IV. Implications of Currency Board Arrangements for the Design of IMF-Supported Programs
- Charles Enoch, and Tomás Baliño
- Published Date:
- September 1997
Although it might be argued that in the setting of a CBA all balance of payments disequilibria should adjust automatically, experience has shown that when sufficiently large shocks impinge upon the domestic economy, the government may need to implement active measures to maintain external balance. Therefore, the fundamental rationale of IMF support for adjustment programs—to provide members with the opportunity to correct disequilibria in their balance of payments without resorting to disruptive measures such as trade and payments restrictions—is as valid in the context of a CBA as in a conventional fixed exchange rate case.
Use of IMF Resources
The IMF has a direct role to play in two particular instances where its resources can be used to bolster the CBA and support the authorities’ adjustment program.
As Initial Backing
At the establishment stage of the currency board the authorities may need official foreign reserves backing to cover domestic monetary liabilities. In Lithuania, an IMF-supported arrangement was designed to prevent the possibility that a potential demand for conversion of the domestic monetary base into foreign currency might lead to the imposition of payments restrictions or raise a question as to the viability of the CBA.
Although, in such cases, the CBA is likely to be introduced within the context of an overall adjustment program, there is no automatic mechanism by which the CBA generates the means to repay its foreign obligations. In recognition of this fact, the Lithuanian authorities allowed the central bank to retain its profits until it could cover its liabilities with its own foreign resources.
As Lender-of-Last-Resort Support
A second case that would call for IMF support is where a sudden shock leads to a liquidity crisis that cannot be handled within the extant domestic financial framework. This was the case in Argentina in 1995. In such cases, a sudden capital outflow may lead to an incipient financial crisis. The knowledge that the central bank cannot provide liquidity support can lead to a self-fulfilling run on banks. Much as in the previous scenario, IMF resources can bolster credibility in the sustainability of the CBA—a key feature for the CBA’s self-adjusting mechanism to continue to operate and avoid a collapse of the adjustment program. It thereby serves to support the maintenance of a liberal exchange regime. Ideally, the right to draw from the IMF would help reestablish confidence and does not need to be used. The capacity of the authorities to draw promptly in such an eventuality may add to the credibility of the CBA.
Although the connection with balance of payments viability is indirect, the maintenance of a sound banking system is increasingly recognized as essential for successful economic stabilization. The IMF should therefore caution countries adopting a CBA about the risks that an unsound financial system poses for the sustainability of the arrangement, and insist that appropriate measures be adopted. As noted earlier, the Argentine authorities, for example, have adopted a number of important measures that have reduced the risk of systemic banking problems.
Targeting in IMF-Supported Programs
In stylized form, the design of a conventional macroeconomic framework for an IMF-supported program begins with the setting of a target path for the balance of payments. A projection is then made for money demand consistent with the program. An implied ceiling on the expansion of net domestic assets (NDA) of the central bank is derived. The central bank is then committed—through control of its net domestic assets—to attaining the balance of payments target, expressed as a floor on the stock of net international reserves (NIR).73 In particular, in the event money demand is less than projected, central bank net domestic assets must be kept below their ceiling.
The problem with this approach in a CBA is that the spirit of the arrangement rules out the active use of policy measures that influence net domestic assets. Nor, of course, does it use the exchange rate as an instrument. Net international reserves therefore are outside the authorities’ effective control. To take account of these factors, the design of IMF-supported programs has been modified. First, in countries operating a CBA, their programs place greater emphasis on the importance of fiscal adjustment and structural reforms. In the monetary area, IMF-supported programs in CBA countries have replaced the conventional floor on net international reserves and ceiling on net domestic assets with a floor on free international reserves, defined as NIR minus the currency board’s monetary liabilities (RM).74 An important quality of free international reserves is that it is unaffected by currency board conversions of reserve money and foreign currency, unlike net international reserves. If, for example, conversion of domestic currency into foreign currency exceeded the program assumption, no action would be required on the part of the CBA authorities to offset this decline in money demand, whereas, in a conventional program, net domestic assets would need to be adjusted downward to attain the NIR target. While, in conventional programs, NIR targets are set with the objective of attaining an improvement in the balance of payments (NIR targets are typically related to months of imports), in CBAs, NIR adjust endogenously to maintain the backing of base money.75
IMF-supported programs recognize, however, that CBAs do have some policy discretion coming from excess reserve cover or, in those cases where it is permitted, the ability to fall below full backing. In light of the risk to credibility that discretion entails, it is more important in CBAs than in conventional cases to outline clearly the extent to which monetary instruments may be used and explain that they will be used only in ways that support the sustainability of the program. While the performance criterion on free international reserves restricts the scope for discretionary credit expansion, permissible uses of domestic credit may also need to be spelled out. Thus, LOLR support—or more conventional rediscount operations or swaps—has been explicitly restricted in CBA programs and, when appropriate, lowering reserve requirements has also been restricted.
When determining the limit an IMF-supported program might place on changes in free international reserves and on central bank borrowing, the country’s access to international capital markets and its vulnerability to external shocks must be considered as well as the likely need and magnitude of lender-of-last-resort support. This, in turn, would depend on the importance and soundness of domestic banks, their liquidity position, access to foreign financing, and the level and nature of the reserve requirement. The need for flexibility to engage in monetary operations would depend, inter alia, on the design and functioning of the payments system and interbank market. Lastly, it would be important to ensure that the maturity of central bank borrowing is sufficiently long so as not to compromise the backing rule.
This paper has taken the view that CBAs may be attractive permanent arrangements for small open economies that wish to preserve the benefits of belonging to a broader currency area. Alternatively, they are useful transitional arrangements for countries that wish to delay the introduction of a full-fledged central bank until they build up central banking expertise or develop financial markets. CBAs may also be attractive to high-inflation countries adopting strong stabilization programs that wish to enhance the credibility of monetary policy. The implied fiscal stringency of a CBA and lack of discretion are not to be accepted lightly, however, because the consequences of a failed currency board experience can be substantial. While it is true that failure to sustain any adjustment strategy may have serious consequences, the greater investment made in generating credibility effects in a currency board means that its failure can be particularly devastating. Under such circumstances, it is essential that the authorities carry through with their adjustment effort with the intention either to exit the CBA from a position of strength or eventually to solidify the arrangement by adopting the long-term structural reforms necessary to limit the costs of operating within the CBA framework. In all cases, strong public finances, wage flexibility, and sound financial systems are key elements of robust CBAs.
As regards the trade-offs between flexibility and credibility in CBAs, recent experiences suggest that monetary and prudential instruments, operational procedures, and institutional arrangements can be adapted in ways that limit the need for discretionary intervention by the authorities. Nevertheless, the paper suggests that some discretionary monetary and LOLR support is generally needed to limit the risks of systemic liquidity crises at an acceptable cost. While less essential, a day-to-day monetary regulation capability may also be developed. In all cases, sufficient access to foreign exchange resources, in excess of those needed to back the CBA’s convertibility rule, must be secured to support such operations.
The paper suggests that the rationale for IMF-supported adjustment programs based on CBAs remains valid, although under slightly different forms and with somewhat different risks. In particular, the availability of IMF resources in case of a balance of payments crisis enhances the credibility provided by a CBA’s reserve backing. Those resources are expected to bolster the CBA and play a catalytic rather than a direct financing role. As compared with programs designed for other situations, IMF-supported programs in countries having a CBA need to put greater emphasis on fiscal adjustment and structural reforms—including in the labor market—and program targets need to be modified to account for the greater endogenity of monetary aggregates and the limited scope for discretion in monetary policy.
CBAs constitute a particular subset of fixed exchange rate arrangements. At any time, any given CBA may be made more or less rigid and rule-bound. In circumstances where credibility, transparency, and simplicity are paramount, little, if any, discretion is typically warranted. In other cases, where credibility has been established, and a role for money market intervention, LOLR capability, and government management of the payments system are important, more flexibility may be desirable in the CBA. In this sense, the line between a CBA and a more conventional fixed exchange rate is an arbitrary one.