Ecuador: Selected Issues and Statistical Annex

The crisis in the banking sector was one of the major contributing factors that led Ecuador to abandon its own currency and introduce the U.S. dollar as legal tender. However, to illustrate the weak growth performance of the country, it is necessary to examine the structural weaknesses in the labor market, the tax system, and the trade system. These weaknesses resulted in the increase in poverty and inequality. This paper provides a brief summary of recent economic developments and statistical data on economic indices of Ecuador.

Abstract

The crisis in the banking sector was one of the major contributing factors that led Ecuador to abandon its own currency and introduce the U.S. dollar as legal tender. However, to illustrate the weak growth performance of the country, it is necessary to examine the structural weaknesses in the labor market, the tax system, and the trade system. These weaknesses resulted in the increase in poverty and inequality. This paper provides a brief summary of recent economic developments and statistical data on economic indices of Ecuador.

IV. Dollarization41

63. Former President Mahuad announced his intention to adopt the U.S. dollar as legal tender in Ecuador on January 9, 2000, and the new administration of President Noboa, who took office on January 21, decided to continue this policy. Thus, the legal framework to implement dollarization was introduced in the Economic Transformation Law (Trole I) and approved by congress on March 13, 2000. From this date, the Central Bank of Ecuador has converted sucre for U.S. dollars at the fixed exchange rate of S./ 25,000 per US$1, the rate in effect since the January announcement of dollarization.

64. The background for this policy decision is discussed in Chapter III, “The Crisis in the Banking Sector.” The present chapter reviews some key arguments in support of and against dollarization, describes how dollarization is being implemented in practice, and provides a preliminary assessment of the impact so far.

A. Some Pros and Cons of Dollarization

65. Until Ecuador announced—in the midst of a political and economic crisis—its intention to adopt the U.S. dollar as legal tender, most of the discussion and analysis of the pros and cons of full dollarization had focused on the potential effects on a country that had already achieved a significant degree of macroeconomic stability (such as Argentina under its currency board system). The main conclusions of this analysis, and its implications for Ecuador, can be summarized as follows:42

On the benefit side:

  • Full dollarization would, if credible, eliminate the risk of future currency crises and thus it might also reduce the risk of default. This would lower spreads on international borrowing, which would help lower fiscal costs and could promote investment and growth. Full dollarization could also lower transaction costs and promote greater economic integration with the United States and the global economy. These benefits would be larger for countries (like Ecuador) with a history of high inflation and frequent currency crises. But higher borrowing spreads or disruptions to capital flows that were associated with other country risks, such as financial sector problems or a debt crisis, would not be removed by dollarization.

On the cost side:

  • The loss of seigniorage earnings from issuing domestic currency. During the 1990-98 period, annual average revenue from seigniorage was estimated at about 1.5 percent of GDP, largely reflecting a relatively high average annual inflation of about 38 percent over the period.43 However, the loss of seignorage compared with a system that maintained a domestic currency but focussed on achieving low inflation would be much less.

  • The loss of the option to pursue an autonomous monetary (and exchange rate) policy.44 The importance of such a loss for a country like Ecuador is difficult to quantify. On the one hand, it is vulnerable to frequent and sizeable external shocks (which are likely to be asymmetric to those of the United States), has significant rigidities in labor and goods markets, and lacks strong fiscal institutions and a tradition of strong fiscal policy (which can help reduce the need for sizeable adjustments of the exchange rate). All of these factors suggest that some monetary/exchange rate autonomy would be useful. On the other hand, the high degree of dollarization of monetary assets and liabilities in the domestic banking system had already made devaluation a costly—and less effective—policy tool.45

  • The loss of the ability to act as a lender-of-last resort for the financial system. This was potentially the greatest cost for a country like Ecuador, which has weak and poorly regulated banks and was already in the midst of a major banking crisis. But the experience of Ecuador has also shown that the ability of a central bank to find its way out of a financial crisis by printing money is limited in a de-facto highly dollarized banking system, since the injection of massive liquidity into the banking system to prevent a complete default on depositors led to greater pressures on foreign reserves and the exchange rate.

66. It is not possible to give a simple answer to where the balance of these costs and benefits lies, since many of the most important considerations are not quantifiable. However, Ecuador’s case also raises a separate question: Can the adoption of dollarization itself be used as a means to halt a banking and currency crisis? Ecuador in January 2000 did not look like a promising case for such an experiment, since the root of the crisis was the lack of sustainability of the fiscal positions and the lack of confidence in the soundness of most banks (including in the government’s deposit guarantee). Dollarization per se would do nothing to address directly either of these problems and, at the time of the dollarization announcement, there was little concrete evidence that congress would be prepared to act decisively on the necessary measures to address them. In these circumstances, some observers feared that dollarization would not halt the deposits flight out of the banking system and that, because the central bank would be unable to provide lender-of-last resort support of any significance, renewed bank runs and widespread arrears would be likely, with further adverse effects on the real economy. However, as the description given below of subsequent events shows, the dollarization announcement does appear to have given the government some breathing room—in terms of a halt to immediate liquidity pressures in the banking system—which the government used to introduce some important structural changes as part of the Economic Transformation Law (Trole I) that implemented dollarization. Moreover, the evolution of the banking system’s liquidity in the five months since dollarization was officially initiated, was also significantly better than expected. As a practical matter, however, it is not possible to distinguish the contributions to this outcome of dollarization per se, the accompanying structural measures, and the announcement of a package of financial support from the multilateral financial institutions.

B. The Legal and Institutional Framework for Dollarization

67. The three key features of the Economic Transformation Law that introduced dollarization were:

  • A prohibition on currency issue in sucres (except as fully backed coins);

  • The obligation on the central bank to exchange sucres for U.S. dollars at a fixed exchange rate and retire from circulation all sucre notes purchased;

  • An obligation of all firms to convert their accounting to dollars.

68. Bookkeeping by natural persons and corporations must be kept in U.S. dollars, an action immediately implemented by financial institutions. All contracts entered into by public institutions and tax assessments will be done in U.S. dollars; they can be settled in either dollars or sucres except for those taxes levied on foreign trade transactions that will be settled in U.S. dollars only. Banks will, for the time being, continue to settle accounts on the books of the central bank, thus reserve requirements deposits will continue to be held at the central bank.

69. The law also included a “conversion mechanism” (known as desagio) to translate previous sucre-denominated loans and deposits into dollars at lower interest rates, as well as a conversion process for previously issued dollar debt to lower rates for a short period (after which they would be rolled over at market rates). This was implemented through a one-time reduction in the interest rate on existing sucre- and dollar-denominated financial contracts. The deposit and lending rates for the periods of contracts remaining after January 11, 2000 (i.e., after dollarization was announced) were reduced to 9.35 and 16.82 percent a year, respectively. At the same time, the maximum lending rate was set at 24 percent.46

70. The Trole I law also introduced some important transitional arrangements, mainly to cushion the elimination of the lender-of-last-resort facility. In early 2000, the banking system was in a severe liquidity crisis, resulting from a combination of deposit flight, high incidence of nonperforming loans, withdrawal on banks’ external credit lines, and the absence of an integrated functioning interbank market. In order to maintain a facility to provide liquidity to troubled banks, the law therefore established:

  • a liquidity recycling mechanism within the banking system, mainly in the form of sales of U.S. dollar-denominated bonds by the central bank combined with repurchase operations; and

  • a liquidity stabilization fund to supplement the resources of the central bank available for providing liquidity assistance.

71. It is envisioned that the operation of the recycling mechanism would be phased out once a viable interbank market has reemerged, and the function of the liquidity fund would be replaced by external credit lines contracted for prudential purposes once Ecuador regains access to international capital markets.

72. To facilitate the operation of these two facilities and to highlight the transparency of the new monetary arrangement, the law mandated four operating accounts at the central bank (Table IV. 1). There is to be full backing with freely disposable international reserves of sucre currency in circulation (account one), as well as of bankers’ deposits at the central bank, and sucre-denominated central bank stabilization bonds (account two). As noted, the central bank would be allowed to operate a banking system liquidity recycling facility, partly funded by remaining disposable international reserves, and the placement of U.S. dollar-denominated central bank instruments in the local market (account three). The final account would cover all remaining assets and liabilities of the central bank.

Table IV. 1.

Ecuador: Central Bank of Ecuador Balance Sheet After Dollarization, March 13,2000

(In millions of U.S. Dollars)

article image
Source: Central Bank of Ecuador

73. The liquidity fund would be sourced by: (i) an initial US$40 million disbursement from the Andean Development Corporation; (ii) a 1 percentage point of the reserve requirement on bank deposits (about US$30 million); (iii) the reallocation of public entities’ financial assets held abroad to the central bank; (iv) external borrowing; and (v) budgetary transfers and government bonds. Access to the liquidity fund will be only in exchange for appropriate collateral.

74. All institutional arrangements are in place to make the liquidity stabilization fund operational. Initial resources for the fund are item (i) and (ii) in the previous paragraph. The institutional framework for the liquidity recycling facility is now also in place; a ruling on access to the facility for banks with a capital adequacy ratio below the regulatory minimum has been incorporated in the Proyecto de Ley Trole II (submitted to the congress as emergency legislation in mid-July 2000).

C. Impact of Dollarization and Challenges Ahead

75. Dollarization has proceeded very fast; of a stock of sucre denominated currency issued equivalent to US$460 million as of March 13, 2000, only US$115 million remained in circulation as of mid-July 2000 (see Table IV.2). Banks reported that even in remote areas, most of the transactions were being conducted in U.S. dollars by June. It is projected that by end-December 2000, currency in circulation denominated in sucre will be only about US$80 million, approximately the amount that the authorities estimate that the public will want to keep in the form of coins.

Table IV. 2.

Ecuador: Central Bank of Ecuador Balance Sheet After Dollarization, July 21, 2000

(In millions of U.S. Dollars)

article image
Source: Central Bank of Ecuador.

76. It is too early to embark on any comprehensive assessment of the impact of dollarization in Ecuador, mainly because many of the supporting reforms have yet to be formulated and implemented. However, it does appear that dollarization, in conjunction with the strong signal of political support for dollarization evidenced by the enactment of Trole I, and the announcement of substantial international financial support, including from the Fund shortly thereafter, has achieved the immediate objectives of reestablishing some confidence in the banking system, as well as averting hyperinflation.

77. The announcement of dollarization on January 9, including the exchange rate of S./ 25,000 per US$1 at which the dollarization would be adopted, introduced a de facto interregnum fixed exchange rate arrangement between the floating rate arrangement, in place prior to the announcement, and the actual implementation of dollarization in mid-March. This fixed-rate arrangement was introduced in the absence of any fundamental immediate change in monetary or fiscal policies.47 Yet, there was no discernible pressure on the exchange rate and virtually no intervention in the foreign exchange market by the central bank during the two months this interim arrangement was in effect.48

78. Could dollarization have been implemented successfully at a more appreciated rate than S/. 25,000 per dollar? A more appreciated rate, if credible, would have reduced the short-term inflationary surge that was essentially a consequence of the massive liquidity expansion and exchange rate collapse that occurred prior to dollarization. But, it is not possible to say now whether dollarization, at a more appreciated rate, would have had sufficient credibility to halt the crisis; at the time the authorities had several good reasons for believing that the use of a more appreciated rate might have weakened credibility and hence been less successful in halting the bank run:

  • The sucre had already reached the level of SI, 25,000 prior to the dollarization announcement. Adoption of a more appreciated rate (for example, S/. 20,000) might have invited a speculative attack—and a continued deposit outflow—before the new monetary arrangements were in place.

  • At S/. 20,000 per U.S. dollar, the available foreign exchange holdings would not have been sufficient to cover all sucre currency and other sucre liabilities of the central bank. While, in principle, the central bank could have borrowed to cover the difference, this would have been difficult in practice and the lack of full foreign exchange coverage could have undermined confidence.

79. Dollarization has also increased confidence in the banking system and has helped financial re-intermediation during the first half of 2000. As discussed in Chapter III, the banking crisis and related events had caused a flight out of deposits. For example, the ratio of currency in circulation to the sucre-denominated components of M2, which had averaged about 12 percent in the period July 1998-November 1998, jumped to 15 percent in December 1998, and steadily rose in 1999 to 26 percent by the end of the year. After the announcement of the intention to officially dollarize the economy, this ratio started to fall, and by end-March, it stood at 24 percent.49 Deposits in the domestic banking system, which fell by about 40 percent between July 1998 and end-1999, in part reflecting the real depreciation of the sucre, have recovered strongly and by June 2000, stood at US$3.6 billion, some US$400 million above end-1999 level. Since the process of unfreezing time deposits started in late March 2000, most of these deposits have remained within the banking system. After the implementation of dollarization, there has been no apparent liquidity need on the part of the banks.50

80. While the inflation rate has continued to increase—consumer prices rose by 65 percent in the first half of 2000—this largely reflects the pass-through of the steep depreciation of the exchange rate in the second half of 1999/early 2000. Moreover, evidence from other countries that have adopted “hard” pegs after bouts of sharp currency depreciation suggests that inflation takes some time to decline. During the last decade, several countries have adopted currency board arrangements (CBA) as their monetary system—including Argentina (1991), Bulgaria (1997), Estonia (1992), and Lithuania (1994), The experience of these countries in terms of the real exchange rate path and of inflation and interest rate convergence to that prevailing in the country whose currency was adopted as an anchor can provide some indication of how these key variables might evolve in Ecuador following dollarization.

  • These four countries introduced CBAs in the context of very high but declining inflation rates and, in the cases of Argentina and Bulgaria, following bouts of hyperinflation in the preceding two years. Convergence to the inflation rate of the countries whose currencies were used as anchor had not materialized after two years of the introduction of the currency board except for Bulgaria, although in each case inflation had fallen sharply (Figure IV. 1).

  • At the time of the introduction of the CBAs, prices for goods and services were mostly unregulated in these countries—administered prices had been largely liberalized in Estonia and Lithuania following independence, and there were frequent adjustments of the few remaining controlled prices in Argentina and Bulgaria. By contrast, Ecuador is still in the process of adjusting prices for goods and services set by the state (these prices have a direct impact on about 10 percent of the consumer price index, and mainly relate to energy and telecommunications) that were sharply reduced in U.S. dollar terms following the collapse of the sucre late last year and in early 2000. This suggests that inflation in Ecuador may not converge to U.S. rates for several years.

  • In each of the four countries, the exchange rate started to appreciate in real effective terms a few months before the introduction of the CBA, a tendency that persisted afterward. By contrast in Ecuador, the exchange rate had a dramatic fall in real terms in the two previous years and up to the eve of the announcement of dollarization.

  • In the countries which adopted CBAs, deposit and interbank interest rate spreads over the corresponding rates of the countries whose currencies were used as anchors narrowed very sharply in a short period of time, a development that already has started in Ecuador. By contrast, lending rates converged less rapidly, and appear to have been influenced mainly by conditions in the local banking system and domestic risk factors. This suggests that Ecuador should be cautious and not attempt to encourage a premature compression of lending rates.

Figure IV. 1.
Figure IV. 1.

Ecuador and Selected Countries: Inflation and Real Exchange Rate Developments Before and After Introduction of Currency Board Arrangements 1/

Annual Consumer Price Inflation 2/

(In percent per year)

Citation: IMF Staff Country Reports 2000, 125; 10.5089/9781451811773.002.A004

uA04fig01

Real Effective Exchange Rate

(Indices, 1990=100)

Citation: IMF Staff Country Reports 2000, 125; 10.5089/9781451811773.002.A004

Sources: Information Notice System and International Financial Statistics.1/ On the horizontal axis, T-0 is the month the currency board arrangement was introduced, T-n the months preceding the currency board and T+n the months following the currency board.2/ Inflation peaked at c. 20,000% in Argentina on T-13, at c. 2,000% in Bulgaria on T-4, at 1,200% in Estonia on T+2, and 1,400% in Lithuania on T-18.

81. Despite the initial quite favorable developments, dollarization will pose significant future policy challenges for Ecuador and it is still too early to declare it a success. Since the country will continue to be vulnerable to external shocks, both structural reforms and more effective policy responses will be necessary to absorb these shocks. In particular, increased flexibility in the labor markets will be critical. In recent years, wages in Ecuador have been in the lower end when compared to other Latin American countries. However, the labor market is still characterized by considerable rigidities arising from hiring and firing, compensation, and other policies (see Chapter V on labor markets). Moreover, some of the most difficult decisions to put in place the strong institutions needed to support dollarization still lie ahead. It is especially important that the initial successes not lead to any perception that dollarization can be a substitute for, rather than a complement to, tough policy decisions to keep fiscal deficits low and to create a more robust financial sector.

41

Prepared by Mayra Zermeño and Mariano Cortes.

42

See “Should Each Country Have Its Own Currency? The Pros and Cons of Full Dollarization” (SM/99/268, 11/02/99) for a more extended discussion.

43

The inclusion of the data for 1999 raises the estimate for seignorage to 1.9 percent of GDP, reflecting the impact of the deposit freeze in the demand for monetary base in that year.

44

For a country that has already adopted a currency board, and therefore already given up monetary autonomy in most circumstances, the additional cost of full dollarization consists of the loss of an “exit option” from the currency board, even if only under extreme circumstances.

45

The share of U.S. dollar-denominated deposits in M2 rose from about 5 percent in 1990 to 35 percent at end-1998, while that of loans rose from virtually nil to 60 percent over this period.

46

The Economic Transformation Law also introduced an interest ceiling, to be set by the central bank according to the formula: LIBOR + 4percent + country risk premium. Initially, the central bank decided on a lending rate ceiling of 24 percent, the government has proposed replacing the formula in the Proyecto de Ley Trole II legislation submitted to congress in July.

47

Late in 1999, the CBE unilaterally announced that it would no longer rediscount AGD bonds, operations that had resulted in significant injections of liquidity earlier in the year. The announcement was not perceived to be a fundamental regime change as the pressure on the sucre did not subside.

48

The only exception being some moderate intervention in the days following the attempted coup-de-etat in late January.

49

As noted, banks dollarized their bookkeeping starting in April, and it is no longer possible to estimate this ratio. At present, there are no estimates of U.S. dollars in circulation that can be used to estimate a more comprehensive ratio of currency to deposits.

50

However, an important contributing factor to the improved liquidity situation in banks is the contraction in new credit.

Ecuador: Selected Issues and Statistical Annex
Author: International Monetary Fund
  • View in gallery

    Ecuador and Selected Countries: Inflation and Real Exchange Rate Developments Before and After Introduction of Currency Board Arrangements 1/

    Annual Consumer Price Inflation 2/

    (In percent per year)

  • View in gallery

    Real Effective Exchange Rate

    (Indices, 1990=100)