The conduct of macroeconomic policy in Estonia and Lithuania in recent years has taken place in the context of currency board arrangements (CBAs). Under the arrangements (introduced in Estonia in June 1992 and in Lithuania in April 1994), the monetary authorities have virtually no discretion in conducting monetary policy since, under a CBA, the monetary base rises or falls in response to the central banks’ sales and purchases of foreign exchange at the fixed exchange rate. Indeed, a currency board can be defined as an “arrangement that legislates a particular monelary rule,” under which a country gives up all monetary sovereignty and allows changes in the monetary base to be determined entirely by the international balance of payments.1 In Latvia, the currency has been pegged to the SDR since February 1994 and the conduct of monetary policy has since been overwhelmingly geared toward maintaining the exchange rate peg. Although the Bank of Latvia does have a range of monetary policy instruments at its disposal (e.g., reserve requirements, open market operations, a refinance facility for overnight loans to commercial banks, repurchase auctions, and auctions of central bank deposits), in practice, limited use has been made of such instruments and developments in the monetary aggregates have reflected, as in Estonia and Lithuania, mainly movements in the balance of payments and, to a much smaller extent, the extension of central bank credit to government. Thus, except for some institutional differences, all three countries have operated for the last several years in the context of a fixed exchange rate regime.

The conduct of macroeconomic policy in Estonia and Lithuania in recent years has taken place in the context of currency board arrangements (CBAs). Under the arrangements (introduced in Estonia in June 1992 and in Lithuania in April 1994), the monetary authorities have virtually no discretion in conducting monetary policy since, under a CBA, the monetary base rises or falls in response to the central banks’ sales and purchases of foreign exchange at the fixed exchange rate. Indeed, a currency board can be defined as an “arrangement that legislates a particular monelary rule,” under which a country gives up all monetary sovereignty and allows changes in the monetary base to be determined entirely by the international balance of payments.1 In Latvia, the currency has been pegged to the SDR since February 1994 and the conduct of monetary policy has since been overwhelmingly geared toward maintaining the exchange rate peg. Although the Bank of Latvia does have a range of monetary policy instruments at its disposal (e.g., reserve requirements, open market operations, a refinance facility for overnight loans to commercial banks, repurchase auctions, and auctions of central bank deposits), in practice, limited use has been made of such instruments and developments in the monetary aggregates have reflected, as in Estonia and Lithuania, mainly movements in the balance of payments and, to a much smaller extent, the extension of central bank credit to government. Thus, except for some institutional differences, all three countries have operated for the last several years in the context of a fixed exchange rate regime.

While there appears to be a broad consensus that the above approach to exchange rate management, supported, for the most part, by appropriately tight fiscal policies and structural reforms, has served the Baltic countries well during the first stage of the transition, a number of questions have recently arisen concerning its medium-term suitability. For example, how sustainable is a fixed exchange rate policy in the context of countries that, notwithstanding the significant gains made on the stabilization front, have rates of inflation that are still substantially higher than those prevailing in the countries whose currencies serve as the anchor for domestic inflation? What are some of the possible disadvantages and costs associated with the continued implementation of these policies and at which point and under what conditions should alternative approaches be considered? While much has been written on the introduction of currency boards, there has been comparatively less debate on the policy issues and choices confronting policymakers as they contemplate the costs and benefits of a move toward more flexible arrangements. In any event, even if a consensus were to emerge that no policy changes are warranted in the short run, alternative policy regimes should be kept under review (e.g., an exit strategy). Also, it would be important to examine contingency measures that might be required if the CBA came under pressure and to consider ways to strengthen “lender-of-last-resort” mechanisms.

Recent Experience

There has been ample discussion in the literature on the macroeconomic advantages of CBAs. These mainly stem from the liberal exchange system implied by the full convertibility of the domestic currency, which is guaranteed by the 100 percent reserve cover; fiscal discipline, which is imposed by the inability of the central bank to provide credit to the government (which forces it either to balance the budget or to seek alternative sources of financing, domestically or abroad); the payments adjustment mechanism to various shocks, which is implicit in the arrangement;2 and, more generally, the transparency that is associated with the underlying institutional arrangements. These advantages are expected to engender confidence in the monetary system and thus create the conditions that foster investment, trade, and growth.3

To a greater or lesser degree all of these features have been present in the three Baltic states. At the same time, in all three, the available evidence points to an improvement in the effectiveness and credibility of the authorities’ anti-inflation policies associated with the presence of the CBA or peg, as witnessed by declining inflation and interest rates under the conditions of output recovery and a strengthening of the balance of payments (Table 3.1).

Table 3.1.

Selected Economic Indicators

article image

End of period.

In percent of GDP.

Including net lending.

Lending rates, in percent.

Includes public, publicly guaranteed, and private debt.

In percent of exports of goods and nonfactor services. Public and publicly guaranteed debt only.

In months' of imports of goods and nonfactor services.

Data prior to 1996 are not comparable.

U.S. dollar value, official data, f.o.b. basis.

In Estonia, the influence of the CBA on the government's stabilization policies has been well documented in a number of papers (e.g., Citrin and Lahiri, 1995). Estonia entered a phase of economic recovery in the second half of 1994, and output growth in 1995–97 has been led by the rapid growth of export volumes and strong domestic demand. Buoyant investment has reflected large inflows of foreign direct investment, sometimes amounting to some 5–9 percentage points of annual GDP. These inflows, in combination with a highly skilled labor force, have facilitated the transfer of technology, the renewal of the capital stock, and have contributed to productivity growth in the tradable goods sector.4 There is a broad consensus that the CBA has made an important contribution to the effectiveness and credibility of policies, as reflected in the favorable macroeconomic indicators: cumulative GDP growth in 1995–97 is estimated to be about 20 percent; short-term lending rates, which stood at 40 percent at the beginning of 1993, fell to some 14 percent in 1997, and annual end-of-period inflation stood at 12½ percent for December 1997, compared with 42 percent at the end of 1994.

In Lithuania, the period of operation of the CBA coincided with a sharp decline in nominal interest rates. Inflation has likewise been on a downward trend; with the end-of-period rate in December of 1997 reaching 8½ percent, the lowest level since the outset of the transition. The country's overall external position has improved in the context of a more open trade regime and a reorientation of trade toward hard currency markets. Foreign investment inflows remaining modest until 1996 (but recovered sharply in 1997) principally reflected delays in structural reform (notably cash privatization) and the more gradual emergence (for instance, in comparison with Estonia) of a favorable legal and regulatory environment with respect to foreign participation in domestic economic activity. As regards output, GDP growth has picked up considerably with cumulative growth in the period 1995–97 estimated at 14 percent.

A similar assessment can be made for Latvia where a strong perception has been created among economic agents that the exchange rate peg is at the center of the authorities’ anti-inflation objectives. Since the peg was first introduced, foreign reserves have increased and been maintained at levels providing adequate import coverage; domestic interest rates, with some temporary interruptions,5 have been on a downward path of convergence to international levels; and domestic monetary assets have grown, reflecting the higher level of reserves but without additional pressures on prices as the increase in money was associated with higher money demand. As in Estonia and Lithuania, inflation has also decelerated and output has entered a period of recovery, supported by some pickup in foreign investment and other capital inflows.

Evaluating the contribution of the exchange rate regime to an external outcome involves the counterfactual problem. Furthermore, the presence of other factors, largely exogenous to the exchange rate regime, is likely to have had an influence on the evolution of the above indicators. A case can also be made that the introduction of new exchange rate regimes in all countries had been preceded by a period of commitment to stabilization and reform (indeed, the perception of the existence of such commitment was an important consideration behind establishing such regimes) and, thus, some of the ex post improvement noted above may have reflected the beneficial effects of such policies. Other favorable elements linked more to the overall macroeconomic climate (and geographical location) than to the nature of the exchange rate regime per se, for instance, the increased availability of foreign financing, are also likely to have played a role. Nevertheless, the weight of evidence suggests that the CBAs in Estonia and Lithuania and the peg in Latvia (coupled with a strong independent central bank) significantly enhanced the credibility of the authorities’ disinflation policies, often against a background of structural weaknesses and constraints (e.g., energy sector financing difficulties in Lithuania in 1994/95, banking crises in Estonia in 1992–94, in Latvia in 1995, and in Lithuania in 1995/96).

Limitations and Constraints

While an exchange rate regime which reduces discretion can be operationally useful and enhance policy credibility, it can also impose constraints and potential costs that need to be kept under close review. This section examines some of the main disadvantages.

Risks of Overvaluation

A potential problem with fixed exchange regimes is the risk of overvaluing the exchange rate that may result if the arrangement is introduced while inflation is still relatively high, even if the level of the exchange rate initially chosen implied substantial undervaluation. The danger is not so much that inflation will not eventually be brought down to international levels but rather that the transition to these lower levels may be long and ultimately result in an overvalued exchange rate. Or, alternatively, that the inability to allow a nominal appreciation of the exchange rate will make it more difficult to contribute to a more rapid convergence of domestic prices to international levels.

The level of the initial peg that was chosen when the fixed exchange rate regimes were introduced in the Baltic states was heavily determined by the need to make the arrangements credible even at the cost of disequilibrium in the goods market implied by an undervalued real exchange rate. It was expected that this disequilibrium would work itself out over time, although there was practically no basis for projecting how speedily the process would be completed.6

The length of the transition period is obviously not independent of public perceptions of the government's determination to adhere to the fixed rate. A perceived weakening of such determination involving, for instance, a loosening of the stance of fiscal policies, is likely to lead to economic agents (wage and price setters) starting to build in premiums on wages and prices, reduce the demand for domestic currency, bid up domestic interest rates, and so on, against possible future changes in the exchange rate, a process which is likely to slow down the process of convergence to permanently lower inflation.7

Relative price adjustments in the transition from central planning to a market economy are a key factor that has contributed to the persistence of higher inflation in the Baltic countries. The prices of previously heavily subsidized goods and services (e.g., food, fuel, housing, health care, among others) with a large weight in the consumer price index have all increased sharply, and such increases have not been offset by decreases in the prices of other goods, leading to upward adjustments in the price level. A measure of the extent to which this process of convergence in the structure of relative prices to that prevailing in the rest of the world was expected to continue following the introduction of the CBA or peg is provided, for instance, by estimates that suggest that consumer prices in 1994 in Latvia and Lithuania stood at levels equivalent to 30–35 percent of 1994 prices in Sweden and Austria.8

The risks of overvaluation have been mitigated in the Baltic countries by gains in productivity that have allowed the maintenance of an adequate degree of competitiveness during the transition, including the process of relative price adjustment. The potential for additional gains in productivity in the Baltic countries is thought to be large as the modernization of these economies continues and as they partake of the traditional benefits of greater foreign participation in domestic economic activity (e.g., nondebt capital inflows, transfer of technology and knowhow, and so on) and a more open trade regime.9 In general, there is legitimate concern that a domestic rate of inflation that is persistently higher than the rate prevailing abroad can eventually erode whatever “headroom” might have existed at the outset of the introduction of the fixed rate regime.

While there are some country-specific nuances, on the whole, there is no evidence that the evolution of the real exchange rate (measured by the usual consumer price–based indices) during the last few years has been associated with a deterioration of their external position.

In Estonia, as in Latvia and Lithuania, dollar wages remain well below levels in Western European trade partner countries, and these relatively low labor costs are often cited as an important contributing factor for the growth of foreign direct investment. Export growth has remained buoyant in all three countries, and while the current accounts deficits have widened, especially in Estonia, international reserves remain adequate.

In addition, all three countries have remained largely free of indexation mechanisms and are perceived to have fairly flexible labor markets that have helped maintain competitiveness. There are no strong employer organizations and trade unions, and the governments have on the whole not intervened in the wage-setting process, with wage settlements generally based on firm-specific productivity developments. Levels of unemployment compensation are relatively low, while eligibility periods for drawing benefits are short. Furthermore, privatization processes have generally eschewed the use of restrictions on the employment levels and production profiles of privatized entities.

A recent updating of the trade weights used in the computation of the real effective exchange rate in Lithuania (which gives a somewhat larger weight to trade with the East and uses a more representative set of trade partners—e.g., inclusion of trade with Estonia and Latvia) shows a real effective appreciation between April 1994 (i.e., introduction of the CBA) and December 1996 of 15 percent (Figure 3.1). In Latvia, the real effective rate based on the consumer price index (CPI) appreciated by 12½ percent between February 1994 (i.e., pegging to the SDR) and December 1996.10 With the CBA in place for a much longer period, the real appreciation in Estonia has been more pronounced—46 percent between January 1993 and December 1996; however, even this real appreciation is less pronounced if the calculation uses unit labor costs and export unit values or a tradable goods price index. Between the first quarter of 1993 and the second quarter of 1996 the real appreciation is estimated to have ranged between 15 percent and 30 percent.11

Figure 3.1
Figure 3.1

Real Effective Exchange Rates

(November 1993 = 100)

Loss of Autonomy

Partly related to the difficulties that a country might face in adjusting to shocks, the constraints imposed by a CBA on the traditional functions of a central bank and the implementation of an active monetary policy can, in certain circumstances, have tangible costs. There may be instances when the presence of a CBA prevents an adaptation of monetary policy warranted by specific domestic conditions.12 The loss of autonomy can be seen as an important potential disadvantage, particularly in the case of countries perceived to be in the early phase of the transition to a market economy, thus still subject to various imbalances and structural rigidities (e.g., distortions in the financial system, difficulties in collecting tax revenues). The extent to which this loss of autonomy is to be regarded as a problem is partly a function of the role that the authorities see monetary policy playing in influencing real economic variables (e.g., high unemployment, weak output growth). If the primary aim of monetary policy is to achieve sustained price stability (leaving to structural and fiscal policies the task of dealing with rigidities and supply constraints), then the relevant question is whether it is easier to control inflation with an “independent” monetary policy in the context of a flexible exchange rate or through some type of fixing, either through a CBA or some targeting mechanism involving little or no autonomy.13

In general, the scope for an effective independent monetary policy will be greater if (1) the authorities do not yield to the temptation of accelerating money growth in the hope of temporarily stimulating output; (2) money demand and supply functions are reasonably stable and predictable; and (3) monetary policy is not held captive to pressures stemming from the need to finance large fiscal deficits or large wage increases, or both. Whether these conditions will be satisfied in the Baltic countries in the period ahead is an open question. Financial policies have been on the whole cautious but some of this caution may have been itself strengthened by the CBA's proscription of lending to government or, in Latvia, by a policy that has put defense of the exchange rate peg at the center of the authorities’ macroeconomic policies and that has been supported by a favorable track record of fiscal discipline. The economic transition itself (including policy instability), together with ongoing processes of financial innovation, is likely to create sufficient “noise” to make money demand functions relatively unstable, at least in relation to the instability confronted by policymakers in other countries further along the transition path. The exceptionally low money-to-GDP ratios in these countries suggest the need for gradual monetization over the medium term; such a process is akin to a structural change in the money demand function, which would imply high instability in the underlying parameters.14 While unstable parameters are dealt with on a daily basis by central banks all over the world, the Baltic countries have to face additionally the disadvantageous combination of large systemic changes and scarce central bank expertise.

By the same token, for inflation control in the context of fixing to work it is necessary that the rate of inflation of the anchor currency be lower and more stable than the domestic rate of inflation and that the authorities’ commitment to the fixed exchange rate be fully credible. The first condition is clearly fulfilled in the Baltic countries. The issue of credibility is more complex but there seems to be consensus that the CBAs in Estonia and Lithuania and the peg in Latvia have indeed enhanced the credibility of government anti-inflation policies. By doing so, they may also have helped minimize the output costs associated with disinflation. Further elaboration on the medium-term exchange rate policy options in the Baltics is presented in Section IV.

Lender of Last Resort

The constraint that in a CBA the money supply can grow only in direct relation to inflows of foreign currency means that the central bank cannot be the lender of last resort for the banking system. Indeed, a view has been put forward that there is an implicit trade off between the lower convertibility risk under a CBA and the possibly higher risk of a domestic financial crisis since an increase in the demand for cash will lead to a contraction of the banking system. (Indeed, in Lithuania, during the first three months of the banking crisis—that is, the first quarter of 1996—the CBA converted sufficient currency into foreign exchange to lead to a contraction in reserve money and broad money, the latter on the order of 13 percent). CBAs have dealt with these risks partly by creating a separate bailout facility that is allowed to provide limited assistance, within the constraints imposed by the excess over the minimum amount required for backing. Unless the cushion of such “excess” foreign reserves is substantial, the CBA's ability to respond to a major financial crisis will be sharply circumscribed. Indeed, an argument often put forward in favor of a conventional peg (as opposed to a CBA) stems from the potential fragility of the domestic banking system in the absence of a lender of last resort and the possibility that a liquidity crisis could turn into a systemic crisis affecting the banking system.15

The CBAs in Estonia and Lithuania have withstood reasonably well these countries’ respective banking crises, both of which originated in the context of ineffective supervision and poor banking practices (weak lending skills, insider abuse, overextension of the banks’ branch network, violations of regulatory provisions, undercapitalization, among others) and both of which eventually resulted in the closing or restructuring of a number of problem banks. Some central bank lending took place in the context of operations to relieve liquidity problems, but this was limited and the costs of bank restructuring were rapidly transferred (or are in the process of being transferred) to the budget.16

A case can be made that the presence of the CBAs during the banking crises (or the authorities’ firm commitment to the peg in Latvia), rather than acting as a serious constraint, may have played the role of a hard budget constraint, sharply limiting the room for maneuver of the government at a time when the temptation to relax policies in the face of strong political pressures was great. By forcing the closure or restructuring of the banking system's worst offenders and focusing the authorities’ attention on some of the underlying weaknesses and the need for reforms, the CBAs may have actually contributed to improving the medium-term viability of the banking system. The CBAs in the Baltic countries have also lessened the extent of the moral hazard problem implicit in a system with a central bank legally able—and perhaps quite willing—to act as a lender of last resort. On the other hand, the outcome of these particular episodes does not prove the more general case that the absence of the lender-of-last-resort function would also be beneficial in future episodes of banking difficulties.

The banking crises have prompted the introduction of medium-term strategies for restructuring the banking sectors, involving, inter alia, tighter prudential rules and international accounting standards, improved monitoring of banks, recapitalization or privatization of former state-owned problem banks, and the introduction of private deposit insurance schemes. Nevertheless, the possibility of future crises cannot be ruled out. Short of an active use of excess holdings of foreign exchange for lender-of-last-resort or monetary operations (which would run counter to the basic premise of a CBA of limited discretion), a case can be made that “some flexibility can add to the sustainability of a currency board and thus enhance its credibility.” Indeed, “institutional arrangements, operational procedures and monetary and prudential instruments can be designed to reduce risks of a systemic liquidity crisis while limiting discretionary interference from the monetary authorities. In addition, public debt policies can be reformed to limit the risk of a debt crisis. Nevertheless, some lender-of-last-resort support is needed—preferably under central bank control—to contain financial sector problems at an early stage and avert contagion risks. This should be done in a manner that addresses systemic problems in the banking system while seeking to avoid bailouts of insolvent banks. Indeed, the existence of such support facilities can enhance confidence in the domestic financial system, and hence lower intermediation spreads.”17

Fiscal Discipline

The limited scope (de facto or de jure) for monetary and exchange rate policies in the Baltics puts a considerably greater burden on fiscal policy in support of the authorities’ macroeconomic and stabilization objectives. It also raises the issue of the extent to which the CBAs have contributed in a tangible way to greater fiscal discipline and whether, in the absence of a commitment to sound fiscal policies, a CBA or fixed rate regime could actually generate it. With few exceptions (Latvia in 1995, Lithuania in 1994, and Estonia in 1996), the Baltics have built a favorable track record of fiscal discipline, and the commitment to a fixed exchange rate has contributed to reinforce that record. By constraining the financing options available to the government, the CBAs have imposed a harder budget constraint than would otherwise have been the case (e.g., by allowing the closure of problem banks, budget support for bank restructuring may turn out to be less costly than would have been the case had the government been able to monetize a large budget deficit) and thus buttressed the authorities’ anti-inflation policies.

Portfolio Constraints

In reviewing the role of currency boards, Schwartz (1993) notes the criticisms made early on in the debate of their relative costs and merits on account of the need to fully back the monetary base by foreign currency. It was argued that since it was highly unlikely that the entire currency issue would have to be redeemed at one time, maintaining the equivalent of a 100+ percent reserve requirement was unduly limiting, possibly denying the monetary authorities (and ultimately the government) more profitable investment opportunities.18 A measure of the seigniorage lost via a currency board would have to take due account of the risk-adjusted yield on domestic assets when compared with the yield on the foreign assets that they would be replacing, as well as other, possibly distortionary, effects associated with, for instance, monetization of government deficits.19 In practice, a case can be made that the seigniorage “losses” in the Baltic CBAs are minimal (if not zero), since, in any event, the reserves held as coverage for the currency issue are at about the levels that they should be (about three months’ worth of imports) for small, relatively open economies, even if the country in question did not have a currency board. Moreover, from market participants’ point of view, it is likely that a high coverage ratio (even in excess of the amount of central bank domestic liabilities) would be perceived as enhancing credibility of the underlying arrangement, since it would leave the central bank room to deal with a banking crisis. It also needs to be noted that the sole reliance on balance of payments flows to supply the monetary base may make adjustment to changes in money demand more severe than necessary.

Solvency Risks

Regardless of the particular exchange rate regime, in a typical banking system balance sheet, assets and liabilities denominated in both the local and the reserve currency coexist. The full convertibility of the domestic currency under a CBA implies that the banking system must be able to convert, on demand, its liabilities in domestic currency into the reserve currency. If the exchange rate peg is perfectly credible, the coexistence of liabilities denominated in two different currencies does not introduce any additional solvency risk. If, however, there is a sudden change in sentiment concerning the government's commitment to the fixed rate (or the underlying institutional arrangements), the resulting rise in the interest rate paid on domestic currency denominated assets will depress the market value of the banks’ long-term domestic currency assets. Given the fixed exchange rate, the reduction in the market value of banks’ assets will not be matched by a reduction in the value of their short-run domestic currency liabilities. This implies that a CBA with such a mixed banking system balance sheet could be subject to changes in asset valuation driven by changes in exchange rate expectations. While this may not be a serious problem in practice (since CBAs reinforce credibility in the parity) it can be an occasional source of potential instability (e.g., the arrival of a new government that the market perceives as being less committed to maintaining the CBA) and can be eliminated only by denominating the entire balance sheet of the banking system in the reserve currency; that is, complete “dollarization,” or by maintaining complete credibility at all times.20

Policy Options Over the Medium Term

This section analyzes some of the limitations (potential or otherwise) associated with the Baltic countries’ approach to exchange rate management and briefly discusses the sustainability of the underlying arrangements.

Choice of Exchange Rate Regime

Any assessment of options for monetary and exchange rate policies over the medium term should take into account the Baltic countries’ stated intentions to join the EU, a goal for which there appears to be a broad-based domestic consensus. Integration with the EU and the process leading to eventual accession will provide a clearly defined framework over the medium term for the conduct of economic policies, as evidenced, for instance, by ongoing broad-based efforts in a large number of areas to adapt institutions and legislation to the standards prevailing in the EU. In this respect, clear evidence of sustained commitment to a stable exchange rate and the policies that support it, as demonstrated, for instance, by the institutional ability to stick to a simple monetary rule (or one involving, as in Latvia, limited effective discretion) and the willingness and readiness to cede some sovereignty in specified areas are conditions that could facilitate the process of integration with the EU.

It was noted recently that “the choice of exchange rate regime is one of the longest-running debates in economics,” and “the fact that it is not resolved must mean that there is no exchange rate system that is superior in all circumstances.”21 Early discussion in the literature argued that pegging would be beneficial if the degree of factor mobility (regional and interindustry) was appropriately high, the size of the economy relatively small, and the degree of openness sufficiently high.22 With the possible exception of relative factor mobility (on which the empirical evidence is lacking), the Baltic countries are certainly small and relatively open, suggesting the desirability of a fixed rate regime. As noted earlier, credibly pegging to a stable currency or basket of currencies of low-inflation countries does ensure a transition to lower inflation. In particular, as pointed out by Fischer: “it helps focus the mind of the government on a very clear constraint on policy” (p. 36).

Beyond issues of the structure of the economy (e.g., relative size, degree of openness) and credibility, a key consideration that argues in favor of a fixed exchange rate in the Baltic countries stems from their governments’ determination to seek membership in the EU and, eventually, its monetary arrangements. Since EMU entry criteria are likely to require a prolonged period of exchange rate stability vis-à-vis the euro, a temporary switch to more flexible arrangements to be followed by a return to a fixed regime might seem unnecessary. With these and other considerations in mind, the Bank of Lithuania, for instance, has recently begun to implement a medium-term strategy for the gradual evolution of the CBA into a traditional peg, such as presently exists in Latvia and that, in its final stage, envisages a permanent fixing of the rate vis-à-vis the euro (see Appendix I for details). This approach raises a number of interesting policy issues; some are identified and discussed below.

Because of the tougher institutional arrangements associated with a CBA, market participants typically perceive that the probability of a change in the exchange rate under a conventional peg is higher. Other things being equal, this lower credibility would normally be associated with a higher level of interest rates under a peg. Furthermore, maintaining a fixed rate in the context of financial liberalization can raise some difficulties. Pursuit of a tight money policy to reduce inflation may push domestic interest rates above foreign rates and lead to capital inflows, which may offset the effects of the tight monetary policy. Unless the source of the inflow is an increase in the domestic demand for money (in which case the appropriate response is to allow the money supply to rise), such capital inflows can exert pressures on the exchange rate.

A key difference between a CBA and a fixed exchange rate regime pertains to the possibility of sterilized intervention. Under a conventional peg the central bank may allow an increase in foreign reserves while maintaining the money supply constant. Under a CBA, the monetary authorities cannot, through purchases of domestic currency bonds, sterilize the potentially contractionary effect on the domestic money base associated with sales of foreign exchange.23 Sterilized intervention can thus, in principle, smooth out such temporary shocks and may allow the monetary authorities to partly neutralize the effects of capital inflows.24

An argument sometimes made against fixing the exchange rate is that it can reduce the authorities’ room for maneuver during macroeconomic shocks. A real shock (e.g., a sharp increase in the price of imported oil, as occurred in the Baltic countries in 1991/92) will require adjustments in relative prices—like the real exchange rate—and the absence of sufficient flexibility in the domestic price level in the short run will mean that flexibility in the nominal exchange rate will make it easier to achieve the desired level in the real rate. The literature on optimal currency areas suggests that the choice of the exchange regime should recognize the existence of unpredictable and persistent shocks, and that the decision to join a currency area should entail careful analysis of the nature of the underlying shocks and of their degree of symmetry. Other things being equal, the more asymmetric the distribution of real shocks across countries, the greater the country's cost of forgoing the option of exchange rate adjustment. Since the process of convergence between the Baltics and their trade partners (particularly in the EU) is far from complete, the possibility cannot be excluded that some real shocks might arise requiring a terms of trade adjustment and an exchange rate realignment. Should the Baltic countries succeed in keeping their labor markets flexible, even after EU accession, then the costs of these future shocks would not be magnified by the presence of a fixed exchange rate.

Whether the monetary authorities operate under a CBA or a traditional peg has important implications for the role that they may play in accelerating the deepening of financial markets and, more generally, encouraging institutional development. As will be argued below, the Baltic countries need to make further progress in this area, particularly in the banking sector and in enhancing the sophistication and flexibility of financial markets. Under a narrow and strict definition of a CBA, the monetary authorities can play only a limited role in these institutional processes. Partly to address some of these limitations and as a first step in a process aimed at broadening the development of other monetary instruments over the medium term, the Bank of Lithuania started limited repurchase operations on treasury bills as a way of facilitating the development of the secondary market, which has remained thin, illiquid, and noncompetitive (see Appendix I).

Preconditions for a Switch

If a country decides to make the switch from a CBA to a conventional peg, a key issue concerns the conditions that need to be in place to ensure a successful transition. This section identifies and examines the relevance of a number of such factors for the Baltic countries. In particular, the focus is on the role and underlying strength of the banking system, the development of the domestic financial markets, and the adequacy of the level of foreign exchange reserves. Central bank independence and the existence of a clear mandate to enforce price stability as well as the need to develop a sufficient degree of institutional capacity for the exercise of the necessary functions are also discussed. While the factors identified here are all desirable policy objectives and, to a greater or lesser degree, should be pursued liidependently of the exchange rate regime, they acquire particular relevance (as will be noted below) in the context of transition economies characterized by structural rigidities and inefficiencies, particularly in the financial sector.

Strength of the Banking System

In the Baltics, as in other countries, the implementation of monetary policy requires the existence of relatively stable relationships between instruments and policy objectives. Disruptions to the financial system adversely affect these relationships and the links between policy instruments (e.g., interest rates, money, and credit aggregates) and policy objectives, such as price stability. Regardless of whether monetary policy is transmitted through direct or indirect instruments, the transmission mechanism is closely linked to and ultimately depends upon the soundness of the banking system. Indirect instruments of monetary policy will be affected adversely by illiquid or insolvent banks, because of their inability to adjust reserves or lending in response to monetary policy signals. Banks with limited balance sheet flexibility may not be able to respond appropriately to policy changes; for example, reserve requirements will not be effective if illiquid banks are not able to meet them. A credit auction or similar market-based liquidity facility may be distorted by adverse selection and moral hazard, since unsound institutions may be willing to borrow at any price to avoid illiquidity. If high-risk borrowers represent a substantial share of total commercial bank credit, the effectiveness of liquidity management through open market operations will be reduced by the low interest-rate elasticity of banks’ credit demand. Similarly, when the authorities employ direct policy instruments, such as credit ceilings, their effectiveness will be reduced if weak banks simply rollover their portfolio of bad loans or if the liquidity provided by new deposits goes to finance banks’ losses on nonperforming loans. In such cases, credit ceilings will not be especially effective in constraining the growth of net domestic assets (NDA) of the banking system.

The strengthening of confidence in the banking sector in turn is essential to the widening and deepening of financial markets. As a result of the banking crises in Latvia and Lithuania, nominal interest rates rose sharply and did not begin to fall appreciably until the strengthened enforcement of prudential banking regulations and the beginning of the implementation of bank restructuring programs. Similarly, the laying out of solid foundations for the development of the capital markets depends on an appropriate, consistent legal framework that clearly lays out the “rules of the game” in the financial sector and in the interactions of the financial sector with the rest of the economy (e.g., collateral). Without a consistent and transparent set of laws, the risk premium will be high and investors will not invest domestically. These structures have only been recently put in place in all three Baltic countries and, as a result, nominal interest rates have come down. It remains to be seen whether the bank restructuring strategies will ultimately be successful (particularly in Latvia and Lithuania) and at what (fiscal) cost. Further strengthening of the banking system and the underlying regulatory climate would be a desirable policy objective in terms of the institutional developments that must take place ahead of EU accession.

Central Bank Independence

In Lithuania (as in Estonia and Latvia), the Law on the Bank of Lithuania recognizes the central bank as an institution “independent from the Government of the Republic of Lithuania and other institutions of executive authority” and it explicitly states that its “principal objective is to achieve stability of the currency.” The bank is called upon to support the government's overall economic policy, “provided said policy is in compliance with the principal objective of the Bank.” The governor is appointed by parliament (for a period of five years in Lithuania25) and cannot be dismissed prior to the expiration of his or her term, unless he or she is convicted of some crime or is not able to perform his or her duties properly due to health problems. From a legal point of view, the basis for an independent monetary policy appears to be well established in the Baltic countries.

Nevertheless, the sequence of events that led to the resignation of the governor of the Bank of Lithuania at the peak of the banking crisis in the spring of 1996 raises questions about a possible gap between the de jure independence of the central bank and the effective interpretation and implementation of that law in practice.26 In general, the literature on central bank independence notes that an otherwise legally independent central bank will maintain its credibility only as long as it operates in a political environment where support for the idea of using monetary policy for the medium-term control of inflation is broad-based, where the track record of adherence to the rule of law is reasonably long and where, more generally, the legal climate is characterized by stability and predictability. Given the early stages of the transition in the Baltic countries, the ample scope for further progress in a number of areas, particularly on the structural front, as well as the need to continue to build up public support for macroeconomic stability and a free market, including the need to generate a broader consensus on the role of medium-term targets for monetary policy, an appropriate legal, institutional, and public opinion context for the exercise of an adequately independent monetary policy may not yet exist in Lithuania. However, there is consensus that the Latvian and Estonian central banks, while operating in a broadly similar formal legal climate as in Lithuania, enjoy a higher degree of effective independence.27

Gross Reserves and Central Bank Liabilities

In assessing the adequacy of foreign reserves in a transition economy, it is desirable to consider the underlying policy environment, as well as the level and structure of liabilities of the central bank. The experience of several countries in recent years suggests that foreign exchange flows can exhibit significant volatility in the presence of uncertainties associated with, for instance, perceptions about the policies of a newly elected government or other, more systemic, factors. At the peak of the banking crisis in Lithuania, for instance, foreign exchange outflows through the CBA during a two-month period reached nearly $150 million (equivalent to some 17 percent of gross reserves prior to the crisis).

The gross reserve positions of the three Baltic countries are shown in Table 3.1. These are in the neighborhood of 2½–3½ months of imports. Because the monetary authorities have significant foreign liabilities (e.g., $350 million in Lithuania, of which 90 percent are to the IMF), the net foreign asset positions are lower and range over 1½–2½ months of imports. For a small, relatively open, market economy, gross reserves equivalent to some three months of imports have typically been considered adequate. The need to further strengthen the domestic financial systems in all three countries and the underlying uncertainties associated with the early stages of the transition (in particular, the need to continue to support the convergence of inflation with respect to regional partners in the EU and to accelerate the process of structural reforms), suggest that a more cautious approach is desirable, calling for a wider margin of safety, that is, a relatively high reserve coverage ratio.

Secondary Market for Treasury Bills and Interbank Lending

An important condition for the adoption of indirect policy instruments is a well-functioning money market that continuously transmits to the authorities up-to-date information on liquidity conditions and interest rate developments. In Latvia and Lithuania, while the primary market for treasury bills has developed rapidly, secondary trading of treasury securities has remained quite limited, with estimated annual turnover on the order of 10–15 percent of the stock of the securities issued. There are several reasons for the thin secondary trading. First, financial institutions are not capable of operating as market makers, either because they have insufficient capital (in the case of brokerage companies) or because they lack the experience and the relevant skills. Second, the actual number of participants is usually very small. Last, in Lithuania, taxation is a limiting factor as capital gains taxes are paid only on securities that are sold before maturity.

Except for Estonia, the interbank market is similarly underdeveloped. In Lithuania, the market collapsed in late 1995 and has not recovered despite government guarantees, registering only a few transactions a week. It is unlikely that before bank restructuring actually takes place and the public regains full confidence in the system, a secondary market for bank reserves that is liquid, deep, and capable of signaling to the authorities interest rate developments will come into existence.

Institutional Development

Institutional capacity for the conduct of monetary policy is difficult to measure. In Estonia and Lithuania, the problem is further complicated since there is no way of using past policy performance as an indicator of institutional strength even though the two central banks have had available to them limited lender-of-last-resort functions, mainly intended to preserve the viability of the banking system. The amounts of financial resources that are, in principle, available to the central banks of Estonia and Lithuania to fulfill lender-of-last-resort functions are not insignificant. One can think of two types of margins for the available resources: those established with respect to the program limits (i.e., floors on net international reserves), which are relatively narrow (1.6 percent and 10 percent of reserve money, respectively). The more fundamental limits are implicit in the requirement of 100 percent cover for reserve money; since actual cover is greater than 100 percent, excess resources that could be used for lender-of-last-resort operations without violating the cover requirement amount to 37 percent and 30 percent of reserve money, respectively.

Under the simple monetary rules of CBAs, the day-to-day conduct of monetary policy does not hinge to any crucial degree on the availability of up-to-date information on total liquidity, nor does it require, as a matter of high priority, the development of a sophisticated forecast capacity and other analytical skills within the central banks. If alternative exchange rate arrangements are to be pursued at some point in the future, the development of the relevant research and analytical functions of the central bank would become an indispensable institutional requirement. Both central banks have made some limited headway in this area; in Lithuania a research division was recently created to prepare short-term forecasts for reserve money, while at the Bank of Estonia the research department produces macroeconomic forecasts for real variables and analyzes monetary developments but is not involved in the preparation of periodic monetary and interest rate projections.

There is obviously scope for the development of additional institutional capacities in the area of monetary instruments, such as open market operations, rediscount facilities, credit auctions, and so on, as in Latvia. The problem is that conventional open market and credit operations go against the spirit of a currency board and are essentially ruled out in the context of the present policy framework. In this regard, the development of these capacities would appear to require the introduction of a transition period during which some measure of flexibility is introduced in the arrangements.28


This section has reviewed the recent experience of the Baltic countries with fixed exchange rate regimes, currency boards in Estonia and Lithuania, and a peg to the SDR in Latvia that has involved little de facto use of monetary policy instruments. Against a background of declining inflation and falling interest rates, along with a recovery of output and a strengthened balance of payments position in all three countries, the section has argued that the existing arrangements have served the authorities’ stabilization policies well, having backed the credibility of stabilization through an institutional anchoring of financial restraint.

The potential disadvantages sometimes associated with fixed exchange rate regimes (and in particular with CBAs), such as the risks of overvaluation, vulnerability to shocks, and the absence of a lender of last resort, do not appear, on balance, to have constrained the implementation of credible macroeconomic policies in the Baltic countries and may, in some circumstances, have actually reinforced the authorities’ determination to pursue tight policies in the face of strong political pressures to do otherwise. On the basis of the evidence, no compelling case can be made that it is necessary at the present time to change in a fundamental way the approach to exchange rate management in the Baltic countries. In addition, the Baltic countries’ stated intention to integrate their economies with the EU, and the policy requirements that this process is likely to entail, would argue in favor of a commitment to a stable exchange rate and the policies that support it.

The relative success to date of the fundamental policy framework in place for the Baltic states should not preclude the consideration of alternative arrangements. Alternative policy scenarios, involving greater flexibility in the conduct of monetary policy, should be developed and kept under review especially given the changing global environment.

Appendix I. The CBA and an Exit Strategy

In January 1997, the Bank of Lithuania completed a medium-term policy paper that lays out a strategy for the gradual evolution of the currency board arrangement into a traditional currency peg. The basic objective underlying the strategy is the strengthening of political and economic links with the EU, including eventual membership and participation in its monetary institutions. The strategy distinguishes clearly the institutional arrangements underlying the CBA and, in particular, the automatic creation of a monetary base associated with foreign exchange inflows, and the choice of a fixed exchange rate regime per se. The program envisages three stages:

(1) The first stage began in 1997. The Bank of Lithuania would keep the CBA in place but gradually introduce new monetary policy instruments, within the room permitted under the ceilings of the program under the Extended Financing Facility. The instruments introduced were repurchase operations in treasury bills and a short-term Lombard credit facility, collateralized by treasury bills. The bank committed itself to limiting the amount of repurchase operations held in its balance sheet to LTL 60 million and to clear the balance of such operations periodically. The provisions were designed to ensure that an adequate margin would be maintained for the lender-of-last-resort function. The Lombard facility would be a standard credit window for commercial banks and, in contrast with the existing liquidity loans, would have the advantage of being collateralized with short-term assets rather than being guaranteed by the government.

(2) During the second stage, which, originally, was to start in early 1998, the government would introduce amendments to the Litas Stability Law, allowing a broadening of the definition of the assets that could be used to back reserve money. While this would alter the essential character of the CBA, under which central bank liabilities may be backed only by foreign exchange reserves, the authorities would like to retain the Litas Stability Law itself, since its various features contribute to confidence (e.g., article 4 of the law, which ensures full convertibility of the domestic currency). With a view to further buttressing confidence in the currency, the existing parity with the dollar would be maintained and the foreign exchange position of the Bank of Lithuania would be strengthened. It would be ready to conduct sterilized interventions, as needed.

(3) In the third and final stage, to be introduced not earlier than 1999, the litas exchange rate would be pegged to a basket made up of the U.S. dollar and the euro, should the latter be in existence, or some other basket of EU currencies. Necessary conditions for entering the final stage of the program would be a monthly rate of inflation below 0.8 percent for a period of six consecutive months and substantial growth in banking sector deposits. This stage would precede the permanent fixing of the rate vis-à-vis the euro.

Appendix II. Monetary Policy Instruments During Transition

The transition from a CBA to a conventional peg—if this were the path chosen—would be accompanied by a shift in the underlying institutional arrangements and a gradual introduction of the monetary instruments currently unavailable to the central banks in Estonia and Lithuania.29 This appendix discusses possibilities in this area, some of which could start without any changes to underlying legislation.

First, the central bank could engage in activities aimed at stimulating the development of the weak interbank markets. With particular reference to Lithuania, the central bank could encourage commercial banks to shift, on an overnight basis, excess reserves across accounts, thus helping improve daily management of reserves. By operating as an overnight broker, the bank would not create additional liquidity and would not be involved in any risk of default or exchange risk. The Bank of Estonia also holds excess reserves but, given the substantial development of the interbank market there, the need for a central bank brokerage role is certainly less urgent.

Second, the central banks could marginally expand available discretionary net domestic asset margins, thus creating the opportunity for a broadening of the central bank's institutional experience, under an otherwise controlled environment. The discretionary margin, currently viewed as a margin of last resort, would be available for open market and credit operations.30 An obvious instrument available to the authorities under this option would be repurchase operations in some liquid short-term asset, such as treasury bills in Lithuania and central bank certificates of deposit in Estonia. Another instrument that could be developed would be a credit auction facility, similar to the one that existed, for example, in Lithuania before the introduction of the CBA.31 Credit auctions can be a way of pricing central bank credit at this stage in the development of the money market and the absence of a reliable interbank reference rate. Credit auctions would need to be collateralized; in Lithuania, with the ample availability of treasury bills, a natural collateral is already available. In the medium term, central banks could gradually introduce a standard credit facility, against the collateral of eligible bills. The development of a well-developed interbank market would provide a natural reference interest rate and would make it possible to introduce directly a standard discount window.

A third way to introduce further flexibility in the CBA would be to allow a small portion of the foreign reserve backing for the monetary base to be made up of government bonds denominated in convertible currency, as in Argentina. This alternative would create additional opportunities for conducting open market operations that could be used to smooth out intramonthly fluctuations in the demand for cash.32

The combination of the above options involving degrees of flexibility appropriately tailored to each country's particular institutional circumstances and preferences would make it possible over time to gradually replace the conventional CBA by a traditional peg. To reduce the possibility of a heightening of tensions in the financial markets the process of changing the legislative framework would need to be managed carefully.


That is, the response of the money supply and interest rates to a balance of payments deficit that, through its effects on absorption, eventually brings about a reversal of the deficit. This process might turn out to be quite traumatic if the banking system is fragile.


For an overview of the issues see Williamson (1995). See also Balino (1997). In a brief survey of the challenges facing central banks Fischer (1996, p. 37) observes that “the monetary theory of the currency board is exactly that of the gold standard. Provided the arrangement is credible, it brings the benefit of rapid convergence toward international inflation and interest rates…. there should be no mistaking the severe demands a currency board puts on monetary policy.” This may partly help explain the relative infrequency with which monetary authorities have opted for CBAs.


Estimates prepared by the Bank of Estonia suggest that average labor productivity in the tradables sector in Estonia increased by some 40 percent between the first quarter of 1993 and the second quarter of 1996.


For example, in 1995 in the aftermath of the collapse of the largest bank, which resulted in some capital outflows and a contraction in the money supply.


While references are made in the literature to a consensus suggesting relatively slow adjustment speeds to real exchange rate disturbances, the issues are not settled, particularly since the extent of initial disequilibria observed for the Baltic states falls far from what has been observed in other cases. The particular consensus that is mentioned implies that disequilibria damp out at an annual rate of 15 percent, implying that it would take about four years to dissipate one half of the initial discrepancy between domestic and international prices (Rogoff, 1996). See also Parsley and Wei (1996) and Engel and Rogers (1996).


An interesting variant of this could be seen in Lithuania in late October 1996 when, following parliamentary elections, uncertainty about the future direction of policies led to a swift and negative market reaction. One aspect of the private sector's response to this heightening of policy uncertainty was to accelerate efforts in the banking system to denominate loan contracts in dollars. See Baliño (1996), p. 20.


For a more detailed discussion see Richards and Tersman (1995), Berengaut (1996), Lorie (1996), and van der Mensbrugghe (1996), where evidence is presented that suggests strong positive correlation between annual inflation and the size of the gap between domestic prices and prices abroad.


Foreign direct investment per capita in Baltic countries still lags behind Central European economies in transition.


The estimate combines a real depreciation against the currencies of the Baltic countries, Russia, and other countries of the former Soviet Union, which account for roughly 50 percent of total trade, and an appreciation with respect to the European Union (EU), which accounts for the rest; to the extent that a growing share of trade is with the EU, the relatively small 6 percent real appreciation may overstate somewhat Latvia's strong competitive position.


The usual methodological and conceptual issues underlying a proper interpretation of such indices in the Baltic countries remain valid. While providing useful signals, about underlying cost, developments, and price competitiveness, the indices need to be interpreted with caution and in the context of a more systematic analysis of other developments in the balance of payments.


Williamson (1995) gives the example of Hong Kong in the early 1990s when the link to the U.S. dollar involved importation of low interest rates from the United States at a time when a domestic asset price boom would have made monetary restraint the desirable policy option.


Capital mobility, interdependence, and the globalization of the world's financial markets are increasingly making the notion of a truly “independent” monetary policy more a theoretical proposition than a policy option. On this point Schwartz (1993, p. 172) notes: “Central banks of the industrialized countries themselves no longer believe that they know how to produce monetary surprises that are stabilizing and that they are masters of discretion.” Nevertheless, it is legitimate to speak of a relatively less constrained monetary policy as might happen outside the context of a CBA (e.g., Latvia).


The average ratio of broad money to GDP in the Baltic countries in 1996 was 20 percent, ranging from 13 percent in Lithuania to 27 percent in Estonia. The same unweighted ratio for a sample of seven small open European countries (Austria, Belgium, Denmark, Finland, Norway, Sweden, and Switzerland) in 1995 was 69 percent.


Actually, two additional features that can help alleviate such pressures would be the presence of a well-developed interbank money market and the presence of foreign banks in the domestic market that could draw upon resources of their parent banks. Neither of these options is yet readily available in the Baltic countries, at least not in a way that might make a tangible difference.


In Lithuania, for instance, loans granted by the Bank of Lithuania to the State Commercial Bank in July–August 1996 were repaid in September 1996, thereby fully restoring the lender of last resort margin under the program's net international reserves target.


Considerations of this type may be behind calls, voiced occasionally, for more active use of foreign reserves in support of domestic economic activity.


Osband and Villanueva (1993) note that revenue from seigniorage stems from the interest earned on investing the foreign reserves minus the “costs of printing bank notes and minting coins.”


The issue is more complicated, since while the banks can protect themselves against the exchange rate risk by reducing their open position, they cannot protect themselves against the risk that their loans will go bad if their customers are adversely affected by an exchange rate change.


Fischer (1996), p. 36.


For an introductory survey see Isard (1995). A discussion of the relationship between relative size and openness and the effectiveness of a given devaluation of the exchange rate is presented in Tower and Willett (1976). A number of arguments are put forward there as to why exchange rate adjustment is relatively less effective in small, open, undiversified economies, than in large, closed ones.


For instance, reserve money fell by 20 percent in Lithuania between December 1995 and March 1996, in the aftermath of the banking crisis.


In this regard, Fischer (1996) also notes the virtues of fiscal discipline and the fact that some countries have sometimes successfully implemented “market-based policies to reduce the returns to foreign investors” (e.g., via increases in reserve requirements on nonresident deposits or other temporary measures).


Article 10 of the Law states that “The chairperson… shall be appointed for a term of five years and his or her salary shall be fixed by the Seimas of the Republic of Lithuania upon the recommendation of the President of the Republic.” The law does not spell out in detail the procedures for appointing the chairperson, however.


In early 1996, following the imposition of moratoriums on several large insolvent banks, the Board and the top management of the bank were forced to resign. The present incumbent is the fifth governor the bank has had during the six-year period through 1997.


It may be recalled that in Latvia in 1995, following the closure of Bank Baltija, a motion of no confidence in the governor and deputy governor of the central bank was introduced in parliament; in the event, the motion failed. It is not clear what would have been its legal status, had it succeeded.


Some elaboration on the kinds of operations that could be developed as a country shifts to a policy involving somewhat greater discretion than under a CBA is presented in Appendix II.


The Estonian authorities have publicly stated that they do not envisage any change to the currency board and the present peg, except with the advent of the European Monetary Union when the kroon's link to the deutsche mark will be replaced by a link to the euro according to the conversion rate at which the value of the deutsche mark will be fixed to the euro.


To protect the lender-of-last-resort function, at least partial clearance of the facilities might be required on a periodic basis.


The authorities would need to carefully explain to economic agents that these credit auctions are conducted with “surplus” resources and, as such, would not compromise the CBA.


The Litas Stability Law, for instance, includes in the definition of foreign exchange reserves “bonds, and other debt securities payable in convertible currency, which are held by the Bank of Lithuania.” This would not appear to preclude the introduction of such domestic bonds denominated in a convertible currency.


  • Baliño,Tomás, Charles Enoch, and others 1997, Currency Board ArrangementsIssues and Experiences, IMF Occasional Paper No. 151 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Bennett,Adam G., 1994, “Currency Boards: Issues and Experiences,” IMF Paper on Policy Analysis and Assessment 94/18 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Berengaut,Julian, and others, 1996, Republic of LithuaniaRecent Economic Developments, IMF Staff Country Report No. 96/72 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Camard, Wayne, 1996, “Discretion with Rules? Lessons from the Currency Board Arrangement in Lithuania,” IMF Paper on Policy Analysis and Assessment 96/1 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Citrin, D.A., and A.K. Lahiri, 1995, Policy Experiences and Issues in the Baltics, Russia, and Other Countries of the Former Soviet Union, IMF Occasional Paper No. 133 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Caprio, G., M. Dooley, D. Leipziger, and C. Walsh, 1996, “The Lender of Last Resort Function Under a Currency Board,” Policy Research Working Paper No. 1648 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation
  • Engel, Charles, and John H. Rogers, 1996, “How Wide Is the Border?American Economic Review, Vol. 86, (December), pp. 111225.

  • Fischer, Stanley, 1996, “Maintaining Price Stability,” Finance & Development, Vol. 33 (December 1996), pp. 3437.

  • Halpern, L., and C. Wyplosz, 1996, “Equilibrium Exchange Rates in Transition Economies,” IMF Working Paper 96/125 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Isard, Peter, 1995, Exchange Rate Economics (Cambridge: Cambridge University Press).

  • Lorie, Henri, and others, 1996, Republic of EstoniaSelected Issues and Statistical Appendix, IMF Staff Country Report No. 96/96 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Osband, K., and D. Villanueva, 1993, “Independent Currency Authorities: An Analytic Primer,” Staff Papers, International Monetary Fund, Vol. 40 (March), pp. 20216.

    • Search Google Scholar
    • Export Citation
  • Parsley, David C., and Shang-Jin Wei, 1996, “Convergence to the Law of One Price Without Trade Barriers or Currency Fluctuations,” Quarterly Journal of Economics (November), pp. 121136.

    • Search Google Scholar
    • Export Citation
  • Richards, A., and G. Tersman, 1995, “Growth, Nontradables, and Price Convergence in the Baltics,” IMF Working Paper 95/45 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Rogoff, Kenneth, 1996, “The Purchasing Power Parity Puzzle,” Journal of Economic Literature, Vol. 34 (June), pp. 64768.

  • Schwartz, Anna, 1993, “Currency Boards: Their Past, Present, and Possible Future Role,” Carnegie-Rochester Conference Series on Public Policy, Vol. 39, pp. 14787.

    • Search Google Scholar
    • Export Citation
  • Tower, E., and T.D. Willett, 1976, “The Theory of Optimum Currency Areas and Exchange Rate Flexibility,” Special Papers in International Economics (Princeton University: International Finance Section).

    • Search Google Scholar
    • Export Citation
  • Saavalainen, Tapio O., 1995, “Stabilization in the Baltic Countries: Early Experience,” Road Maps of the Transition, IMF Occasional Paper No. 127 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • van der Mensbrugghe, Emmanuel, and others, 1996, Republic of LatviaRecent Economic Developments, IMF Staff Country Report No. 96/143 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Williamson, John, 1995, What Role for Currency Boards? (Washington: Institute for International Economics, September).

From Economic Stabilization to EU Accession