bank credit operations were in virtual suspension, and a liquidity crisis loomed as banks
prepared for March 2000, time deposits, frozen in March 1999 until one year past their original
maturity dates, were due to be released. Real GDP, which had been stagnating through the
1990’s, was more than 7 per cent lower in 1999 than in 1998 (see Figure 1). With hyperinflation
apparently under way, dollarization seemed inevitable, sooner or later.
Figure 1. Ecuador: Consumer prices, exchange rate, and per-capita real GDP, 1989-2002 1
100 $1,400
Per-capita real
90
$1,200
U.S. dollars at 2000 prices and exchange rate

80
70 $1,000
Annual percentage rate
Grow th of consum er
60
$800
50
$600
40
30 $400
20
Sucre depreciation against the U.S. $200

10
0 $0
1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
Percentage growth rate of consumer prices (Dec.-Dec.)

Percentage loss of sucre value against the U.S. dollar (Dec.-Dec.)
Per-capita real GDP (US$ at 2000 prices and exchange rate)
Data source: International Monetary Fund
2. Two weeks later, popular discontent forced President Jamil Mahuad to resign. Groups
opposed to dollarization figured vociferously in the opposition. After a military-civilian junta
attempted unsuccessfully to seize power, Vice President Gustavo Noboa took office
constitutionally as President. Concluding that there was no alternative, he decided to press ahead
with dollarization. The Government quickly drafted the requisite legislation and sent it to the
Congress. After a quick debate, boycotted by parties opposed to dollarization, the Congress
approved it, and the President signed it in March 2000. Before the end of the year the Central
Bank had retired the outstanding sucre stock, using its dollar holdings. In September 2000, as
1
The percentage loss in the sucre’s value in terms of U.S. dollars is used instead of the exchange rate to
limit the vertical size of the chart.
2
scheduled by the dollarizing legislation, the dollar became Ecuador’s legal currency unit. 2
3. This paper discusses the causes and consequences of Ecuador’s move to dollarization and its
many lessons for other economies. Section 2 discusses the dynamics of the 1998-99 “pre-
dollarization crisis,” focusing particularly on the role of the economy’s “bimonetarism” -- the
fact that the economy, in particular, the banking system, was operating in two currency units, the
sucre and the U.S. dollar, with a fluctuating exchange rate. Section 3 discusses the longer-term
origins of the crisis, notably the use of exchange-rate depreciation throughout the 1980s and
1990s, to sustain the export surplus required to cope with Ecuador’s external debt. Section 4
discusses the spontaneous dollarization that resulted from the heavy use of exchange-rate
depreciation. The high external debt forced Ecuador to rely on exchange-rate depreciation to
maintain an export surplus sufficient to limit the external borrowing requirement, but this
depreciation then discouraged residents from holding sucres, leading to the dual-currency
banking system that proved so destabilizing in the pre-dollarization crisis.
4. Section 5 describes the specific form dollarization took in Ecuador. Ecuador retained its
central bank after dollarizing, and although substantially reduced in size it retains a significant
role in macroeconomic management. Section 6 discusses the problems that arose in the
transition to dollarization, principally the surge in the price level as it moved to parity with world
prices and the accompanying relative-price movements. Section 7 discusses Ecuador’s slow
macroeconomic recovery since 2000. Section 8 discusses Ecuador’s challenges and prospects as
a dollarized economy, and draws conclusions that might apply in other economies considering
dollarization. The price-level increase appears to have reduced the economy’s international
competitiveness, and many Ecuadorians are persuaded that a new national currency must be
introduced to deal with this problem. This section argues that this would probably serve only to
restore the unstable semi-dollarized system prevailing before dollarization, and that Ecuador’s
best hope of improving competitiveness would be to focus on public- and private-sector
structural reform and careful debt management while remaining dollarized. Section 9 draws
lessons and conclusions for other economies.
2. Ecuador’s pre-dollarization crisis, 1998-1999
5. The macroeconomic crisis that precipitated Ecuador’s move to dollarization began in late
1997, when sharply declining oil-export earnings and torrential rains associated with the 1997-98
El Niño weather phenomenon together affected the economy’s fiscal and external flows. Oil
exports slid to US$923 million in 1998 from US$1.8 billion and US$1.6 billion in 1996 and
1997 respectively. Between December 1997 and March 1998 the rains caused massive damage
throughout the coastal areas. Some 300 people lost their lives, about 30,000 lost their homes,
and many more lost income and wealth. A large amount of infrastructure was damaged, and
crop destruction reduced exports and forced additional food imports. Failing water and
sanitation systems increased infectious disease. The resulting tax revenue losses and the
emergency and reconstruction expenditure widened the fiscal deficit (see Figure 2).
2
The sucre retained a de jure legal status. A constitutional amendment would have been necessary to end
it, and the government took the view that the political obstacles to doing so were too daunting.
3
Figure 2. Ecuador: Indicators of macroeconomic balance, 1998-2002 (per cent of GDP)
Per cent of GDP (current account, non-financial public-sector surplus) Annual percentage rate (inflation)
8 100
6 90
4 80
2 70
0 60
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
-2 50
-4 40
-6 30
-8 20
-10 10
-12 0
Balance-of-payments current-account surplus

Central-government surplus
Percentage increase in consumer prices (Dec.-Dec.)
Data source: Central Bank of Ecuador
6. In principle, Ecuador might have considered using external financing to tide over the crisis.
In 1998, however, world financial markets, shaken by successive crises in east Asia, Russia, and
Brazil, were essentially closed to Ecuador, whose ratio of external debt to GDP was – even
before the massive 1999 exchange-rate depreciation -- among Latin America’s highest (see Table
1).
Table 1. Selected Latin American economies: Public and publicly guaranteed external debt (per cent
of GDP, 1997-2002)
Country 1997 1998 1999 2000 2001 2002
Argentina 22.9 25.9 29.7 30.5 26.5 73.0

Bolivia 52.1 50.6 51.2 49.8 38.8 43.3
Brazil 10.8 12.5 17.1 15.5 18.4 21.3
Colombia 14.5 17.0 23.4 24.8 26.8 26.2
Ecuador 65.1 66.4 99.0 81.6 58.4 51.1
Peru 32.5 34.0 37.7 36.0 35.0 36.0
Source: World Bank
7. In 1998, Ecuador recorded a US$1.7 billion deficit (8.7 per cent of GDP) in net exports of
goods and non-factor services, compared with a US$837 million surplus (4.4 per cent of GDP)
and a US$55 million deficit (0.2 per cent of GDP) in 1996 and 1997 respectively. Twice during
1998, in March and September, the widening current-account deficit (see Figure 2) forced the
monetary authority to devalue outside the pre-announced band it had maintained since 1994. In
February 1999, reserve loss forced the monetary authority to float the exchange rate. It delayed
doing so for several crucial weeks as it waited for the Congress to reach agreement on the 1999
4
annual budget. This was because the float would have rendered the budget’s exchange-rate
assumption was unrealistic and would therefore have made it even harder to secure legislative
agreement. Meanwhile, the fiscal deficit widened from 4.6 per cent and 3.5 per cent of GDP in
1996 and 1997 respectively to 7.3 per cent of GDP in 1998 (see Figure 2), largely because oil-
based revenue slid from 8.3 per cent and 6.4 per cent of GDP in 1996 and 1997 to 4.6 per cent of
GDP in 1998.
8. Exchange-rate depreciation proved destabilizing for Ecuador in 1998 and 1999 because its
banking system had become semi-dollarized. With the financial-sector liberalization that had
taken place over the 1980s and early 1990s, Ecuadorian banks were permitted to carry out
operations in U.S. dollars, and they did so, accepting growing proportions of their deposits and
making growing proportions of their loans in dollars. Depositors increasingly preferred dollar
accounts to protect their wealth from exchange-rate depreciation. Borrowers accepted dollar-
denominated loans because, since the pre-announced crawling-peg policy seemed prospectively
to limit exchange-rate depreciation, their interest rates seemed significantly lower. Commercial
banks made a point of carefully matching assets and liabilities by currency, instituting asset-
liability committees whose main task was to ensure that dollar-denominated deposits were
adequately backed by dollar-denominated securities and loans.
Figure 3. Ecuador: Monthly average crude oil-export price (US$/bbl.)
$35.00
Dollarization announcement --->

$30.00
$25.00
$20.00
$15.00
$10.00
$5.00
Jun-95
Dec-95
Jun-96
Dec-96
Jun-97
Dec-97
Jun-98
Dec-98
Jun-99
Dec-99
Jun-00
Dec-00
Jun-01
Dec-01
Jun-02
Dec-02
Jun-03
Dec-03
9. Over the course of 1998 the banking sector became the focus of the crisis. The oil-sector
downturn and the El Niño damage affected bank portfolios, but the exchange-rate slide
eventually proved even more damaging. While the banks were generally matching their own
sucre and dollar assets and liabilities, many of their dollar borrowers were unmatched. In
retrospect, this was inevitable. As the crisis progressed, depositors allowed the exchange-rate
depreciation to move their wealth (measured in dollars) out of sucres and into dollars (see Figure
4). As this happened, the banks themselves ran through the borrowers whose earnings were
likely to keep pace with the exchange-rate depreciation and lent increasingly to borrowers whose
earnings were not. 3 As a consequence, exchange-rate depreciation rapidly increased the banks’
3
The fact that city taxis were sometimes purchased using dollar-denominated bank credit was often cited to
characterize the problem. Exchange-rate depreciation affected taxi owners’ ability to meet debt service
obligations not only by increasing their sucre equivalent but also by making gasoline more costly in sucres.
5
non-performing portfolios (see Figure 5). 4
10. Supervisory inadequacies compounded the problem. In retrospect, as they liberalized the
financial system, the authorities ought to have strengthened prudential regulation more than they
did. Until late 1998 Ecuador possessed no deposit-insurance system, and the authorities had no
way to deal with insolvent banks short of the traumatic step of liquidation. Although the
Banking Superintendency had undergone substantial technical improvement as and after
financial liberalization took place, it was still unable to cope effectively with such long-standing
practices as connected lending; inadequate capital-adequacy standards; and outmoded,
unrevealing accounting practices. Nor could it cope with such newer practices as predatory
interest-rate competition, growing use of off-shore banking (now fully legalized), and dollar-
denominated lending to entities vulnerable to exchange-rate depreciation. Indeed, it is important
to recognize that the problem was larger than the Banking Superintendency’s lack of capacity.
For example, it is sometimes argued that banking inspectors ought to have questioned lending in
dollars to borrowers exposed to exchange-rate depreciation. Had they done so, however, they
would have made it more difficult for the banks to acquire dollar assets to back their growing
dollar-deposit bases. In retrospect, it would have been better if the banks had discouraged the
growth of dollar deposits beyond what they could back with the right kinds of asset. At the time,
however, it would have been difficult for banking supervisors to argue that there was a
significant risk that the exchange-rate policy – a float within a pre-announced band -- would
prove unsuccessful.
Figure 4. Ecuador: On-shore commercial-bank deposits, December 1996 – February 2000

(US$ million)
$6,000
U.S.-dollar deposits Sucre deposits
$5,000
$4,000
$3,000
$2,000
$1,000
$0
Mar-97
Mar-98
Mar-99
Mar-00
Dec-96
Jun-97
Sep-97
Dec-97
Jun-98
Sep-98
Dec-98
Jun-99
Sep-99
Dec-99
Banks eventually repossessed many taxis, and some ended up with the Central Bank when banks defaulted
on their liquidity support.
4
As in most banking systems, banks classified their loans, subject to Superintendency review. To the
extent banks tended to err on the side of optimism, Figure 5 should be regarded as indicative rather than
precise.
6
Figure 5. Ecuador: On-shore commercial-bank credit performing normally and in arrears,

December 1996 – February 2000 (US$ million)
U.S.-dollar credit performing normally U.S.-dollar credit in arrears

$6,000
Sucre credit performing normally Sucre credit in arrears
$5,000
$4,000
$3,000
$2,000
$1,000
$0
Dec-96
Mar-97
Jun-97
Sep-97
Dec-97
Mar-98
Jun-98
Sep-98
Dec-98
Mar-99
Jun-99
Sep-99
Dec-99
Mar-00
11. Exchange-rate depreciation was also a problem for the fiscal accounts. One peculiar
consequence of the declining oil prices was that the stabilizing effect of exchange-rate
depreciation on the fiscal accounts diminished sharply. When oil-export prices were at more
“normal” levels, the government’s net foreign-exchange flow – basically revenue from oil and
tariffs less expenditure on debt-service – was positive. In 1998, however, low oil prices meant
that the fiscal accounts’ net foreign-exchange balance was far smaller. Oil and trade-based
revenue less interest due was US$692 million in 1998, compared with US$1.1 billion and
US$840 million in 1996 and 1997 respectively.
12. The administration of President Jamil Mahuad, which had taken office in August 1998 after
winning a two-round election, drafted emergency legislation to address the banking and fiscal
crises. In November 1998 it submitted emergency legislation -- that is, legislation requiring
rapid consideration -- to the Congress incorporating drastic reforms to deal with the fiscal and
the banking crises. The fiscal reform was the complete replacement of the personal and
corporate income tax with a new one-per-cent tax on all financial transactions, including checks
and financial transfers. The financial-transactions tax was expected to raise more revenue than
the poorly performing income taxes. The Government would have preferred not to eliminate the
income taxes and to institute the transactions tax at a lower rate, but it controlled far less than a
majority in Ecuador’s (unicameral) legislature, and this was the only revenue measure for which
it could secure a legislative majority. The banking reform was the establishment of a new
deposit-insurance system, providing a cash guarantee for all banking-system deposits, even for
commercial banks’ trade-credit lines, and establishing a Deposit Guarantee Agency (DGA) with
the legal and technical capacity to carry out bank-rescue operations.
13. The Congress approved the legislation, and it went into effect almost immediately. The new
Deposit Guarantee Agency began operations on December 1, 1998 and immediately took control
7
of the nation’s largest bank, a Guayaquil-based bank heavily affected by all aspects of the crisis.
The Agency kept the bank open and operating, recapitalized through an infusion of bonds
provided by the central government and continuing Central Bank liquidity support (collateralized
by the bonds). Over the first few months of 1999, the Deposit Guarantee Agency took control of
several other banks, keeping some in operation and liquidating some of the smaller ones. The
transactions tax went into effect in January 1999. As anticipated, its revenue yield was higher
and steadier than the income tax. Given the reality that wealthier people carry out more
checking transactions, a case could be made that it was progressive. Moreover, it provided the
tax authorities a wealth of information relevant for collection of other taxes. Unfortunately, it
also discouraged financial activity and bank deposits, encouraging their contraction at a moment
when the commercial banks were in particular trouble.
14. The basic problem of the deposit insurance was its fundamental lack of credibility. It
covered the full value of all commercial-bank deposits, including those held off-shore, and
provided that deposits of banks deemed to have failed must be paid rapidly in cash. Since no
fund had been accumulated, and the entire banking system was manifestly under severe pressure,
depositors inevitably asked whether the authorities could conceivably cover potential losses.
While it is impossible to know whether it slowed the ongoing decline in overall deposit stocks
(see Figure 4), it is clear that by early 1999 many commercial banks came under very intense
liquidity pressure. Several smaller banks were allowed to fail, and the government was able to
pay their depositors. The Central Bank provided liquidity credit to larger banks to keep them in
business, among other reasons because their deposit-insurance liability would have been too
large for the government to handle.
15. Once the exchange rate began floating freely in mid-February 1999, it depreciated sharply,
losing about 30 per cent of its U.S.-dollar value over four weeks, generating inflationary
pressure: consumer prices were 13.5 per cent higher in March than in February. The
depreciation induced acute crisis for several banks, especially a Guayaquil-based bank called
Banco del Progreso that had undertaken aggressive high-interest-rate operations for several years
under a flamboyantly enterprising owner. In mid-March the Government first declared a bank
holiday, and then, several days later, a deposit freeze, to last one year for checking and savings
deposits and one year from the original maturity date for time deposits. The President promised
to hire international auditing firms to determine the banks’ true capital positions and to have the
Banking Superintendent take action on the basis of the results. Although the freeze temporarily
reversed the exchange-rate depreciation and slowed the inflation, it had a devastating effect on
real economic activity. Banking-system credit operations, already shrinking, virtually ceased
with the freeze. This, together with curtailment of access to transactions balances, largely
explains why real GDP tumbled so deeply in 1999 and why urban unemployment roughly
doubled between mid-1998 and December 1999.
16. On July 30, on the basis of the auditing firms’ results, the Banking Superintendent stated in a
televised speech that of the 32 banks examined (including three already closed), nineteen were
sound, six more would be closed and four relatively large banks would undergo enhanced
monitoring, recapitalization and restructuring under the Deposit Guarantee Agency. Three of
these four banks failed within two months, however. The authorities kept these three banks
open, merging them with other banks the Agency already owned. In all, by December 1999
8
banks representing more than 60 per cent of total commercial-bank assets were in public hands.
17. The collapse of credit operations and declining real GDP may have contributed as much as
exchange-rate depreciation to the plunge in merchandise imports. During 1999 merchandise
imports ran at about half their 1998 rate. Meanwhile, oil-export prices recovered, offsetting
declining non-oil exports. As a result, the current account swung from an 10-per-cent-of-GDP
deficit in 1998 5 to a surplus exceeding 6 per cent of GDP in 1999 (see Figure 2), a strikingly
large adjustment. Meanwhile, however, the public finances remained under intense pressure.
The overall 1999 non-financial public deficit reached 6.4 per cent of GDP, a modest
improvement from the 7.3-per-cent deficit recorded in 1998 (see Figure 6). Oil-based revenue
reached 7.7 per cent of GDP, compared with 6.4 and 4.6 per cent in 1997 and 1998. It would
have been higher, but revenue from domestic motor-fuel sales was relatively lower because the
Government froze domestic motor-fuel prices in July 1999 to settle a strike by transport workers.
The exchange-rate depreciation forced up interest charges: external interest due reached 5.9 per
cent of GDP in 1999, compared with 3.6 per cent in 1998. Largely because oil revenue
recovered, the primary public surplus reached 2 per cent of GDP, compared with a 2.5-per-cent
deficit in 1998.
Figure 6. Ecuador: Non-financial public deficit, 1991-2002 (per cent of GDP)
10.0
8.0
6.0
4.0
2.0
0.0
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
-2.0
-4.0
-6.0 Non-financial public-sector primary surplus

Domestic interest due
-8.0 External interest due
Non-financial public-sector deficit
-10.0
18. In late August 1999 the crisis took on an additional dimension when its tightened cash flow
forced the Government to interrupt servicing of its Brady bonds, issued in 1994 as part of a
comprehensive debt and debt-service reduction operation. The Government called on
bondholders to discuss relief. Brady bonds were designed to make coordinated negotiations
difficult to organize, however, and the Government could not organize discussions during the
month-long period before bondholders could take legal action or request payment from
5
Merchandise imports had been relatively high in the latter part of 1998 because importers, anticipating
that exchange-rate depreciation was likely in 1999, purchased for inventory.
9
collateral. At the end of September, the Government announced that it would service only the
uncollateralized bonds, and invited holders of collateralized bonds to request the collateral.
Bondholders took the view that it was wrong to treat one class of creditor preferentially in a
default situation, and elected to accelerate the principal. This activated cross-default clauses,
thrusting Ecuador effectively into default on all its external bonds (US$6.5b out of total public
external debt of US$13b). In the weeks following, the Government succeeded in opening a
dialogue with selected bondholder representatives, but could not begin serious negotiations
(although bondholders did refrain from litigation). Bondholder representatives were
understandably unwilling to discuss any deal on which they would have to stake their own
credibility. (Later, in August 2000, the Government got around this obstacle by making an
exchange offer directly to bondholders, who accepted new bonds against the Brady bonds at
60 cents on the dollar.)
19. Over 1999, the government found it increasingly difficult to cope with the banking crisis.
With economic activity now seriously disrupted, the government felt it had little choice but to
unfreeze transactions balances. Within weeks of the freeze announcement, the authorities began
unfreezing checking deposits, and over the remainder of 1999, they accelerated the unfreezing of
checking and savings deposits. They also allowed some conversion of time deposits into
marketable “Reprogrammable Certificates of Deposit,” which could be used to service bank
loans at par. Inevitably, as their funds became available, many depositors withdrew. Holders of
savings deposits – typically people of modest means – feared they could lose their holdings if
they waited. Between April and December, some US$465m (about 3.2 per cent of 1999 GDP) in
checking and savings deposits, both on- and off-shore, were unfrozen. About a third of these
deposits were estimated to have been fully withdrawn from – i.e., not deposited back in -- the
banking system, fuelling capital flight and exchange-rate depreciation. In November 1999
Ecuador’s Constitutional Tribunal ruled that the freeze had been unconstitutional, and had to be
reversed as soon as possible.
20. The Central Bank had little choice but to provide the banks liquidity support. From
December 1998 through December 1999 the Treasury issued about US$1.6b in bonds to
recapitalize banks that were still open and to finance payment of guaranteed deposits of banks
that had been closed. Banks recapitalized in this way used a large proportion of the bonds they
received in liquidity operations with the Central Bank, rediscounting them or engaging in
repurchase operations. The Central Bank acquired some US$1.2b (on the order of 10 per cent of
1999 GDP) in bonds. The monetary base grew 136 per cent over 1999 (it grew at annualized
rates of 101 and 522 per cent respectively in the third and fourth quarters), with the liquidity
operations accounting for more than all of this. This was the monetary mechanism through
which exchange-rate depreciation and inflation worsened over the year. The monetary
expansion would have been even larger, but the Central Bank sterilized about a quarter of the
liquidity support it provided by selling its own interest-bearing obligations. These were acquired
mainly by a small group of “better” banks that had excess resources as a consequence of
depositors’ “flight to quality” within the banking system.
21. By the end of 1999, virtually all checking and savings accounts had been unfrozen.
Unfreezing of time deposits was due to commence in mid-March 2000 (deposits were to be freed
a year from their original maturity date). Well before the end of 1999, however, the authorities
10
concluded that they would need somehow to limit the unfreezing of time deposits, because banks
would possess only a fraction of the resources they needed to meet withdrawal demand.
Between the banks’ on- and off-shore offices, about US$2.2b in deposits were due for release
from March through June 2000. Heavy withdrawals would either thrust banks into failure, with
contagion effects throughout the system, or oblige the Central Bank to provide liquidity credit
and create money, intensifying exchange-rate depreciation, inflation, and capital flight.
22. Late in November 1999, with the exchange rate depreciating rapidly, the Central Bank
announced that it would tighten policy -- in particular, it would no longer automatically provide
banks liquidity credit nor try to moderate interbank interest rates. This change had revealing
consequences. As expected, interbank interest rates surged. Throughout the year, the Central
Bank had been absorbing money through open-market operations in its own one- and two-week
instruments, and so had taken on a large debt of its own. As a consequence, its own interest bill
had become a significant source of monetary expansion. This meant, however, that the soaring
interest rates resulting from the tightened policy now served, perversely, to increase the rate of
monetary expansion. This made it all too evident that the Central Bank had lost the capacity to
manage the money supply and the exchange rate. Financial markets responded accordingly: over
December and then during the first week of January the exchange rate went into apparent free
fall, depreciating massively on a daily basis, precipitating the decision to dollarize.
23. The banking system’s semi-dollarization played a central role in the crisis dynamics, turning
what would have been a severe economic crisis into a catastrophe. Once exchange-rate
depreciation got under way in 1998, the banking system’s semi-dollarization made its
consequences more explosive, first because it directly increased the sucre equivalent of the
money supply’s dollar component, and then because borrowers’ unmatched positions meant that
depreciation increased non-performing loans and reduced service payments. Once depositors
realized that the banks were in crisis, they withdrew, intensifying the banks’ illiquidity. Indeed,
to the extent they withdrew from dollar accounts, exchange-rate depreciation swelled the sucre
equivalent of what they could withdraw. Over the course of 1999, the banking crisis simply
overwhelmed the authorities’ capacity to cope. It is a retrospective truism that the deposit freeze
was a disastrous policy error. All the same, it is doubtful that the authorities had it within their
power to prevent the banking-system meltdown once the exchange rate began to slide.
3. Oil and external debt: historical origins of Ecuador’s pre-dollarization crisis
24. Even as it went into its “pre-dollarization” crisis, Ecuador’s economy was performing poorly.
Over the 1980s and 1990s, Ecuador’s GDP grew more slowly than the public external debt and
at about the same rate as the population. Per-capita real private consumption remained
essentially unchanged. Over the 1990s the measured poverty incidence worsened, partly because
inadequate and volatile public revenue and the debt-service burden together constrained
government provision of health, education and social-welfare services. The measured
(consumption-based) poverty incidence rose from 34 per cent in 1995 to 46 per cent in 1998.
Extreme poverty (food consumption below minimum nutrition standards) rose from 15 to 17 per
cent.
11
Figure 7. Ecuador: Per-capita real GDP, real private consumption, and year-end per-capita public
external debt in 1999 U.S. dollars and prices,* 1972-2000
1400
1200
1000
800
600
400
200
0
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
Per-capita real GDP Per-capita real private consumption Per-capita public external term debt (2000 U.S. dollars)
Data source: International Monetary Fund, World Bank

* Data are reflated using the GDP deflator.
25. If it is true, as Section 2 above argues, that during 1998 and 1999 Ecuador’s dual-currency
system amounted to grease on the exchange-rate skids, then whatever it was that brought the
dual-currency system into existence bears a substantial share of the ultimate blame for the pre-
dollarization crisis. Ecuadorians came increasingly to prefer the dollar to denominate their
wealth over the 1980s and 1990s, for the obvious reason that it preserved its value more reliably
than the sucre (see Section 4 below). The reason the sucre depreciated so heavily over the 1980s
and 1990s was that this was necessary to hold the external flows where they needed to be to limit
the external-borrowing requirement, given the overwhelming external debt-service burden. On
this logic, one underlying cause of the pre-dollarization crisis and the forced move to
dollarization is Ecuador’s external-debt accumulation over the 1970s and 1980s.
26. Moreover, what made it possible for Ecuador to accumulate external debt so rapidly during
the 1970s was the start of large-scale oil exports in the early 1970s. Ecuador’s oil economy also
figures heavily in the causes of the crisis, in several ways. First, oil dependence implied that the
fiscal and external accounts were vulnerable to volatile world markets. This meant that
whenever oil-export proceeds slipped, as in 1997, the authorities faced difficult choices.
External adjustment would require either exchange-rate depreciation or, if available, additional
external financing. Fiscal adjustment would require offsetting tax increases, expenditure cuts, or,
if available, additional external financing. Worse, when oil exports commenced in the 1970s, the
military regime then in power reduced taxes and increased public employment and subsidies, all
with the objective of hastening Ecuador’s development. These changes were premised on
relatively high oil-revenue flows. To the extent oil revenue turned out lower, the fiscal
consequences were accordingly aggravated. The expanded public sector that the military
governments set in place in the 1970s remained largely unadjusted thereafter. One way or
another, Ecuador has had to bear the cost of its oversized public sector, either through taxes or
borrowing. Higher taxes made productive sectors less efficient. Increased borrowing served to
increase the public debt. Together, the oversized public sector and its accompanying debt have
been a drag on the economy’s performance over the past three decades.
12
27. Commodity booms and crashes have played a significant role in Ecuador’s history since its
establishment of the Republic in 1830 following the break-up of the Gran Colombia
confederation. In the latter part of the Nineteenth Century Ecuador became one of the world’s
leading cacao producers, and this became the basis of the growth of the coastal export economy.
Just after World War I, however, world cacao prices collapsed as other producers entered the
international market, leading to a sustained period of economic stagnation for Ecuador. Growth
resumed in the late 1940s, when Ecuador took a significant share of the world banana market.
Growth remained vigorous until the late 1950s, when declining world banana prices induced
lower growth rates -- although to this day Ecuador remains the world’s largest banana exporter.
28. Discovery of significant oil reserves in Ecuador’s Amazon region in the 1960s led to the next
commodity cycle. Production and export commenced in 1972, fortuitous timing to catch the
surge in world oil prices after October 1973. In 1972 the Armed Forces seized power, saying
that they intended to retain power to ensure that oil revenues were used for national
development. Underscoring their nationalist purpose, they renegotiated contracts with foreign
oil companies, and placed several officials from the earlier civilian governments on trial for
corruption associated with negotiation of the contracts. In 1973 they took Ecuador into the
Organization of Petroleum Exporting Countries (OPEC). On the basis of the surging oil revenue,
they increased public sector employment and capital formation rapidly, raising overall
government expenditure by about two thirds between 1972 and 1975. At the same time they
reduced domestic taxation: non-oil public-sector revenue fell from 18.7 per cent of GDP in 1972
to 13.8 in 1975, while oil revenue rose from 2 to 8.4 per cent. They applied part of the oil
earnings to subsidize domestic electricity and oil derivatives. From 1970 to 1977 annual real
GDP growth exceeded 9 per cent (compared with just under 6 per cent in the 1960s). During
1974 and 1975, rising aggregate demand induced inflationary and balance-of-payments
pressures. The military government, which had never developed a solid popular base, found its
political position too narrow to maintain itself in power, and in the late 1970s it decided to
depart. A convention drafted and approved a new constitution, and in 1979 a new constitutional
government after elections.
29. As the economy grew, Ecuador’s private sector -- mainly commercial banks -- began
borrowing heavily from foreign banks engaged in “recycling” OPEC surpluses. Meanwhile,
however, oil revenue slipped as world recession drove down world oil prices. A 1974-75 El
Niño episode affected coastal agriculture and fishing, reducing government revenue. In the latter
part of the 1970s, to avoid raising non-oil taxes and reducing expenditure and subsidies, the
Government also began heavy external borrowing to finance its deficit. At the end of 1979 the
overall public external-debt stock stood at US$4.5b (about 28 per cent of GDP), compared with
US$324m (20 per cent of GDP) at the end of 1970 (see Figure 8).
13
Figure 8. Ecuador: Public and publicly guaranteed external debt, 1971-2000 (US$ million)
16000
14000
12000
10000
8000
6000
4000
2000
0
1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999
Data source: World Bank
30. Over the first half of the 1980s, like many other developing economies, Ecuador underwent
an external-debt “inflation” as the “debt crisis” took hold. Public and publicly guaranteed debt
roughly doubled between 1980 and 1984. This was partly because oil-export prices had declined
after their 1979 surge, widening the public deficit. More important, however, the interest due on
Ecuador’s floating-rate external debt surged as the United States Federal Reserve’s vigorous
contractionary policy drove world interest rates to unprecedented levels. Non-interest
expenditure began to increase after August 1979, when new government increased public-sector
wages. To make matters worse, the 1982 devaluation and prospects of further devaluation
affected private-sector external borrowers (mostly banks), who persuaded the authorities to
provide relief. In 1983, in the context of an IMF program, the Central Bank assumed about
US$1.5b in private external debt (about 11 per cent of 1982 GDP) in exchange for sucre-
denominated debt. With subsequent devaluations, this “sucretización” policy amounted to a
substantial (and highly unpopular) transfer to private debtors (see Bayas and Somensatto 1994). 6
(In 1992 the Central Bank transferred the debt remaining from the sucretización transactions,
amounting then to about 8 per cent of GDP, to the Treasury.)
31. Ecuador dealt with the debt crisis in broadly the same ways as other affected economies,
undertaking IMF programs to control the public and external deficits while negotiating with bank
creditors to reschedule amortization due and secure new credit to cover interest due. Exchange-
rate adjustment inevitably figured heavily in the policy mix as the authorities struggled to
maintain a sustainable export surplus and so limit external borrowing. Soon after the first
devaluation in August 1983, the authorities carried out further devaluations. In the mid-1980s a
6
Even so, sucretización proved insufficient to resolve the private debt problem. By 1986, a large
proportion of the private sector’s sucre obligations to the Central Bank had become non-performing. That
year, the Government authorized commercial banks to use external debt valued at par to service obligations
to the Central Bank arising from the “sucretized” debt -- a further transfer from the public to the private
sector, since this debt was purchased at substantial discounts.
14
“liberalizing” government attempted to implement a free exchange-rate float. It retreated to a

fixed rate in mid-1987, however, when the exchange rate became too volatile. On the whole,
however, over the 1980s the authorities did succeed in bringing about significant exchange-rate
depreciation in real-effective terms (see Figure 9).
Figure 9. Ecuador: Nominal- and real-effective exchange rates, December 1970 – August 2003
(real) 1990 = 100 (nominal) February 2000 = 100

120.0 180.0
160.0
100.0
140.0
80.0 120.0
100.0
60.0
R eal-effective ex change rate --> 80.0
40.0 60.0
40.0
20.0
Nominal-effective ex change rate --> 20.0
0.0 0.0
Dec-70
Dec-72
Dec-74
Dec-76
Dec-78
Dec-80
Dec-82
Dec-84
Dec-86
Dec-88
Dec-90
Dec-92
Dec-94
Dec-96
Dec-98
Dec-00
Dec-02
32. As in other Latin American economies, however, this was a costly policy approach, for at
least two broad reasons. First, the export surplus was inimical to medium- and longer-term
growth, since it amounted to external dissaving. Figure 10 summarizes the longer-term shifts in
some of Ecuador’s key macroeconomic flows. From the mid-1970s until the early 1980s
Ecuador was a net importer of goods and non-factor services, but after the debt crisis it became a
net exporter (except for 1987, when an earthquake interrupted oil exports, and in 1998, when the
exogenous shocks that set off the crisis reduced exports while anticipated exchange-rate
depreciation encouraged imports). Domestic saving remained essentially unchanged as a
percentage of GDP after the debt crisis, so that total saving and capital formation declined,
holding real-GDP growth to a lower level. The correlation of capital formation and net imports
can be seen in Figure 10. 7
7
Over the twenty-seven years 1976-2002, the correlation coefficient of gross fixed capital formation and
net imports of goods and non-factor services was 0.716, while the correlation coefficient of gross fixed
capital formation and real GDP growth was 0.349.
15
Figure 10. Ecuador: Real GDP growth, capital formation, and net imports of goods and non-factor services,
1976-2002
Per cent
30.0
25.0 Gross fix ed capital formation
20.0
15.0
10.0
<-- R eal GDP grow th
5.0
0.0
1976 1980 1984 1988 1992 1996 2000
-5.0
Net imports of goods and non-factor services

-10.0
-15.0
Net imports of goods and non-factor services (per cent of GDP)

Gross fixed capital formation (per cent of GDP)
Real GDP growth (annual percentage rate)
Data source: National accounts of Ecuador
33. Second, the exchange-rate depreciation generated direct and indirect inflationary pressure,
which continually undid the effects of the depreciation and made further depreciation necessary
to maintain the export surplus. As so often happened in economies that used repeated
devaluation to sustain a depreciated real-effective exchange rate, depreciation continually
induced inflationary pressure, and further exchange-rate depreciation became necessary. Annual
consumer-price inflation averaged 13.1 per cent between 1971 and 1981 but 41.6 per cent
between 1982 and 1994.
34. One of the many damaging consequences of this cycle of depreciation, inflation and further
inflation was that it induced spontaneous dollarization. At first this dollarization, ranging from
the purchase of dollar currency to overseas placements of wealth, was either illegal or semi-legal.
Over the 1980s and 1990s, however, Ecuador carried out a gradual process of financial
liberalization, and, as part of that process, it gradually authorized financial institutions to carry
out operations in dollars as well as sucres.
35. It is difficult to overstate the importance of Ecuador’s becoming an oil exporter in the 1970s
– not just the consequences of the oil exports per se, but the government’s decision to expand the
public sector and the private sector’s decision to begin taking on external debt. Even in the
1970s, the external-debt stock grew considerably faster than net imports of goods and non-factor
services: between 1977 and 1981 the total external debt rose by US$5.2 billion, while the
accumulated net imports of goods and non-factor services was just US$1.1 billion. Between
1977 and 2001, however, the gap was extraordinarily large: while the total external debt
increased by a total of US$12.1 billion (including the debt reductions secured in 1994 through a
Brady transaction and an August 2000 debt-reduction transaction), Ecuador ran an accumulated
surplus of net exports of goods and non-factor services amounting to US$8.4 billion, a difference
16
of US$20.6 billion, partly to cover interest, but in large measure to cover other kinds of financial
outflows, including those generally described as “capital flight.” No less important, the
oversized public sector remained a burden to the economy throughout the 1980s and 1990s.
Through such mechanisms as conventional direct and indirect taxes and the inflation tax,
Ecuadorian residents bore the burden of debt service, excess staff costs, energy subsidies,
inefficient publicly-owned enterprises, and an increasingly unviable pension system. The
pension problem derived in part to a highly complex and outmoded system for adjusting wages
in formal sectors. Ecuador’s political parties, badly fragmented along ideological, regional and
even ethnic lines after constitutional government resumed in 1979, were unable to agree on the
kind of comprehensive public-sector restructuring that almost everyone agreed was necessary.
As a consequence, productive enterprises found themselves not only with inadequate resources
for capital formation, but with diminished incentives to carry it out.
4. Causes and implications of Ecuador’s semi-dollarization
36. Over the 1980s and 1990s, as their currency’s purchasing power not only diminished but
became increasingly unstable, Ecuadorians sought a stable unit of account for wealth and
contractual relationships. In addition to this, two additional causes of dollarization were
(i) “globalization,” which meant that Ecuadorians engaged in an increasing number of dollar-
denominated transactions with non-residents; and (ii) the increasing “density” of contractual
relationships in the economy as it modernized, which deepened the need for a unit of account
with stable purchasing power. The growing number of people who rented out housing, placed
savings at interest, sold professional services, undertook commercial contractual relationships,
and so on came increasingly to insist on the dollar as the unit of account, to the extent they
possessed the “market power” to do so. People on purchasing sides of contractual relationships
presumably preferred the depreciating local currency as the unit of account. Depending on their
bargaining strength, however, to use local currency they had to pay premia high enough to
compensate not only for the expected depreciation of the local currency against the dollar, but
also for the associated uncertainty. Given a choice between high premia and dollar
denomination, purchasers often settled for dollar denomination – although they sometimes
learned ex post that they would have been better off in sucres even with the high premia.
37. It stands to reason that dollarization has advanced most in economies with long inflation
histories, such as Argentina, Bolivia, Perú, and Ecuador. 8 Experience suggests that government
attempts to reverse spontaneous dollarization do more harm than good. In 1982 Bolivia’s
government directed banks to convert dollar accounts to local currency. Subsequent withdrawals
and consequent destabilization helped precipitate Bolivia’s slide into hyperinflation. In 1986
Perú’s government, directed banks to convert dollar accounts to local currency, with similar
results: conversion of dollar accounts discouraged depositors and produced withdrawal demand
that had to be met largely through liquidity creation. In both cases, while other aspects of
economic policy contributed to hyperinflation, conversion of dollar accounts was a destabilizing
trigger. No doubt mindful of this experience, Ecuador’s authorities concluded it was best, or
8
One reason dollarization had not proceeded as fast and as far in Brazil as in other economies despite high
inflation was its use of financial indexation, which afforded wealth holders some protection from
inflationary uncertainty. Dollarization has advanced in Brazil since formal indexation ended in 1994.
17
least bad, to allow dollarization to follow private preferences.
38. The dollar-denominated proportion of Ecuador’s broad on-shore money supply (including
quasi-money) rose from 3.9 per cent at the end of 1990 to 34.6 per cent at the end of 1998 (see
Table 2). This measure understates the extent of dollarization, for at least two reasons. First,
dollar currency was increasingly used within Ecuador for transactions. Second, Ecuadorians
made increasing use of off-shore deposits, not included in official money-supply figures.
According to estimates prepared for July 1999, overall banking-system deposits totaled US$6.4b.
On-shore deposits totaled US$3.5b, while off-shore deposits totaled US$2.9b. Of the on-shore
deposits, US$1.7b was in sucres, and all off-shore deposits were in dollars.
39. Table 2 shows that dollarization progressed slowly at first, but became pervasive by the mid-
1990s, especially after exchange-rate depreciation mechanically increased the sucre equivalent of
dollar balances. (Again, these indicators cover only on-shore banks, and so under-report the
extent of dollarization.) In December 1996, 24 per cent of all on-shore bank demand, savings
and time deposits were in dollars rather than sucres; in December 1998, this percentage had risen
to 41, and in March 2000 it stood at 63 (see Figure 4). In December 1994, 33 per cent of all bank
loans were in dollars; in December 1998, this percentage had risen to 60, and in March 2000 it
reached 91 (see Figure 5). Over this period, the commercial banks’ overall deposit base declined
about thirty per cent in dollar terms, with sucre deposits falling more than two thirds while dollar
deposits grew.
40. Where an economy’s money supply is semi-dollarized, exchange-rate depreciation directly

increases the local-currency value of its dollar part. The inflationary impact of exchange-rate
depreciation is therefore larger and probably more rapid than it would otherwise be. Since
inflation following exchange-rate depreciation erodes some of the real-effective effect,
achievement of any particular external-accounts objective is likely to require more depreciation,
all other things being equal. Moreover, once the price level and exchange rate begin rising, they
affect expectations and uncertainty regarding future inflation and depreciation, and encourage
further spontaneous dollarization.
18
Table 2. Ecuador: Semi-dollarization

Year Year-end percentage in U.S. dollars of:
Quasi- Deposits Loan
money portfolio
1989 9.7 14.7 1.9
1990 7.4 13.3 1.5
1991 7.5 14.5 3.0
1992 10.8 20.0 6.8
1993 12.6 16.9 13.4
1994 15.7 15.6 20.3
1995 24.3 19.2 28.3
1996 28.0 22.3 32.6
1997 36.9 23.6 45.1
1998 43.9 36.9 60.4
1999 47.4 53.7 66.5
41. Evolving inflationary expectations and uncertainty and shifts in the money stocks
denominated in the two units of account are likely to affect local-currency and dollar interest
rates in complex ways. In a semi-dollarized system, with some interest rates in dollars and some
in the national currency unit, relative interest-rate movements can become difficult for market
participants and policy-makers alike to interpret. Even in an undollarized system, it is usually
impossible to know to what extent interest rates reflect inflationary expectations and uncertainty
rather than real scarcity of financial resources. In a semi-dollarized system, relative domestic-
currency and dollar interest rates respond to and influence relative supplies of and demand for
financial resources in the two units of account. For example, banks receiving a sudden influx of
dollar deposits might reduce dollar lending rates and raise local-currency lending rates, to
encourage dollar borrowing to meet asset-liability matching objectives. Policy-makers
conceivably could respond to rising local-currency interest rates by easing monetary policy. This
might produce expectations of higher inflation, however, and induce further surges in dollar
deposits, hence further reductions in dollar interest rates and further increases in local-currency
rates. Where the authorities attempt to support an exchange rate by deliberately propping up
local-currency interest rates, the circumstances can become especially difficult to interpret. The
role of interest rates is all the more complicated where nominal deposit rates are high, because
capitalization of interest into deposit stocks implies that local-currency deposit stocks tend to
grow at rates on the order of the interest rates.
42. In addition, the fact that the economy is operating in two units of account with an unstable
exchange rate implies that at least some entities have unmatched positions. One of the obvious
lessons from Ecuador’s pre-dollarization crisis, as discussed in Section 2 above, is that
exchange-rate depreciation poses significant risks not only to entities whose dollar obligations
are inadequately matched with dollar earnings, but also to other entities whose dollar assets
include the obligations of inadequately matched entities.
43. In the event, however, the most important problem the dual-currency system created for the
economy was that it made exchange-rate depreciation exceedingly dangerous. Once exchange-
rate depreciation became unavoidable, the dual-currency system effectively collapsed.
19
5. Ecuador’s approach to dollarization
44. For several days following the January 2000 announcement, it was unclear whether
“dollarization” would mean (i) a conventional fixed exchange rate; (ii) “loose” convertibility,
under which the Central Bank could carry out a broad range of transactions so long as it
maintained full foreign-exchange backing for all local-currency issues at a fixed exchange rate,
(iii) “strict” convertibility, under which the only allowable Central Bank transactions would be
purchase and sale of foreign exchange at a fixed exchange rate; (iv) complete substitution by the
dollar for the sucre, but with the Central Bank retaining some functions; or (v) complete
substitution of the sucre by the dollar, with the Central Bank closed and liquidated. After
internal discussion and advice from foreign experts, the Government decided on the fourth of
these alternatives. The domestic currency would be withdrawn, but the Central Bank would
operate as a limited liquidity regulator and as something like a “lender of last resort,” using
allocated or borrowed foreign-exchange resources. With time-deposit unfreezing just ahead, the
authorities felt that the Central Bank should retain as many of these functions as possible. 9
45. The basic dollarizing legislation approved early in March 2000 was called the “Fundamental
Economic Transformation Law.” 10 It directed the Central Bank to use part of its foreign-
exchange holdings to repurchase the outstanding sucre stock (at the sharply depreciated
exchange rate the Central Bank’s total foreign exchange stock was nearly double the sucre
currency stock); and also provided for adoption of the dollar as an official monetary unit. It also
provided that legal accounting records would henceforth be maintained in dollars, with public
contracts, tax assessments, and public accounts to be converted at the fixed rate. It directed the
Central Bank to issue coins in denominations smaller than one dollar (as in Panamá, on a
currency-board basis against foreign-exchange holdings). It specified that the Central Bank
would continue providing commercial-bank clearing-house services, for which it would continue
holding bank-reserve deposit accounts.
46. The Law reorganized the Central Bank into four “systems” with segregated balance sheets.
In the first, foreign exchange holdings back an equal quantity of old sucres, until their retirement,
and new small-value coins. In the second, foreign-exchange holdings stand behind an equal
stock of bank reserve deposits and Central Bank “stabilization bonds.” In the third, the Central
Bank’s remaining foreign-exchange holdings, Treasury bond holdings, and holdings of
repurchase agreements stand behind the Central Bank’s external obligations (including those to
the IMF), the public sector’s deposit accounts, and Central Bank stabilization (open-market)
bonds. The foreign exchange in this account would henceforth be considered “available” gross
international reserves, since amounts held in reserve against coins and against banks’ reserve
deposits would be unavailable for external transactions (under dollarization, of course, private
entities would not need to purchase foreign exchange from the Central Bank). Finally, the fourth
9
A semantic issue arises here: a true “liquidity regulator” and “lender of last resort” presumably must be
able to create money. The idea was that the Central Bank would be a limited entity, operating basically
through the size of its financial position, able to prevent minor problems from getting out of hand but
unable to create money.
10
Informally, the Law came to be called the “Trolleybus Law” – i.e., more encompassing than an
“omnibus” law and advancing at high speed.
20
balance sheet comprises the Central Bank’s remaining assets and liabilities and its accumulated
capital position.
47. That is, the Central Bank’s foreign exchange is assigned first to back coins and banks’
reserve deposits, then all other liabilities. Some of these other liabilities, particularly external
obligations (including those to the IMF), are long-term. The Treasury’s deposit balance at the
Central Bank is a sight account, however. Net withdrawals from this account imply direct
reductions in the country’s international-reserve position. Table 3 shows the four balance sheets
as of March 10, when the system commenced, and then at the ends of the subsequent semesters
through June 30, 2003. 11 The retirement of sucres and the introduction of coins can be seen in
the progress of the first balance sheet. The virtual stagnation of the second balance sheet is
indicative of the slow pace of recovery in banking-system operations over the course of the year
(see Section 6 below). The table also shows the increase over 2000 and 2001 in what is now
defined as the international-reserve position, deriving basically from the current-account surplus
and credit from multilateral lenders.
48. The dollarization law authorized the Central Bank to carry out “liquidity” operations with
banks, including transactions in bonds with commercial banks under its second system and
repurchase operations under its third system. The Central Bank’s available foreign-exchange
resources and its liabilities to external lenders and to the Treasury limit its scope to carry out
such operations, however. The Central Bank is able, in effect, to carry out some “recycling”
operations, accepting resources from banks with more liquidity than they require and on-lending
them to illiquid banks. Recycling of this kind may be useful because depositors have been
observed to “fly to quality” in moments of crisis, moving deposits from “weaker” to “stronger”
banks. In addition, to supplement the Central Bank’s capacity to cope with short-term liquidity
problems, a separate Liquidity Support Fund was set up outside the Central Bank, capitalized
through a US$70m loan from the Andean Development Corporation (CAF). Banks were
required to place deposits with the Fund amounting to one per cent of their deposit base, a
supplemental reserve requirement (on top of their conventional eight-per-cent reserve
requirement on all deposits).
11
The monetary base is defined as the sum of the monetary emission and bank reserve deposits. It is not
comparable with the monetary base before dollarization, since currency in circulation after dollarization
consists of dollar currency not issued by the Central Bank.
21
Table 3. Ecuador: The Central Bank’s four balance sheets, March 10 – June 30, 2003
US$million: 10-Mar-00 30-Jun-00 31-Dec-00 30-Jun-01 31-Dec-01 30-Jun-02 31-Dec-02 30-Jun-03
(1) Exchange system (net): 0 0 0 0 0 0 0 0

Assets: 425 153 35 35 27 34 40 45
Reserves backing monetary emission 425 153 35 35 27 34 40 45
Liabilities: -425 -153 -35 -35 -27 -34 -40 -45
Monetary emission * -425 -153 -35 -35 -27 -34 -40 -45
Sucres -425 -153 -12 -10 0 0 0 0
Small-denomination coins 0 0 -24 -26 -27 -34 -40 -45
(2) Financial reserve system (net): 0 0 0 0 0 0 0 0
Assets: 299 215 226 144 261 228 264 279
Reserves backing bank reserves 299 215 226 144 261 228 264 279
Liabilities: -299 -215 -226 -144 -261 -228 -264 -279
Reserve deposits -287 -212 -226 -144 -261 -228 -264 -279
Monetary stabilization bonds -12 -3 0 0 0 0 0 0
(3) Main operations system (net): 0 0 0 0 0 0 0 0
Assets: 821 1203 1579 1651 1454 1522 1319 1384
Available reserves 141 523 919 1025 785 911 704 787
Repurchase operations 38 1 0 0 50 1 3 1
Treasury obligations 642 679 661 625 618 610 611 596
Liabilities: -821 -1203 -1579 -1651 -1454 -1522 -1319 -1384
Non-financial public-sector deposits -512 -661 -1065 -1144 -988 -1159 -1026 -1188
Private deposits -26 -9 -8 -14 -16 -24 -9 -13
Central-bank term obligations -283 -533 -500 -473 -396 -330 -275 -169
Central Bank security issues 0 0 -6 -19 -55 -7 -9 -14
(4) Other operations system (net): 0 0 0 0 0 0 0 0
Assets: ** 1627 1712 1613 1571 1653 1653 1653 1653
Liabilities and capital: ** -1627 -1712 -1613 -1571 -1653 -1653 -1653 -1653
Other liabilities: *** 497 270 180 438 403 392 489
Capital and reserves -2291 -1883 -1804 -2091 -2056 -2046 -2143
of which, accumulated profit 83 0 53 0 0 0 0
Memorandum:
International reserves 865 891 1180 1204 1074 1173 1008 1111
Monetary base 712 365 261 179 289 262 304 323
* Small coins and sucres still outstanding.

** Includes capital positions in international and Andean organizations and collateral for interest on Brady bonds.
*** Includes exchange adjustment and provisions.
49. The loss of central-bank seigniorage earnings is an issue often raised in connection with
dollarization. For present purposes, the seigniorage flow is best defined as the earnings flow a
monetary authority receives from assets held against its monetary issue less any interest paid on
that monetary issue. A monetary authority that retires its currency issue gives up whatever
interest it could earn on the assets it would have held against that issue. 12 Under its new
structure, Ecuador’s Central Bank gives up most, but not all, of its seigniorage earnings. To the
extent it receives interest on its own foreign-exchange holdings, it earns something very much
like a seigniorage flow, since it pays no interest on its small-change currency issues nor on its
holdings of bank reserves.
50. A transitory disposition in the dollarization Law provided for a one-time reduction in interest
rates on existing sucre and dollar-denominated financial assets. For financial contracts
outstanding on or after January 11, asset rates were reduced to a maximum of 16.82 per cent for
assets and 9.35 per cent for liabilities. Contracts already at lower rates would retain their
original rates. The Law also provided for conversion of sucre interest rates on loans outstanding
12
Some observers argue that the United States’ Federal Reserve should share some of its seigniorage
earnings with nations that dollarized. Their basic reasoning is that by dollarizing a nation increases overall
demand for dollars, thus enabling the Federal Reserve to increase its dollar issue and hence its receipt of
seigniorage. At this writing, however, it seems doubtful that the United States could be persuaded to
compensate dollarizing economies for the transfer of seigniorage revenues.
22
at the moment of dollarization (a “desagio” table). These measures were intended to address a
familiar problem in economies undergoing sudden disinflation. Prices and interest rates on
contracts formulated before the disinflation presumably reflected relatively high inflationary
expectations, and the fact that these assumptions now changed would have implied windfall
gains and losses.
51. The point can be made with a simple numerical example. Suppose a buyer agrees to pay a
seller purchasing power equivalent to 100 pesos at current prices one year from now, at a
moment when buyer and seller both anticipate a doubling of the price level over the year. In this
case, they would presumably agree on 200 pesos as the amount the buyer would pay the seller
one year from now. Suppose, however, that immediately after the contract is set the expected
inflation rate falls to zero. In this case, if the price were not revised back to 100 pesos, the buyer
would pay the seller twice the purchasing power originally intended. In practical circumstances,
of course, inflationary expectations are almost never so exactly and explicitly held, and contract
revision of this kind is likely to produce not only winners and losers, but differing and confused
perceptions about gains and losses. This clearly happened in Ecuador, but since contracts had
become overwhelmingly short term the problem was transitory.
52. In addition, in something of a reversal of financial liberalization, the Law subjected interest
rates to a usury ceiling, to be set periodically by the Central Bank as a “risk” spread over LIBOR,
up to 1.5 times the banking system’s average lending rate (the so-called “reference” rate
calculated by the Central Bank from data obtained from the banks). Although this was intended
to prevent banks from engaging in dangerous high-interest-rate practices, many observers argued
that existence of a usury ceiling, however high, would discourage would-be financial-system
entrants. (In general, forceful prudential regulation -- requiring banks to show that high-interest
operations are safe enough, and perhaps setting additional capital requirements if necessary -- is
a better approach to this problem than usury ceilings.) In July 2000, under some pressure from
the IMF, the Government issued a “clarification” to the effect that the usury limit would be
exactly (i.e., not up to) 1.5 times the banking system’s prevailing average lending rate, a level
presumed to be well above what the market would regard as binding.
53. Recognizing that the success of dollarization would depend on resolution of problems outside
the monetary system per se, the dollarization law approved in March and follow-up legislation
(the so-called “Second Trolleybus Law”) approved in September 2000 included measures
regarding financial-sector regulation, non-performing bank loans, and public-sector structural
reform. (To be sure, “success” refers here to whether economic performance improved, not to
whether the authorities simply succeeded in dollarizing.) For the banking system, the
dollarization legislation clarified the rules defining capital adequacy and set explicit rules
regarding the time commercial banks would have to remedy capital deficiencies. Accompanying
regulations further provided for gradual adaptation of international standards for commercial-
bank accounting, to be used to gauge compliance with Basel norms.
54. In addition, the dollarization legislation and subsequent regulations also provided for
reprogramming of private debt to financial institutions. Some 800,000 debtors who each owed
less than US$50,000 in all to the financial system were authorized to reprogram it up to seven
years, according to standard formulas applicable to broad categories of debt (credit cards,
23
mortgages, etc.). At the same time, a legal framework was set out to enable the several hundred
debtors owing more than US$50,000 to renegotiate their loans. The Government hoped this
approach would enable banks and businesses to put the crisis behind them. After the legislation
was approved, however, the Government came under political pressure to make debt
reprogramming compulsory and generalized, for larger as well as for smaller debtors, and to
reduce the interest charges. The Government resisted, hoping to limit reprogramming of large
loans that are still viable, to ensure that banks maintain adequate cash inflows. It took a
relatively long time – about three years -- to complete the full reprogramming task.
55. The public-sector structural reforms also included changes to expedite privatization of the
telecommunications and electricity sectors and to open the hydrocarbons sector to private
participation, as well as some “flexibilization” of formal labor-market restrictions. While a full
discussion of the very large reform agenda that still remains for Ecuador’s public sector and
financial system is beyond the scope of the present paper, it is relevant that Ecuador has lagged
behind other economies in the public-sector reform process. The telecommunications and
electricity sectors have been reorganized as saleable companies, but have not yet been privatized.
There are no plans to privatize the hydrocarbons sector, although recent reforms widened the
scope for private operations under contract. The public pension system is not financially viable,
but, unlike other Latin American economies it has not thus far been possible to reform it. As in
many economies, Ecuador badly needs to carry out a deep labor-market reform. Formal-sector
restrictions, including job-tenure provisions and a centralized system of wage determination,
continue to discourage job creation. In 2002 an important reform simplified the unusually
complex wage-adjustment system and introduced a limited regime for temporary hiring. Even
with these changes, however, the formal labor market remains unusually restrictive.
6. Ecuador’s transition to dollarization in 2000 and 2001
56. Over the course of 2000 and 2001, Ecuador’s policy-makers had to deal with several
transition and implementation problems. These included (a) the unfreezing of time deposits;
(b) the price-level increase brought about by the sucre’s severe undervaluation at the outset of
dollarization; (c) adjustment of motor-fuel, electricity, and cooking-gas prices, along with wages;
and (d) a wide range of practical difficulties for people unfamiliar with dollar currency. The
need to unfreeze deposits was a unique problem, but Ecuador’s experience with the other
transition circumstances affords lessons relevant for other economies.
57. Although the economy remained in deep crisis, macroeconomic conditions during 2000 were
relatively favorable for the transition. Rising world oil prices (see Figure 3) lifted export
earnings, compensating for the sluggish performance of other commodity exports. (Sector-
specific problems affected many commodity exports: disease afflicted shrimp exports, for
example, while banana exports had market-access problems.) Strong oil-export revenues
(accounting for nearly half of total government revenue) and steady enough tax performance
afforded breathing space for the fiscal accounts. Meanwhile, the sharply depreciated real-
effective exchange rate and sluggish economic performance -- real GDP growth was only 1.9 per
cent in 2000, and important manufacturing and construction sub-sectors continued to decline --
held imports low. The continuing current-account surplus helped the Central Bank accumulate
24
foreign exchange: even with the purchase of sucres from circulation, the Central Bank’s total
foreign-exchange holdings ended December 31 at about US$1.1b, an increase from US$865m on
March 10 (see Table 3).
58. Lending by multilateral institutions during 2000 also helped shore up the Central Bank’s
foreign-exchange base. In April 2000, after more than a year of discussions and negotiations,
Ecuador secured a twelve-month, US$304m IMF stand-by arrangement. (Although some
observers criticized the IMF’s “intransigence” during 1999, it is hard to see how IMF lending
could have affected the crisis dynamics.) The first of six scheduled tranches was released
immediately. The IDB disbursed tranches of several sectoral adjustment loans originally
approved in 1994 but delayed since then because Ecuador failed to satisfy various points of the
conditionality. In June 2000 the World Bank approved a US$150m structural-adjustment loan
with conditionality covering comprehensive tax reform, public-sector financial management,
financial-sector reform, and social-sector expenditure protection. The World Bank also
approved a US$10m Financial Sector Technical Assistance Loan. In June the IMF approved
disbursement of the second tranche of its stand-by program, and then in September approved
release of the third tranche. 13 Thereafter, however, the IMF became concerned about the slow
pace of structural-adjustment and tax reform, and held up release of the subsequent tranche until
June 2001.
59. When they first announced dollarization, the authorities’ largest concern was that banks
would have inadequate resources to cope with the time-deposit release due to commence in mid-
March 2000. Accordingly, they announced that no more than US$4,000 would be made
available in cash from each time-deposit account. The remainder would be provided in the form
of three- to seven-year marketable Treasury bonds, which banks would purchase from the
Treasury with their own promissory notes. The bonds would take deep discounts, but could be
used at par to pay taxes and to service debt to banks. In the weeks leading up to the start of the
unfreezing process, however, the state prosecutor raised legal objections to this use of Treasury
bonds. 14 The authorities then decided that for balances exceeding US$4,000 depositors would
receive longer-term certificates of deposit from the banks instead of Treasury bonds.
60. Release of time deposits commenced on March 13. To the relief of all concerned, from that
day and over the weeks following, the banks succeeded in limiting deposit withdrawals.
Deposits actually increased somewhat over 2000. Several factors contributed to this success.
One is that in early March 2000 the IMF, World Bank, IDB and CAF made a joint statement,
timed to buttress depositor confidence, that they would provide Ecuador financial support
totaling US$2.2b over 2000, 2001, and 2002. Another is that banks themselves carried out
13
Funds borrowed by the Central Bank from the IMF increase both the Central Bank’s foreign-exchange
reserves and its liabilities to the IMF. Funds borrowed by the Government from multilateral institutions are
deposited by the Government in its Central Bank account, and so also increase the Central Bank’s foreign-
exchange balances, but increase the Government’s – not the Central Bank’s – foreign-exchange liabilities.
14
Ecuador has a lengthy history of government assumption of insolvent private debt, including, of course,
the widely resented sucretización program (see Section 3). Many government officials were accordingly
troubled by the idea of having the government assume private banks’ deposit debt. While it is true that
bank obligations to the government would offset the government’s bond issue under the scheme,
Ecuadorians generally believed that at least some banks would prove unable to honor these obligations.
25
effective marketing campaigns. Dollarization may have persuaded many depositors that
circumstances were changing for the better, and that the end of the dual-currency system would
make banks safer. The rulings by the Constitutional Tribunal and the state prosecutor’s stand
against paying depositors in Treasury bonds may also have contributed to depositor confidence,
since they seemed to promise that that the authorities would henceforth find it difficult to carry
out arbitrary property seizures. The unpopularity of the freeze and the fact that the President
who had done it had been forced from office seemed to make it unlikely that anything similar
could ever happen. Finally, depositors may have reasoned that the withdrawal limit would help
ensure that banks had adequate liquidity. In the event, the Central Bank did not have to provide
net liquidity support to banks over the course of the unfreezing process. Banks able and willing
to do so to were permitted to make the full amounts of deposit balances available. Foreign-
owned banks and certain smaller banks did so, although not the large banks under AGD control.
61. Rising prices turned out to be a more serious transition problem. Consumer prices rose
91 per cent over 2000 (compared with 60.7 per cent over 1999). The main cause was the deeply
depreciated exchange rate at the moment of conversion (see Figure 1), which implied that prices
would have to rise to restore parity with world prices. Strictly speaking, this was a one-time
price-level increase, not a true “inflation,” but for most Ecuadorians this was an academic
distinction. The price-level rise was widely predicted when dollarization was announced.
Indeed, if one assumed that the real-effective exchange rate would revert to something its
average in (say) 1996-9, it was straightforward to estimate the price-level increase that would
bring about parity. If it took a year to reach parity, for example, the required price-level increase
would have presumably have been between 120 and 140 per cent. It seemed logical to suppose
that once parity was reached, prices would level off and then rise only at the world inflation rate.
62. This reasoning missed several important points, unfortunately. One was the possibility –
indeed the likelihood – that the price-level rise would overshoot, making the real-effective
exchange rate overvalued. Deflation would then have to occur, or the price level would have to
remain stable while world inflation took place, to induce real-effective depreciation. The
“Hume” mechanism would presumably operate -- dollars would be “exported” in exchange for
imported goods and services, the economy’s dollar stock would shrink, and this would drag
down the price level, producing real-effective depreciation. (Tradables prices in dollars would
presumably be set to external parity values, so it would be non-tradables prices that actually fell.)
“Downward price stickiness” implies that deflation could take a painfully long time to work,
however. As Argentina’s experience in the 1990s suggests, the real-effective exchange rate
simply cannot adjust so rapidly under dollarization and convertibility as with floating or less
firmly fixed exchange rates. In the event, Ecuador’s real-effective exchange rate appreciated
sharply (see Figure 9). The standards for gauging whether and by how much it may have
become overvalued are inevitably somewhat arbitrary, but there is reason to believe that by late
2002 or 2003 the country’s international competitiveness had diminished dangerously
(Sections 7 and 8 discuss this point further).
63. Meanwhile, during 2000 and 2001, even with the price level rising at an annual rate just short
of three digits, deposit rates in the (now fully dollarized) banking system remained relatively
steady, at several points over LIBOR. At the measured inflation rate, they were highly negative
in real terms. Although some observers found this phenomenon mystifying, the reason was clear
26
enough. With the exchange rate firmly fixed, the relevant opportunity cost for deposit funds was
external deposit rates. For most would-be depositors, purchasing and holding a commodity
bundle for subsequent sale was simply impractical. Moreover, with the economy still in
recession and the deposit freeze just ending, demand for credit was limited. If the banks had had
good lending opportunities, they would presumably have set higher deposit rates. 15 In any case,
the banking system remained in disarray, with banks accounting for more than half the system’s
asset base having failed and remaining under public control.
Figure 11. Ecuador: Evolution of domestic motor-fuel prices (deflated by consumer prices;
December 1999 = 100)
250
230
210
190
170
Gasoline "super"
1999 = 100
150
130
110
Gasoline "extra"
90
70
Diesel
50
Dec-94 Dec-95 Dec-96 Dec-97 Dec-98 Dec-99 Dec-00
64. A large problem associated with the price-level increase during 2000 and 2001 was the
adjustment of motor-fuel, cooking-gas, and electricity prices, as well as formal-sector wages.
These prices were set by the authorities. In September 1998, when they introduced a new cash-
subsidy program for poorer households, the authorities had raised but then fixed the sucre price
of household cooking gas. In July 1999 the government had frozen motor fuel prices for one
year to settle a transportation strike (see Figure 11). Over 1999 electricity rate adjustment had
lagged behind the surging price level. Taken together, the repressed prices implied forgone
government and public-enterprise revenue amounting to several percentage points of GDP.
Political and equity considerations meant, however, that any energy-price adjustments would
require simultaneous wage adjustment, which would, inter alia, affect public-sector staff
expenditure. Real wages had been declining for months in real terms. (Figure 12, which charts
15
Some economists believe that negative real interest rates indicate, ipso facto, that financial markets are,
or are likely to be, distorted. In circumstances of limited credit demand, however, interest rates can
perfectly well clear at non-positive real values. The fact that the price-level adjustment was a move to
international parity rather than a genuine inflation suggests that if there was a mystery about the array of
prices and interest rates, it was that the adjustment to parity required so long a period of time to take place.
27
the evolution of the real earnings of a private-sector worker earning the minimum wage and
mandated complementary pay, shows the decline in real-wage levels over the course of the
crisis.)
Figure 12. Ecuador: Total remuneration received by a typical private-sector worker earning the
minimum wage
10000000 140
120
Real
100
Sep 94 - Aug 95 = 100

1000000
Sucres (log)
80
Nominal
60
100000
40
20
10000 0
Jan-90
Jul-90
Jan-91
Jul-91
Jan-92
Jul-92
Jan-93
Jul-93
Jan-94
Jul-94
Jan-95
Jul-95
Jan-96
Jul-96
Jan-97
Jul-97
Jan-98
Jul-98
Jan-99
Jul-99
Jan-00
Jul-00
January 1990 - September 2000
65. For policy-makers, the energy-price adjustments ought to have involved delicate balances
among (i) allocative efficiency (and international competitiveness); (ii) the non-financial public-
sector finances; and (iii) social equity. Unfortunately, the troubled political context made it
almost impossible to apply these criteria in meaningful ways. For the IMF stand-by program
agreed in April 2000, the Government promised to raise motor-fuel prices and wages in two
steps, in July and October, and to increase the cooking-gas price by 40 per cent. The
Government then concluded, however, that it would be best to address the problem in one go,
rather than face two periods of heightened social tension. Accordingly, on May 25, it announced
60- to 80-per-cent motor-fuel price increases. It decided not to raise the politically sensitive
cooking-gas price, however, but instead raised prices of jet fuel and industrial oil derivatives by
about 300 per cent. It also authorized 48- to 70-per-cent public-sector pay increases, and raised
the monthly stipend it had been providing mothers of poorer families since September 1998 by
75 per cent. On May 31 it raised electricity-rates. If the price level had then stopped rising,
these prices would have been more or less “in line.” Unfortunately, however, the price rise
continued, causing energy prices and formal-sector wage rates to lag once again in relative terms.
In January 2001 the Government announced a new round of corrective increases, including a
100-per-cent increase in the cooking-gas price. Widespread popular demonstrations ensued.
The Government negotiated an agreement with demonstration leaders, under which, inter alia, it
reduced the increase in the cooking-gas price to 60 per cent, promised that motor-fuel prices
would remain unchanged for at least twelve months, and agreed to maintain a dialogue with
representatives of demonstrating groups. While the details of the authorities’ struggle to reset
public-sector energy prices during 2000 are perhaps no longer of great interest in themselves,
they do illustrate how hard it is to set public-sector prices on rational criteria in the midst of an
28
inflationary process.
66. Along with the financial-sector, inflation and price-realignment issues, some practical
problems arose in the transition to dollarization. Many people found it difficult to adjust to using
dollars (the fact that dollar bills were all the same color was a problem for illiterate people, for
example). The authorities carried out a public-information campaign, but it was poorly funded.
Shortages of small-denomination coins (see Associated Press 2000) were a continuing problem.
Counterfeiting has occurred, although its extent is (by its nature) difficult to determine. On the
whole, however, as of this writing four years after the dollarization announcement, the country
has become quite used to using dollar currency.
7. Ecuador’s macroeconomic performance since 2000
67. At the time of the transition to dollarization Ecuadorians’ opinions about the likely
consequences of dollarization were quite mixed. Some Ecuadorians believed it would lead to
faster growth, at least after the transition process was complete. Others, looking to Argentina’s
experience with a hard fixed exchange rate, believed that dollarization would either render the
economy uncompetitive or force severe deflation. One or the other of these views may yet turn
out correct, but the first four years of dollarization have turned out neither euphoric nor
disastrous (see Table 4). Although the 2000 and 2001 growth rates were relatively high as
Ecuador pulled out of the 1999 disaster, the 2002 and 2003 growth rates were more subdued. In
any case, recovered oil-export prices and the sluggishness of import growth have held the current
account in surplus. Higher oil-export proceeds, invigorated tax administration and tightened
expenditure management have held the public finances in overall surplus. At the same time,
however, the continuing positive differential between the Ecuadorian and world inflation rates
suggests that the economy’s competitiveness has been gradually eroding, with inevitable
consequences for medium- and longer-term growth prospects.
Table 4. Ecuador: Recent macroeconomic performance, 1997-2003

1997 1998 1999 2000 2001 2002 2003
Gross domestic product:
US$ million 19768.6 19722.7 13689.1 13927.1 19968.5 23267.7 25561.4
Real growth rate 3.4% 0.4% -7.3% 2.3% 5.1% 3.4% 2.9%
US$ per capita at prices and exchange rate of... 2000 $1,224.2 $1,205.9 $1,096.7 $1,101.0 $1,135.2 $1,152.0 $1,162.9
Inflation (growth rates):
Consumer prices (Dec.-Dec.) 30.7% 43.4% 60.7% 91.0% 22.4% 9.4% 7.1%
Implicit GDP deflator 25.9% 34.3% 56.9% 119.3% 36.7% 12.6% 6.8%
Exchange rate (25,000 sucres/US$):
Year average 0.160 0.216 0.453 0.998 1.000 1.000 1.000
December average 0.176 0.264 0.728 1.000 1.000 1.000 1.000
Real-effective exchange rate -- base 100 =: 1995 99.4 101.2 137.1 140.9 105.1 94.9 90.3
Per cent of GDP:
Gross fixed capital formation 21.4% 26.5% 15.9% 21.3% 18.5% 18.8% 19.0%
Current-account surplus (deficit): -2.3% -10.6% 6.7% 6.6% 3.2% 3.6% 3.1%
Exports of merchandise and non-factor services 30.6% 25.4% 39.0% 43.0% 29.5% 26.5% 25.6%
Imports of merchandise and non-factor services 30.8% 34.0% 30.7% 36.0% 26.6% 23.8% 23.5%
Non-financial public-sector surplus 1.5% 0.4% 0.6% 0.2%
Public and publicly guaranteed external debt 65.1% 66.4% 99.0% 81.6% 58.4% 51.1% 47.7%
Source: International Monetary Fund, World Bank
68. Figure 13 shows the recent evolution of the Central Bank’s monthly economic activity index
(not adjusted for seasonality). The chart shows the recovery of economic activity from the 1999
29
collapse during 2000 and 2001, but also shows the loss of growth momentum in 2002 and 2003.
Unemployment subsided from its 1999 levels, although inadequate employment remains (or
should remain) the core problem for macroeconomic management.
Figure 13. Ecuador: Monthly real economic activity, January 1992 – November 2003
250
200
150
100
50
0
Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03
Figure 14. Ecuador: Open unemployment in the three largest cities, March 1998 – December 2003
25%
Per cent of economically active labor force
20%
Guayaquil
15%
Quito
10%
Cuenca
5%
0%
Mar-98
Mar-99
Mar-00
Mar-01
Mar-02
Mar-03
Jun-98
Sep-98
Dec-98
Jun-99
Sep-99
Dec-99
Jun-00
Sep-00
Dec-00
Jun-01
Sep-01
Dec-01
Jun-02
Sep-02
Dec-02
Jun-03
Sep-03
Dec-03
69. Figure 15 and Figure 16 show Ecuador’s recent quarterly current-account and fiscal
performance. (The current-account chart shows inflows as positive values and outflows as
negative values, with the deficit (or surplus) and the balance flow. The fiscal chart shows
30
revenue flows as positive values and expenditure flows as negative values, with the deficit (or
surplus) as the balancing flow. In both diagrams, for each quarter, the total flows above and
below the horizontal axis are identically equal.) Notable points in the current-account chart
include the resurgence of oil earnings after 1998 and their persistence thereafter (apart from a
slight dip around the end of 2001 and the beginning of 2002) and the recovery of imports after
the dollarization year 2000. The growth of imports came about through a combination of the
real-effective exchange-rate appreciation, sluggish real-GDP growth, and the resumption of
normal banking activities. Notable points in the fiscal-performance chart include the strong
revenue performance after dollarization, including non-oil as well as oil revenue and growing
non-interest expenditure. Non-oil revenue performance was enhanced by improved
administration: the internal revenue service had been completely replaced in mid-1998, and, after
dollarization, vigorous leadership helped enhance the reconstituted agency’s effectiveness
(indeed, prestige).
Figure 15. Ecuador: Quarterly current-account performance, 1998 Q1 – 2003 Q3
US$ million
2500
2000
1500
1000
500
0
1998.1 1998.3 1999.1 1999.3 2000.1 2000.3 2001.1 2001.3 2002.1 2002.3 2003.1 2003.3
-500
-1000
-1500
-2000
-2500
Oil exports Merchandise imports Other merchandise exports

Interest due Other current-account items Current-account deficit

31
Figure 16. Ecuador: Quarterly fiscal performance, 1998 Q2 – 2003 Q2
US$ million
2000
1500
1000
500
0
1998.2 1998.4 1999.2 1999.4 2000.2 2000.4 2001.2 2001.4 2002.2 2002.4 2003.2
-500
-1000
-1500
-2000
Oil-export revenue Revenue other than oil-export revenue

Non-interest expenditure Interest expenditure
Deficit (- = surplus)
70. Figure 17 shows the evolution of Ecuador’s inflation rate since December 1997. As noted in
Section 6, Ecuador’s price-level rise toward parity took place over at least two years, and the
inflation rate appears to have settled to world rates only during 2003.
Figure 17. Ecuador, United States: Monthly inflation (percentage change over previous month,
December 1997 – August 2003
16.0
<-- Dollarization announcement

14.0
<-- Ecuador
12.0
10.0
8.0
6.0
4.0
2.0
<-- U.S.
0.0
Dec-97 Dec-98 Dec-99 Dec-00 Dec-01 Dec-02
-2.0
71. Figure 18 shows the recent evolution of Ecuador’s “reference” interest rates, weighted
32
averages calculated by the Central Bank on the basis of surveys of commercial banks. Liability
interest rates have remained well above international rates. Although they declined over 2000
and 2001 along with international rates, over 2002 and 2003 they remained at about 5 per cent,
even as LIBOR declined from about 2 to about 1.2 per cent. Meanwhile, although commercial-
bank asset rates declined from an average of 15.4 per cent in 2000 and 2001 to 13.3 per cent in
2002 and 2003, the asset-liability spread was actually higher in the later period – 8.1 per cent
compared with 7.9 per cent – and considerably more variable from month to month. The liability
interest rate is probably now on the order of the inflation rate, an indication of the sluggishness
of economic growth over 2002 and 2003.
Figure 18. Ecuador: Banking-system “reference” interest rates in U.S. dollars,

December 1999 - December 2003
18.0%
16.0%
14.0%
12.0%
10.0%
8.0%
6.0%
4.0%
2.0%
0.0%
Dec-99 Jun-00 Dec-00 Jun-01 Dec-01 Jun-02 Dec-02 Jun-03 Dec-03
Six-mo.US$ LIBOR (annual pctg. rate) Liability rate less LIBOR

Asset rate less liability rate
72. During 2002 and 2003 the government carried out several significant structural reforms.
During 2002 it secured legislative approval of a “Fiscal Responsibility Law,” which introduced a
system to set limits on expenditure increases and required that oil proceeds deriving from prices
exceeding historical averages be directed to debt repayment rather than expenditure. In 2003 it
secured legislation reforming the wage-determination system. In addition, the government
secured legislation requiring that the Central Bank monitor and periodically publish information
about sectoral competitiveness. This is an innovative approach, and it will be interesting to see
whether it significantly influences policy-making and company management in coming years.
73. To summarize, while dollarization enabled Ecuador to stop a hyperinflation, macroeconomic

performance since then has been mixed. There are grounds for concern that the economy is
danger of stagnating. The inflation rate has declined sharply, if rather more slowly than
originally anticipated, and economic activity has recovered to pre-crisis levels. Competitiveness
does appear to be a deepening problem, however. Recent structural reforms can be expected to
improve both competitiveness and business confidence, but it is doubtful whether these will be
33
sufficient to enable Ecuador to compete through productivity with emerging economies

elsewhere that appear to have cost advantages. The fact that Ecuador now has a stable currency
unit has made it easier for private firms to gauge their own competitiveness, and the fact that
exchange-rate depreciation is not an option has meant that they have been forced to carry out
structural reforms of their own. It remains uncertain, however, just how solid a basis has been
set for sustained economic growth in coming years.
8. Ecuador’s prospects under dollarization: the tradeoff of stability and

competitiveness
74. Since the early 1980s, the exchange rate has been the focus, if not the obsession, of
Ecuadorian macroeconomic policy-making. During the 1980s this economic variable very
nearly became the authorities’ basic instrument for addressing the full range of the country’s
macroeconomic problems, including price stability, external-debt management, competitiveness,
and even fiscal management. Predictably enough, the authorities found it impossible to deal with
all these objectives using a single instrument. The exchange rate was a tempting policy
instrument precisely because it is easy to activate and is generally, if crudely, effective, at least in
the short term. It is a problematic instrument, however, first because it is almost impossible to
set it correctly for any given moment, let alone over extended time periods; and, moreover,
because – as Ecuador’s experience demonstrates all too clearly -- overuse blunts its
effectiveness.
75. To understand why using the exchange rate to address multiple macroeconomic objectives is
so problematic, imagine a simple economic system with significant export and import flows, but
in which (a) the private and public external-debt level is insignificant; (b) the exchange rate has
little direct effect on government revenue and expenditure flows; (c) the banking system operates
entirely in national-currency units; and (d) the international-reserve stock is not an immediate
problem. Even in an economy of this kind, exchange-rate setting is likely to be something of a
dilemma. A more depreciated exchange rate would favor exports and shorter-term export-led
growth, but might generate inflationary pressure; a more appreciated exchange rate would favor
imports, in particular capital-goods imports, and perhaps longer-term growth. This tradeoff is
accompanied by a similar tradeoff involving the monetary system: the increased foreign-
exchange inflow resulting from a more depreciated exchange rate would lead to monetary
expansion, which favors credit expansion and growth in the short term, but also price-level
instability; a more appreciated exchange rate would favor price-level stability, and to this extent
favor longer-term growth. Summarizing, exchange-rate depreciation tends to favor short-term
growth at the cost of inflation, while appreciation tends to favor longer-term growth.
76. If the private sector has substantial external debt, however, these exchange-rate management
tradeoffs become more complex. While exchange-rate depreciation increases the domestic-
currency value of external debt, damaging the financial viability of private firms, it sets an
incentive favoring the export surplus needed to limit further external-debt growth. In itself, this
is favorable for capital formation, and to this extent favorable for longer-term growth.
Moreover, where the government finances depend heavily on the exchange rate, the exchange-
rate dilemma takes on yet another dimension of complexity. To the extent the government itself
34
has a heavy external-debt burden, exchange-rate depreciation widens the public deficit – and so,
inter alia, attempts to limit the growth of the external debt by using exchange-rate depreciation
may run the risk of increasing the government’s borrowing need.
77. A clear lesson from Ecuador\s experience is that repeated exchange-rate depreciation
encourages dollarization. Once a financial system begins to operate in a mix of dollars and
national currency, the economic system in general and the financial system in particular become
far more difficult to manage. With a semi-dollarized financial system, exchange-rate
depreciation directly increases the money supply and increases the national-currency equivalent
of debt to the banking system, which is likely to prove damaging to business enterprises that lack
dollar income. Indeed, once an economic system becomes semi-dollarized, it comes to operate
through economic mechanisms with highly intricate knock-on and feedback effects, and may
become virtually impossible to control. Dollarization became necessary in Ecuador for
essentially this reason. Indeed, once spontaneous dollarization advanced to the point it did, the
exchange rate became virtually ineffective. Since hardly anyone was prepared to hold sucres, the
exchange rate no longer had a base on which to operate.
78. Ecuador’s international competitiveness has clearly diminished substantially under

dollarization, and many Ecuadorians have drawn the conclusion that this can only be corrected
by reintroducing a national currency. It is probably a mistake, however, to believe that the
nation’s currency arrangement is a fundamental aspect of Ecuador’s deepening
uncompetitiveness. The deeper problems include (i) the reduced accumulation of capital since
the onset of the debt crisis, which has held back labor productivity; (ii) the debt-service burden,
which must be borne through taxes that figure directly in enterprises’ cost structures; (iii) the
burden of the oversized public sector, which – again -- must also be borne through taxes that
figure in enterprises’ cost structures; and (iv) the high costs of the financial system, which – yet
again -- figure in enterprises’ cost structures.
79. An additional dimension of the competitiveness problem is that oil exports tend to reduce the
competitiveness of non-oil activities. That is, Ecuador, like other oil exporters, tends to have a
“Dutch-disease” problem. Different analysts seem to mean different things by this expression,
but the meaning intended here is that oil exports tend to lead to real-effective exchange rate
appreciation, rendering non-oil activities less profitable and competitive than they would
otherwise be. Thus, the larger the flow of oil exports, and the larger the resulting exchange-rate
appreciation, the higher the incentive to import foodstuffs rather than purchase them
domestically. It is the essence of a Dutch-disease problem that relatively few people benefit
themselves from the export activity, even as their own activities are made less profitable.
80. It is therefore more than doubtful that Ecuador could significantly improve its
competitiveness by reintroducing a new national currency. One need only note that while
Ecuador had a national currency, it was unable to sustain adequate competitiveness. The greater
likelihood is that reintroduction of a national currency would restore the unstable dual-currency
system. It is difficult to imagine how, after the experience of the late 1990s, Ecuadorian
residents could be persuaded to hold a new currency, particularly if it were introduced with the
announced aim of helping to restore and maintain competitiveness. Ecuador would be best
advised to retain full dollarization, and pursue the goal of achieving and sustaining greater
35
competitiveness by improving public- and private-sector efficiency.
81. Ecuador’s own experience in the 1980s and 1990s shows clearly enough that while
exchange-rate depreciation can compensate temporarily for the costs of inefficient productive
structures, high taxation levels, and thick banking spreads, sooner or later those costs will rise
and restore the underlying uncompetitiveness. Only sustained structural reform to deepen
productive efficiency, to cut the public-expenditure commitments -- including debt service -- that
force high taxation, and to squeeze financial-system spreads can genuinely improve
competitiveness. Indeed, dollarization should prove helpful for structural reform, insofar as a
more stable unit of account enables policy-makers and private company managements to gauge
costs more accurately. It is unrealistic to assume, however, that a depreciable national currency
unit can permanently improve competitiveness.
9. Conclusions
82. Ecuador’s experience offers a cornucopia of lessons, not only about spontaneous and official
dollarization but also about the macroeconomic processes that, in retrospect, made official
dollarization inevitable. These lessons are likely to be especially relevant for economies that
already have significant spontaneous dollarization and run a danger of being forced to dollarize,
as well as for economies that choose or are forced to dollarize.
83. Ecuador was forced to dollarize because the financial system, and therefore the economy,
had become excessively vulnerable to exchange-rate depreciation. What made the economy so
vulnerable was that Ecuadorian residents responded to repeated exchange-rate depreciation by
shifting increasing proportions of their wealth into dollars. The economy accordingly came to
function in two units of account with a flexible, uncertain exchange rate. As a consequence,
some entities were long and some entities were short in dollars. While the banking system per se
was basically matched, it was exposed to a large number of entities that were exposed to
depreciation. In 1997 and 1998 Ecuador had the bad luck to undergo just the kind of external
shock and depreciation to which such borrowers were vulnerable. These led, through a sequence
of events that seems retrospectively inexorable, to banking-system collapse, incipient
hyperinflation, and a forced move to full dollarization.
84. Over the decades that preceded the pre-dollarization crisis, and indeed in the crisis itself,
Ecuadorian governments took many policy choices that had unfortunate consequences. The
military government of the mid-1970s made a fateful error when it assumed that oil-export prices
would remain sufficiently high to sustain the expanded public sector it set in place. To be sure,
Ecuador would have grown faster in any case if it had allocated relatively more of its oil earnings
to productive investment rather than to public-sector expansion. In retrospect, private borrowers
and their creditors should not have acted on the assumption that the exchange rate would remain
stable, or that interest rates on floating-rate debt would never surge. These were widespread and
all too common assumptions in the 1970s, however, and they set the stage for the debt crisis of
the early 1980s.
85. Again in retrospect, Ecuador needed deeper debt relief than it received after the debt crisis
36
began. In particular, while it may seem in retrospect that the government’s assumption of the
private debt was a costly error, the alternative – unrelieved private-sector distress – would also
have had devastating consequences. In any case, the government acted on expert advice, and
according to what were then considered best debt-management practices. Taking a retrospective
view once again, financial liberalization ought to have taken place gradually enough to enable
banking supervision to acquire sufficient competence to limit the banking system’s vulnerability.
Again, however, as noted in Section 2 above, even if they had been as skilled and forceful as
they really needed to be, it is doubtful that banking inspectors could have prevented banks from
becoming dangerously exposed to exchange-rate depreciation. It would have been very difficult
for banking supervisors to challenge the government’s assumption that the pre-announced
crawling-peg policy made the exchange rate stable enough. A common theme in all these
examples is that the widely held assumptions at given times can blind policy-makers and
business leaders to the need to plan for seemingly unthinkable macroeconomic contingencies. 16
86. In explaining experiences like Ecuador’s, economists conventionally focus on the

presumable inadequacy of structural reform or on the presumable failure of fiscal and monetary
policy to tighten sufficiently. These arguments often have a tautological character, in that the
only measure of the policy inadequacy is often the economy’s failure to adjust. In any case, it is
often pointed out in response that political exigencies make it impossible for governments to
carry out necessary reforms or to tighten policy sufficiently. To a large degree, debates about
macroeconomic adjustment are variations on these themes.
87. Thus, for example, it is generally accepted that Ecuador went into the debt crisis with an
oversized public sector. If its government had undertaken a sufficiently ambitious public-sector
reform in the mid-1980s, for example, it would have set a better basis for its international
competitiveness. The tax burden on the private sector could have been lower, the banking
system would have been able to provide more resources for productive purposes, aggregate-
demand pressure from public expenditure would have been smaller, and so on. Less exchange-
rate depreciation would then have been needed to generate the resources for debt service. The
spontaneous dollarization that took place over the 1980s and 1990s might not have been so
extensive, and the banking system might not have been as vulnerable as it turned out to be. By
the same token, however, it is only fair to ask what might have happened if Ecuador’s
government had refused to assume the private external debt, or if it had resisted accepting the
increased interest charges on the private-sector debt that resulted from the floating rates. It had
little choice, of course, given the international political realities of the time. But it is equally true
that, once the debt crisis thrust the economy into crisis, a tax reform or a public-sector
restructuring sufficient to make a difference was also unfeasible, given Ecuador’s internal
political realities, and in particular the reality that many Ecuadorians resented the idea of bearing
higher taxes or lower government expenditure in order to pay what had been private debt service.
88. Moreover, what is less frequently said of macroeconomic catastrophes is that it is sometimes
true that no conceivable policy could have prevented them. Ecuador’s pre-dollarization crisis
was probably a case in point. With Ecuador’s banking system so heavily exposed to exchange-
16
It is noteworthy that even in the mid-1970s, as they struggled to recycle the OPEC surplus, international
banks had the foresight to lend at floating rates, to this extent protecting themselves against what then
seemed unlikely world interest-rate surges.
37
rate depreciation, once the large exchange-rate depreciation of 1998 took place, the crisis was
probably unstoppable. It is now difficult to deny that generalized deposit insurance and then the
deposit freeze were policy errors; indeed they were widely recognized as such at the time. Even
so, as long as the dual-currency system remained in place, the authorities simply did not have it
within their power to prevent the banking crisis. To be sure, as a matter of political reality, had
they attempted nothing, they would probably have been intensely criticized for that.
89. Ecuador’s experience raises inevitable questions about the dangers of “dual-currency”
financial systems, particularly since such systems have evolved in several developing economies.
A dual-currency financial system increases the importance of maintaining exchange-rate
stability. Even so, dual-currency financial systems tend to become so because of a history of
exchange-rate instability, which implies that future exchange-rate instability can never be
entirely ruled out. Indeed, one important concern is that potential damage to dual-currency
financial systems could persuade an economy’s authorities to delay an exchange-rate
depreciation needed for balance-of-payments purposes.
90. One lesson from Ecuador’s experience would seem to be that bank supervision must be
allowed to limit the vulnerability of bank portfolios to exchange-rate depreciation. This
imperative may conflict, however, with the imperatives to ensure (a) that financial institutions
maintain their assets and liabilities matched in currency terms and (b) that financial institutions
maintain adequate profitability. To see why, suppose the dollar composition of liabilities
increases in a given economy. Banks may not be able to find borrowers genuinely able to
assume dollar liabilities in dollars, however, and might be forced to take on lower-yielding dollar
assets in order to maintain adequate matching.
91. Ecuador’s experience in 2000 and 2001affords important lessons about how best to undertake
a process of dollarization and what is likely to happen as the process evolves. One obvious point
is that domestic prices must rise or fall to parity following the introduction of dollarization, and
this process can take some time. Depending on the initial distance from parity, the adjustment
process can be painful and complicated, and there may be undesirable consequences for the
economy’s international competitiveness. While Ecuador’s deposit-unfreezing process is
unlikely to be a problem for other dollarizing economies -- strictly speaking, it had nothing to do
with dollarization per se -- it was successful despite many observers’ expectations, and it is
probable that dollarization contributed decisively to making it so. Another positive, perhaps
counterintuitive, lesson from Ecuador’s experience is that a central bank can have a continuing
role even with dollarization – although, to be sure, a government that wanted to emphasize its
commitment to dollarization might prefer to make a point of closing the central bank, to make
reintroduction of a national currency that much more difficult.
92. Finally, Ecuador’s present experience is likely to provide important lessons about how an
economy with a hard fixed exchange rate can deal with a deepening competitiveness problem.
Moreover, if the proponents of a new national currency get their way, Ecuador may yet provide
lessons on how to go about doing so and then on its consequences. All the same, however,
Ecuador would do best, or least badly, to stay with dollarization, and indeed to take advantage of
having a hard currency to deepen public-, private- and financial-sector efficiency.
38
REFERENCES
Acosta, Carlos Albert (2000). “La trampa de la dolarización.” (Contrapunto: Boletín

Mensual de la Red de la Sociedad Civil; Quito, Ecuador) (May) pp. 7-10.
Arteta, Gustavo (1999). “Opciones cambiarias para el Ecuador: El dilema de la

elección” (CORDES: Quito, May)
Associated Press (2000). “Ecuador Struggles with U.S. Dollars” (August 14).
Banco Central del Ecuador (2000). “Dolarización” (http://www.bce.fin.ec/avisos).
Banco Central del Ecuador (2000). Boletín Mensual,” various issues.
Baliño, Tomás J. T., Charles Enoch, et al (1997). “Currency Board Arrangements: Issues
and Experiences.” (Internatiional Monetary Fund Occasional Paper 151), August.
Bayas, Santiago and Eduardo Somensatto (1994). “Programa de sucretización

ecuatoriano: historia de los efectos monetários de la conversión de la deuda
externa del sector privado.” Cuestiones Económicas (Banco Central del Ecuador:
Quito, Ecuador) No. 23 (December).
Beckerman, Paul (1986). "Inflation and Dollar Accounts in Peru's Banking System,
1978-1984," World Development, August, Vol. 15, No. 8, pp. 1087-1106.
Beckerman, Paul and Andrés Solimano, eds. (2001). Ecuador: Crisis and Dollarization.
(Washington, DC: World Bank).
Berg, Andrew and Eduardo Borensztein (2000). “Full Dollarization: The Pros and Cons,”
International Monetary Fund, Economic Issues No. 24 (December).
Berg, Andrew, Eduardo Borensztein and Paulo Mauro (2003). “Monetary Regime
Options for Latin America,” in Finance and Development (International Monetary
Fund), September, pp. 24-27.
Berg, Andrew, Eduardo Borensztein and Paulo Mauro (2003). “Monetary Regime
Options for Latin America,” in Finance and Development (International Monetary
Fund), September, pp. 24-27.
Calvo, Guillermo and Carlos Végh (1992). “Currency Substitution in Developing

Countries: An Introduction.” Revista de Análisis Económico, Vol. 7 No. 1,
pp. 48-57.
Calvo, Guillermo (1999). “On Dollarization.” (draft) (April)
continues
39
REFERENCES (continued)
Carrasco V., Carlos Marx (1998). Ecuador y el Consenso de Washington: La Hora

Neoliberal (Cuenca, Ecuador).
Clements, Benedict and G. Schwartz (1992). “Currency Substitution: the Recent

Experience of Bolivia (IMF Working Paper: Washington, DC).
El Comercio, Quito daily newspaper (various issues)
Enoch, Charles and Anne-Marie Gulde (1998). “Are Currency Boards a Cure for all
Monetary Problems?,” Finance and Development, Vol. 35 (December) pp. 30-43.
Fernando-Arias, Eduardo and Ernesto Talvi (1999). “Devaluation or Deflation?

Adjustment Under Liability Dollarization” (October).
Fischer, Stanley (2001). “Exchange Rate Regimes: Is the Bipolar View Correct?”
Distinguished Lecture on Economics in Government, delivered at the Meetings of
the American Economic Association, New Orleans, U.S.A., January 6.
Ghosh, Avitash, Anne-Marie Gulde, and Holger C. Wolf (2002). Exchange Rate
Regimes: Choices and Consequences (Cambridge, Massachusetts: MIT Press,
2002).
Gomis Porqueras, Carlos Serrano, and Alejandro Somuano (2000). “Currency

Substitution in Latin America: Lessons from the 1990s” (World Bank Policy
Research Working Paper 2340: Washington, DC), May.
Government of Ecuador (2000). Letter of Intent for the International Monetary Fund
Standby Program (http://www.bce.fin.ec/avisos/intent_fmi.html) (April)
Gulde, Anne-Marie (1999). “The Role of the Currency Board in Bulgaria’s Stabilization”
(IMF Policy Discussion Paper: Washington, DC), April.
Jaramillo, Fidel (1994). “Ecuador: Estabilización, ingreso de capitales externos y

conflictos de política macroeconomica.” (Multiplica working paper: Quito,
Ecuador (April).
Kopits, George, E. Haindl, E. Ley, and J. Toro (1999). “Ecuador: Modernización del
Sistema Tributario. (Washington, DC: International Monetary Fund) (November)
Mancheno, Diego, J. Oleas, and P. Samaniego (1999). “Aspectos teóricos y prácticos de

la adopción de un sistema de convertibilidad en el Ecuador.” Notas Técnicas,
Banco Central del Ecuador (Quito, Ecuador).
continues
40
REFERENCES (continued)
Moreno, Juán Luís Villalaz (1999). “Lessons from the Monetary Experience of Panamá:
A Dollar Economy with Financial Integration.” Cato Journal, Vol. 18 No. 3
(winter).
Pareja, Mauricio (2003). “Canje, conversión y reducción de deuda pública: el caso

ecuatoriano,” Apuntes de Economía No. 28, (Banco Central del Ecuador,
Dirección General de Estudios, Quito, Ecuador) (June).
Rogoff, Kenneth S., Aasim M. Husain, Ashoka Mody, Robin Brooks, Nienke
Oomes (2003), “Evolution and Performance of Exchange Rate Regimes,”
International Monetary Fund Working Paper WP/03/243, December
Rojas, Mónica (1998). ”Dolarización en Ecuador: Economía ‘verde’… pero no por lo

ecológico.” Gestión (Quito, Ecuador) December, pp. 25-27.
Savastano, M. (1996) “Dollarization in Latin America: Recent Evidence and some Policy
Issues.” International Monetary Fund Working Paper (January).
El Universo, Guayaquil daily newspaper (various issues)

Beckerman P20041011 Ecuador Dollarization

Uploaded by

Copyright:

Available Formats

Beckerman P20041011 Ecuador Dollarization

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Beckerman P20041011 Ecuador Dollarization

Uploaded by

Copyright:

Available Formats

EXTERNAL DEBT, OIL DEPENDENCE AND A NATION’S CURRENCY:

WHY AND HOW ECUADOR DOLLARIZED

Paul Beckerman* October 11, 2004

Data are available from the writer on request ([email protected]).

Table 1. Selected Latin American economies: Public and publicly 3

U.S. dollars at 2000 prices and exchange rate

Sucre depreciation against the U.S. $200

Percentage growth rate of consumer prices (Dec.-Dec.)

Data source: International Monetary Fund

2. Ecuador’s pre-dollarization crisis, 1998-1999

Figure 2. Ecuador: Indicators of macroeconomic balance, 1998-2002 (per cent of GDP)

Balance-of-payments current-account surplus

Data source: Central Bank of Ecuador

Argentina 22.9 25.9 29.7 30.5 26.5 73.0

Source: World Bank

Figure 3. Ecuador: Monthly average crude oil-export price (US$/bbl.)

Dollarization announcement --->

Data source: Central Bank of Ecuador

non-performing portfolios (see Figure 5). 4

Figure 4. Ecuador: On-shore commercial-bank deposits, December 1996 – February 2000

Data source: Central Bank of Ecuador

Figure 5. Ecuador: On-shore commercial-bank credit performing normally and in arrears,

U.S.-dollar credit performing normally U.S.-dollar credit in arrears

Figure 6. Ecuador: Non-financial public deficit, 1991-2002 (per cent of GDP)

-6.0 Non-financial public-sector primary surplus

Data source: Central Bank of Ecuador

3. Oil and external debt: historical origins of Ecuador’s pre-dollarization crisis

Data source: International Monetary Fund, World Bank

Data source: World Bank

“liberalizing” government attempted to implement a free exchange-rate float. It retreated to a

(real) 1990 = 100 (nominal) February 2000 = 100

25.0 Gross fix ed capital formation

Net imports of goods and non-factor services

Net imports of goods and non-factor services (per cent of GDP)

Data source: National accounts of Ecuador

4. Causes and implications of Ecuador’s semi-dollarization

least bad, to allow dollarization to follow private preferences.

40. Where an economy’s money supply is semi-dollarized, exchange-rate depreciation directly

Table 2. Ecuador: Semi-dollarization

5. Ecuador’s approach to dollarization

US$million: 10-Mar-00 30-Jun-00 31-Dec-00 30-Jun-01 31-Dec-01 30-Jun-02 31-Dec-02 30-Jun-03

(1) Exchange system (net): 0 0 0 0 0 0 0 0

* Small coins and sucres still outstanding.

Data source: Central Bank of Ecuador

6. Ecuador’s transition to dollarization in 2000 and 2001

Data source: Central Bank of Ecuador

Sep 94 - Aug 95 = 100

Data source: Central Bank of Ecuador

7. Ecuador’s macroeconomic performance since 2000

Table 4. Ecuador: Recent macroeconomic performance, 1997-2003

Source: International Monetary Fund, World Bank

Data source: Central Bank of Ecuador

Data source: Central Bank of Ecuador

Figure 15. Ecuador: Quarterly current-account performance, 1998 Q1 – 2003 Q3

Oil exports Merchandise imports Other merchandise exports

Data source: Central Bank of Ecuador

Figure 16. Ecuador: Quarterly fiscal performance, 1998 Q2 – 2003 Q2

Oil-export revenue Revenue other than oil-export revenue

Data source: Central Bank of Ecuador